Page 1


Roman Republican Colonization New Perspectives from Archaeology and Ancient History


Editorial Board Prof.dr. Harald Hendrix Dr. Marieke van den Doel Dr. Jeremia Pelgrom Dr. Arthur Weststeijn Editorial Staff Malena B. McGrath Carla M. Summers Correspondence regarding editorial material and contributions should be addressed to: Papers of the Royal Netherlands Institute in Rome Royal Netherlands Institute in Rome (KNIR) Via Omero 10/12 - 00197 Roma www.knir.it e-mail: papers@knir.it subscription and order of single volumes: Palombi & Partner Srl Via Gregorio VII, 224 00165 Roma www.palombieditori.it


Papers of the Royal Netherlands Institute in Rome Mededelingen van het Koninklijk Nederlands Instituut te Rome


PAPeRS Of tHe ROyAl NetHeRlANDS INStItute IN ROMe - VOluMe 62 - 2014

Roman Republican Colonization New Perspectives from Archaeology and Ancient History

edited by

tesse D. Stek and Jeremia Pelgrom


© 2014 Palombi & Partner srl Via Gregorio VII, 224 00165 Roma www.palombieditori.it layout, graphics and editorial assistance care of the publishing house All rights reserved. Without limiting the rights under copyright above, no part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the written permission of both the copyright owner and the editors of this book. Cover: Inscription from Aquileia mentioning the triumvir lucius Manlius, 2nd c. B.C. (courtesy of the Museo Archeologico Nazionale di Aquileia, inv. no. 1). Page 8-9: B. Rhenanus, P. Vellei Paterculi Historiae Romanae, 1520 (Basel), 9. Page 42-43: the ager Cosanus from the air (after Castagnoli, f. 1956. “la centuriazione di Cosa”, MAAR 24, tav. 24). Page 122-123: A view of the landscape across the liri Valley (G. R. Bellini, A. launaro and M. Millett). Page 276-277: Black glazed pottery with Herculean stamps (M. Vitale, SAA). Page 332-333: Relief commemorating the founding ceremony of the Augustan colony of Aquileia with the representation of the sulcus primigenius (courtesy of the Museo Archeologico Nazionale di Aquileia, inv. n. 49100). ISBN 978-88-6060-662-4


Contents PR efACe

7

INtRO DuCtION Roman Colonization under the Republic: historiographical contextualisation of a paradigm Jeremia Pelgrom and Tesse D. Stek

10

PAR t I CO Nte x tuAlIzING ROMAN Re PuBlI CAN COlONIzAtION. BACKG RO uND S, DefINItIONS AND C OMPARANDA Private Vis, Public Virtus. family agendas during the early Roman expansion Nicola Terrenato

45

the nature of Roman strategy in Mid-Republican colonization and road building Guy Bradley

60

Roman colonization and the city-state model Jeremia Pelgrom

73

the city-state model and Roman Republican colonization: sacred landscapes as a proxy for colonial socio-political organization Tesse D. Stek livy 27.38 and the vacatio militiae of the maritime colonies Luuk de Ligt

87

106

PAR t II COlONIAl l AND SCAP eS. COlONI StS AND NAtIVeS SHA PI NG tH e uRBA N, NAtuRAl AND SOCIAl eNVI RONMeN t Gellius, Philip II and a proposed end to the ‘model-replica’ debate Jamie Sewell

125

Republican colonization and early urbanization in Central Adriatic Italy: the valley of the River Flosis 141 Frank Vermeulen Strangers in Paradise. latins (and some other non-Romans) in colonial context: a short story of territorial complexity Michel Tarpin Colonisation romaine et ‘espaces ripariens’ dans les Civitates Campaniae de Sylla aux triumvirs Ella Hermon

160

193


early colonization in the Pontine region (Central Italy) Peter Attema, Tymon de Haas and Marleen Termeer

211

le colonie di luceria e Venusia. Dinamiche insediative, urbanizzazione e assetti agrari Maria Luisa Marchi

233

Roman colonial landscapes: Interamna lirenas and its territory through antiquity Giovanna R. Bellini, Alessandro Launaro and Martin Millett

255

PAR t III t He Re lIGI OuS DI MeNSION Of ROMAN COlONIzAtION tutelary deities in Roman citizen colonies Marion Bolder-Boos

279

Il culto di Apollo nella colonizzazione romana Andrea Carini

295

Il santuario di ercole ad Alba fucens: nuovi dati per lo studio delle fasi pi첫 antiche della colonia latina 309 Daniela Liberatore

PAR t IV t He CR eAtI ON Of ROMAN Ce NtRA lI t y Effigies parvae simulacraque Romae. la fortuna di un modello teorico repubblicano: leptis Magna colonia romana Mario Torelli

335

Qua aratrum ductum est. la colonizzazione romana come chiave interpretativa della Roma delle origini Simone Sisani

357

Publications of the Royal Netherlands Institute in Rome

405


Preface

Roman colonization has generally been seen as a primary model for colonization and colonialism in more recent historical periods. the direct relevance of Roman colonization to wide-ranging historical interpretations and to the self-perception of modern nation-states and empires explains the continuing fascination exerted by Roman colonization in both academic and non-academic circles. trying to unravel the character and development of the key phase of Roman colonization in the Roman Republican period is therefore essential not only for ancient history, archaeology, and related disciplines but also for a better understanding of the modern world. the most comprehensive study on Roman colonization remains edward togo Salmon’s Roman Colonization under the Republic. this study, published in 1969, is one of the very few monographs written on Roman colonization and it is by far the most authoritative and frequently cited study in both the Anglo-Saxon and Continental research traditions. In the almost 50 years since the publication of Salmon’s seminal book many crucial revisions have been proposed for different aspects of the traditional view of Roman colonization. Despite the obvious importance of these new studies, their impact on our general understanding of Roman colonization and their deeper significance for understanding Roman imperialism has yet to be fully appreciated. the increasing fragmentation of the research field is an important reason that an overarching, radically new, understanding of Roman Republican colonization has not, as yet, been brought forward. Issues that are central to the character of Roman colonization are studied in separate disciplines including Roman historiography, urban archaeology, architecture studies, landscape archaeology, Roman religion studies and Roman law. this volume brings together recent insights from a range of different academic traditions, lifting language and cultural barriers. By presenting both new theoretical insights and new archaeological discoveries, it explores the potentially productive interplay between different emerging research areas currently isolated. the initial stimulus for this volume was provided by a european Science foundation exploratory Workshop held in Ravenstein, the Netherlands, in November 2010, which was co-funded by the Radboud university Nijmegen. thanks goes to all participants of the workshop enduring the stay of three days in an isolated monastery to discuss the direction of Roman colonization studies. Many of the contributors to the present volume were present on that inspiring occasion. Where it was felt necessary to balance the disciplines reflected in the present volume, other scholars were invited to participate, especially regarding landscape archaeological approaches. the editors are particularly grateful to all the authors and to the anonymous reviewers of the papers for their invaluable feedback. A major debt of gratitude goes to Rogier Kalkers and Corinne tetteroo for their practical editing assistance. the editors are grateful for the financial support of the Netherlands Organization for Scientific Research (NWO), to Bert Brouwenstijn and Bas Kinsbergen for the graphics, and for language editing to Anita Casarotto, John Dillon, Antonella lepone, Malena B. McGrath, elizabeth Robson and Carla M. Summers. Bibliographical abbreviations follow the American Journal of Archaeology guidelines and, for classical texts, the Oxford Classical Dictionary, 3rd edition. the illustrations have been provided by the authors unless otherwise indicated.

ďœˇ


INTRODUCTION


fIG. 1. Roman colonies founded in Italy from the early Republic to the Gracchi.




Roman Colonization under the Republic: historiographical contextualisation of a paradigm Jeremia Pelgrom and tesse D. Stek

Introduction Recent years have witnessed increased interest in Roman colonization1 and previous conceptions of Roman colonization have been fundamentally challenged (fig. 1). At present, under the influence by postcolonial theory, there seems to be a strong consensus that previous views of Roman colonization are outdated and that the development of new ones is needed to move forward.2 An influential approach to this challenge has been the adoption of alternative perspectives to mainstream Romanocentric ones and the development of methodologies to bring these perspectives into focus. especially since the 1990s, the attention paid to the subaltern part of society, to the downtrodden, and more generally to the variegated experiences and voices in ancient colonial situations has been especially productive in this respect and has opened our eyes to fundamentally different approaches to the subject.3 In certain circles, this has produced a new orthodoxy constructed around key themes such as resistance, appropriation, hybridization, variegation, discrepant experiences, fluidity of identities and more. this type of approach is sometimes associated with, or rather claimed by, archaeologists more than ancient historians,4 but in reality this undoubtedly healthy corrective development is based on a dynamic and fruitful debate of both critical-historical insights and new archaeological theoretical approaches. After an initial, perhaps overenthusiastic deconstruction of top-down perspectives, which resulted in an underestimation of Roman agency, a more productive approach now takes Roman power into account, but also considers the different and intricate ways in which it was exercised.5 Without doubt, recent studies of Roman colonization have made important progress by applying such new approaches to the primary sources and archaeological evidence. yet, they have tended to do so in opposition to what they view as outmoded interpretations of Roman colonization or in relative isolation of the long and complex history of research. framing new interpretations in opposition to previous work or not engaging this work at all has risks. One such risk is the simplification of earlier work caused by preferring one prominent study to many other, less visible ones, or by fusing elements from different contributors into a chimerical ‘standard view’. the imminent risk here is that, by losing sight of the complexity of the past debate, new interpretations are framed according to a very particular, unintentionally circumscribed line of thought. As will be demonstrated below, such a process may ultimately narrow a debate that actually is much wider in scope and potentially obscure alternative approaches to the subject. Another approach, also of postcolonial pedigree, is to delve deeper into the complex historiography of Roman colonization and to trace the emergence of some concepts and the decline of others against their historical and intellectual background. yet, despite the momentous historical importance of Roman colonization, the history of modern thinking about Roman colonization has been surprisingly little scrutinized

1

2

In this article, with ‘Roman colonization/Roman colonies’, we refer to both latin and citizen colonies. On this primarily modern rigid distinction, see also further below. Cf. recently the important contributions in Hurst and Owen 2005 and in Bradley and Wilson 2006.

3 4 5

Van Dommelen 1998 is essential. e.g. Webster and Cooper 1996. Cf. e.g. Huskinson 2000 and for Republican Italy Wallace-Hadrill 2008 and Stek 2009; cf. now Stek 2014 with bibliography.




and problematized.6 Whereas the importance and impact of new postcolonial theoretical and methodological approaches for the study of Roman imperialism have been amply reviewed in recent scholarship, much less attention has, as yet, been paid to the historical contextualization of the formation of modern views of Roman colonization. therefore, we have chosen to focus in this introductory contribution on the historiography of Roman colonization, which we consider one of the essential first steps toward a new and hopefully more balanced understanding of the phenomenon of colonization in the Republican period. We will illustrate the importance of understanding the historiographical tradition by discussing what is beyond a shadow of a doubt the most influential study of Roman colonization down to the present day, Roman Colonization under the Republic by edward togo Salmon7 (1905-88). this study, published in 1969, is not only one of the very few monographs written on Roman colonization,8 but it is also by far the most authoritative and frequently cited study in both the Anglo-Saxon and Continental fIG. 2. edward togo Salmon (after evens 1974, 2). research traditions. It is usually the earliest modern study quoted. Salmon’s work, however, is often represented in a very schematic and stereotypical way. At the same time, it also is commonly regarded as a balanced overview of prior scholarship. this is encouraged by Salmon’s own claim that he merely “seek[s] to collate what is known or guessed about the colonies” and “provide an upto-date synthesis”.9 Salmon’s habit of seldom referencing other scholars contributed to his preeminent place in the historiographical tradition. Roman Colonization under the Republic accordingly has become synonymous with traditional, statist, top-down views of Roman colonization as advancing imperialism and romanization (fig. 2). By analyzing Salmon’s reasoning in detail and tracing its development over time, his seminal study can be situated more firmly in the wider, complex modern historiography of Roman colonization. In fact, Salmon’s work ultimately represents a very specific and therefore partial line of thought in Roman colonization studies. We will demonstrate that his book definitely cannot and should not be considered the starting point of Roman colonization studies. Roman Colonization under the Republic does not objectively

6 7 8



terrenato 2005. Salmon 1969. the most important synthetic studies of Roman colonization before Salmon are discussed below. Recent syntheses on colonization include Bernardi’s Nomen

9

Latinum (Bernardi 1973) and laffi’s Colonie e municipi nello stato romano (laffi 2007), both of which focus on legal aspects of colonization (e.g., the origin of latin rights, questions about government, and so forth). Salmon 1969, 11. Cf. also below.


reflect the very diverse lines of thought that had emerged prior to its publication. the following offers a glimpse of the vast panorama of the highly refined and intense debate outside Salmon’s own preoccupations, or those of similar champions of his generation like frank Brown (1908-88), that have attracted the most attention and obtained an iconic and somewhat isolated status. the aim is to show that engagement with these sometimes highly technical debates is essential for laying a solid foundation for new approaches to the study of Roman colonization. Such a survey may help a better understanding of the relationship between conceptions of Roman colonization, on the one hand, and topics such as power dynamics, administration and imperialism, land and property issues, legal and civic questions, romanization, and other cultural processes on the other, all of which are wide-ranging topics treated in the diverse contributions to this volume.

Collating a history of Roman colonization So much has been said about this great Roman institution in the past that anyone who writes about it today is hardly likely to shed a blinding new light on it, much less to revolutionize traditional conceptions of it. He can, however, seek to collate what is known or guessed about the colonies, provide an up-to-date synthesis, describe their vicissitudes and men’s changing attitudes towards them, appraise their varying purpose and importance, and perhaps suggest some new approaches to several old problems. Such is the intended scope of the present work. (Salmon 1969, 11).

Salmon clearly states what his book was intended to be: an up-to-date synthesis of existing knowledge gained in the long and erudite scholarly tradition leading up to his monograph. Such a synthesis was undeniably a desideratum of the time. New data and insights had increased enormously, and the technical debate had become too complex for the non-specialist audience. In addition, German and Italian scholars had dominated the research field for long, and their findings did not always easily reach the englishspeaking world. Despite this favorable momentum, Salmon only partly filled this lacuna. On closer examination, his book is not a passive collection of recent knowledge about Roman colonization, but is more polemical and selective than one might suspect at first sight. Paradoxically, Salmon’s originality is not the result of integrating all the new data and technical insights that had become available but derives from his reexamination of the canonical ancient texts. In this section, the focus will be on the more conventional aspects of Salmon’s work; the sections that follow explore, in greater detail, how his work deviated from other established views. On a basic, structural level, Salmon’s book closely follows the time-honored antiquarian practice of composing chronologically ordered descriptions of Roman colonies. After the introduction, which is thematic in character, Salmon adopts a rigidly chronological narrative framework. Colonies are discussed in the order of their foundation dates and are grouped together in chapters that cover broad historical periods.10 this scholarly tradition can be traced back to Roman times. the best example is Marcus Velleius Paterculus who, in an excursus in his compendium of Roman history, summarizes the names and dates of

10

His narrative is subdivided into the following chronological phases: 1) colonies before the latin War, 2) colonies founded between the latin War and the Second Punic War (latin colonies and maritime colonies are discussed in two separate chapters, resp. three and four, 3) colonies founded during the Second Punic War, 4) colonies founded from the Second

Punic War to the Gracchi, 5) colonies of the Gracchan Age, 6) colonies of the age of revolution, and finally 7) colonies in the Roman empire – which is at the same time the conclusion. Beyond summarizing the main conclusions of the book, this last chapter is structured as an epilogue describing colonial policies of the Imperial Age.




colonies founded between the capture of Rome by the Gauls and circa 100 B.C (fig. 3).11 By Salmon’s time, the quality of such colonial lists obviously had improved markedly; more information had been recovered not only from literary sources, but also from epigraphic, numismatic, and to a lesser extent, archaeological evidence. the development of colonial histories can be clearly traced by comparing influential lists (historiae) of colony foundations compiled in the early Modern period. early examples such as francesco Ruperti’s De coloniis romanorum tempore liberae Rei publicae deductis (1838)12 are based almost exclusively on information provided by Velleius and livy, who, besides names and foundation dates, occasionally provide information on the number of colonists, the size of distributed allotments, and the involved triumviri; the studies that followed steadily incorporated further information.13 A good illustration of this increasing elaboration is the list compiled by Karl Julius Beloch (1854-1929), published in 1880 as part of the book Der italische Bund unter Roms Hegemonie. Beloch gives the probable location of colonies, estimates the size of their territory, and includes information derived from epigraphy, primarily on the titles of the colonial magistrates.14 fIG. 3. front cover of an early edition of Velleius’ Historiae Without doubt the best list available to Salmon was Romanae (after Rhenanus 1520). the list describing Roman that compiled by ettore Pais (1856-1939) and colony foundations covers just one and a half page. published in two articles (1923 and 1925) under the self-explanatory title Serie cronologica delle colonie 15 romane e latine dall’età regia fino all’Impero. this excellent study remains, even today, the most comprehensive since it also discusses the colonies founded during and after the Civil Wars16 and includes numismatic information.

11

12 13

14



Vell. Pat. 1.14-15. He legitimizes this excursus with the following words: “as related facts make more impression upon the mind and eye when grouped together than when they are given separately in their chronological sequence.” this scholarly tradition goes back at least to a Greek historical tradition (e.g., the list of Greek settlements on Sicily in thuc. 6.2-6.5, which in turn may be based on the History of Sicily by Antiochus of Syracuse). Ruperti 1838, 95-148 is devoted to such a list. Significantly, Ruperti ends his list in 100 B.C., precisely where Velleius does. Beloch 1880. Pages 135-58 provide a list for the latin

15

16

colonies founded between 495 (Signia) and 180 B.C. (luca). His emphasis on territorial sizes derives from his interest in demographics, which require a territorial parameter in order to calculate population densities. localizing the colonies made it also possible to draw maps pinpointing colonies and their territories. Pais 1923 and id. 1925, who in turn leans heavily on the work of De Ruggiero (De Ruggiero 1896, 96-130; republished with minor modifications in Diz. Epigr 2 (1900) 415-64, s.v. “Colonia” (e. De Ruggiero)). Pais was not the first to include these later colonies. His work is surely based on the extensive list published by Kornemann (Kornemann 1901).


lAtINAe fidenae Cora Signia Velitrae Norba Antium Ardea labici Vitellia Circeii Satricum Setia Sutrium Nepet

COlONIAe

PRISCAe COlONIAe

Romulus 501 495 494 492 467 442 418 395 393 385 393 382 382 COlONIAe

lAtINAe

COlONIAe CIVIuM

ROMANORuM

(MARItIMe) 338 338 334 329 328 314 313 313 313 312 303 303 299 298 295 295 291 289-283 289 273 273

Ostia Antium Cales tarracina fregellae luceria Saticula Suessa Aurunca Pontiae Interamna lirenas Sora Alba fucens Narnia Carseoli

lAtINAe

268 268 264 264 264 263 247 246 245 241 218 218 199 199 197 197 194 194 194 194 193 192 189

IuRIS

ARIMINeNSIS COlONIAe CIVIuM ROMANORuM (MARItIMe)

Ariminum Beneventum firmum Castrum Novum Pyrgi Aesernia Alsium Brundisium fregenae Spoletium Placentia Cremona Puteoli Salernum Volturnum liternum Sipontum Buxentum Croton tempsa thurii Copia Vibo Valentia Bononia COlONIAe CIVIuM

Minturnae Sinuessa Venusia Hadria Sena Gallica Cosa Paestum

ROMANORuM

(AGRARIAN) Potentia Pisaurum Saturnia Mutina Parma Graviscae

184 184 183 183 183 181 181 177 157 128 124 123 122

Aquileia luna Auximum Heba fabrateria Nova Neptunia Scolacium

tAB. 1. Salmon’s list and categorization of Roman colonial foundations.

Besides the wealth of their information, these modern lists also differ from the Velleian prototype in their attention to legal matters. While Velleius differentiates only between colonies founded before and after the start of the Civil Wars,17 modern scholars subdivide their lists by distinguishing between different types of colonies, notably the distinction between latin and citizen colonies.18 As a rule, the more explicitly argumentative parts of these studies concentrate on the correct interpretation and classification of the various types of colonies. Although Salmon hardly cites these scholars, he undoubtedly relied heavily on this tradition, as can be seen from the many similarities in the structure of his works and the attention he devotes to legal differences between different groups of colonies (tab. 1).19

17 18

19

He does not, however, discuss the latter category. In Ruperti’s book, the list is not subdivided by category. Instead, a discussion of the various legal statuses of colonies is offered in a separate chapter (chap. 4). His book is divided into two parts: first, a thematic part titled Commentatio and second, a historical part titled Historia, which describes the various colonial foundations. Pais, although certainly aware of the difference be-

tween latin and citizen colonies (in fact, he makes an important contribution to these legal definitions), does not group the colonies according to legal categories, but rather retains the chronological treatment in order of their foundation. Before Salmon, lists of colonies were available to an english-reading audience. A highly condensed list is found in frank 1959, 40-42, 59-60, the first edition of which was published in 1933.




From listing to explaining Roman colonies While it can be safely stated that Salmon’s work stands at the end of a long tradition20 and, in that sense, may be considered a genuine compilation of current knowledge, the same cannot be said of his analysis of the function of colonies. Since the Renaissance, a second line of research developed alongside chronologies (historiae) of Roman colonies. this line of research aimed at understanding the function of Roman colonization in both internal and external Roman politics (commentarii). from the start, scholars acknowledged that the Roman colonial program, independent of military objectives,21 had been crucial to several other developments in Roman society. Since Antiquity, one of the most influential scholars to emphasize the multifaceted role of Roman colonization is Carolus Sigonius (circa 1523 or 1524-84).22 In his De antiquo jure Italiae, published in 1560, Sigonius discusses Roman colonization at length and firmly establishes that, in addition to military purposes, a variety of demographic and social factors motivated the Roman fIG. 4. Carolus Sigonius (after Vedriani 1665, 150). colonization program.23 Sigonius was not the first to observe that colonies performed multiple functions (fig. 4). Niccolò Machiavelli (1469-1527) in his Discorsi on livy stressed that colonization, besides being an inexpensive way to keep conquered territory under control, had the additional benefit of relieving the mother city of excessive numbers of inhabitants.24 Another, slightly later scholar who stressed the multifaceted role of colonies is Justus lipsius (1547-1606). In his Admiranda, sive, de magnitudine Romana, published in 1598, lipsius devoted a long section of the first book, under the heading De coloniis, et modus fructusque deducendi, to the function of colonies (fig. 5). In particular, lipsius stressed their mimetic qualities vis-à-vis Rome and their important role in romanization.25 It was Sigonius, though, who developed the most detailed argument in a study that remains useful to readers today.26 especially important to the academic tradition that came after is the fact that he discusses Roman colonial history in the context of Roman agricultural

20

21

22 23



A rare example that continues this tradition is Bernardi 1973 for latin colonies. See also Northwood 2008, who in a paper on Asconius’ 53 colonies proposes to modify Salmon’s list of colonial foundations. Which are accepted unanimously, e.g., Cic., Leg. agr. 2.73 and in the Renaissance period Macchiavelli 1531, book 2, chap. 6 and lipsius 1598, book 1, chap. 6. 4. Sigonius 1560. De antiquo jure Italiae is volume 3 of the larger work De antiquo jure Romanorum, Italiae, provinciarum. the sections relevant to colonization are: de coloniis (book 1, chap 2) and de jure coloniarum (book 1, chap 2). He describes the six main motives as follows:

24 25

26

“[U]nam ad priores populos coercendos, alteram ad hostium incursiones reprimendas, tertiam stirpis augendae caussa, quartam plebis urbanae exhauriendae, quintam seditionis sedandae, sextam ut praemiis milites veteranos afficerent.” Machiavelli 1531, book 1, chap. 2. lipsius 1598, book 1, chap. 6; for the mimetic qualities, see especially lines 1-3. for an example of the authority this text still enjoyed in the 19th century, see Smith 1842, s.v. “colonia” and Ruperti 1838, 10-23, who offers an improved and more detailed overview of the various functions of the Roman colonization program.


history, connecting colonization directly to Roman agrarian reforms and thereby to the Struggle of the Orders.27 essentially, the classical paradigm to which Sigonius refers, and one that would remain very influential after him, considered the Roman colonization program an ingenious means of significantly increasing Roman manpower while at the same time diminishing the risks of social unrest in the city.28 Colonial land allotments offered the impoverished urban plebs a means of living and, most importantly, the economic resources needed to purchase the military equipment that qualified them to serve as soldiers. On this view, colonization significantly increased Roman military resources by transforming proletarii into assidui. Simultaneously, this policy of sending the poor to newly conquered lands relieved demographic pressure in Rome and reduced the risk of social unrest and secessions.29 Obviously, these beneficial effects could also be achieved by agrarian reform but the advantage of colonization was that it did so without upsetting the status quo in Rome and threatening concordia.30 the colonization of newly conquered lands left existing property claims untouched. Particularly in the case of latin colonies, the promotion of large numbers of proletarii to the property-owning class did not seriously upset the socio-political balance since these colonists lost their Roman citizenship. It is this long scholarly tradition of the sociofIG. 5. Justus lipsius (after lipsius 1615, frontispiece. Print made by Cornelis Galle). economic function of colonization that Salmon distances himself from and indeed intentionally marginalizes. It would, of course, be beside the point to blame Salmon for adopting a clear perspective.31 yet, in light of Salmon’s own claim to have produced merely a synthesis, it seems worthwhile to explore in detail how Salmon’s study in fact differed from other leading viewpoints, giving it a better-defined place in the history of Roman colonization studies. this may eventually do more justice to his work because it underscores its originality.

27

28

Machiavelli also briefly discusses colonies in the context of the agrarian laws (Machiavelli 1531, book 1, chap. 37). In his view, colonies were used by the nobles as a way to avoid drastic agrarian reforms, which, however, was not very successful, since the plebs did not want to go to colonies. Sigonius’ views, of course, should be seen in the context of humanist Renaissance thought, in which the ideals of Republicanism and the self-government of free Italian cities were paramount. the emphasis on internal social conflict in his presentation of Roman history may thus be explained partly in this light. Cf. Smith 2006, 68-69.

29

30

31

for a discussion of the primary sources that support this view, see Patterson 2006, 194-98. for a skeptical position, see for example Càssola 1988. lipsius also argues that that colonization had the additional benefit of acting as a social filter: the purest and best people remained in Rome, while the weak were sent away (lipsius 1598, book 1, chap. 6.2). Crucial for this view is liv. 3.1, which describes the establishment of the colony in Antium; on this passage, see Patterson 2006, 194-95. But see Crawford 1971, 253.




Military strategy and the topographical location of colonies Having established the general background of the different traditions in colonization studies – which will be further discussed later – we may now consider what is unique about Salmon’s own view. to be brief, Salmon argued that Roman colonization was motivated primarily by military strategy (cf. Bradley in this volume, also for the ancient sources in this regard). Significantly, Salmon presents this strategic interpretation as a natural fact. the opening line of his book reads, “the major role that the coloniae of Rome played in helping her to win and hold an empire is one of the important facts of history”.32 Although presented as a basic truth, this statement is in fact a scholarly position: a conclusion rather than a point of departure. from the better-annotated articles that led up to what is, after all, a very synthetic treatment in his monograph, it is clear that Salmon’s overriding attention to military strategy was grounded in a particular academic position and, as such, determinedly moved away from other interpretative schemes. In the following, the background against which Salmon adopted this position and his argumentation will be reviewed as well as the development of his thought over time. It can be demonstrated that Salmon’s approach is in fact based on a very specific conception of military strategy. this narrow understanding places an extraordinarily strong emphasis on topographical-strategic considerations. It comes at the expense of other perspectives, not only those unrelated to military considerations (such as the internal socio-economic motives just mentioned, and discussed in more detail below), but even perspectives based on other military-strategic explanations of Roman colonization. Although Salmon is well known in modern scholarship for his view of the role of colonies in Roman imperial strategy, the specific nature of his argument is less often appreciated, and the motives behind it remain unexplored. Salmon’s focus is decidedly narrow, even among military models explaining Roman colonization. Imperial success and the strategic function of colonies can in fact be construed in a variety of ways and with different emphases. for instance, in light of manpower requirements, demographic considerations may play and indeed have played a key role in scholarly debates. But also technological and moral aspects have been regarded as essential factors in Rome’s imperial success. However, Salmon does not pursue these avenues in any detail, but rather chooses to focus very specifically on strategic military topography. A key argument in Salmon’s view concerns the location of colonies in the landscape. According to him, from the earliest times onwards, Republican colonies were located at key strategic and well-defended points. even the sites of colonies that, according to Salmon, were founded by the latin league rather than under Roman auspices, such as fidenae and Suessa Pometia, both founded before 500 B.C., would have been selected on the basis of their strategic importance.33 fidenae, for instance, would have been established to hold “the line of the tiber against the etrusci and falisci” and Cora and Signia the line “against the Volscians”.34 this emphasis on strategic military topography is clear in almost all Salmon’s writings, including his famous monograph on Samnium and the Samnites (1967). It is hard to ignore that almost without exception, whenever the site of a colony is mentioned (whether already established or yet to be installed) adjectives or nouns suggesting its strategic importance abound. to give an impression of this, Ardea (442 B.C.) is a “stronghold”, Vitellia (395 B.C.) the “custodian” of a route, and Circei a “sentinel against Aurunci”. the colony of Satricum (385 B.C.) was a “key point”, just as the “twin fortresses” of Nepet and Sutrium were “the keys to etruria”.35 Indeed, in Salmon’s own words, all these early colonial sites were “skillfully chosen, and defense considerations were uppermost in their selection. […] they were in fact powerful bastions, and

32 33



Salmon 1969, 11. Id. 1953a, 101; id. 1953b, 123; he follows Enciclopedia Italiana di scienze, lettere ed arti (1931) 843, s.v. “la colonizzazione greco-romana” (P. fraccaro) for this view of the role of the latin league.

34 35

Salmon 1969, 42. Ibid., 42-43.


their strategic contribution was notable. they formed a network of fortresses, controlling river crossings, mountain passes, roads and tracks, and they could frustrate enemy combinations”.36 this emphasis on the military role of colonies – and their strategic positioning – came perhaps more naturally for the latin colonies founded after 338 B.C. Previous scholarship had already linked these colonies firmly to Roman expansion in Italy at that time. Salmon indeed describes the period of the Samnite Wars as “the golden age” of latin colonies.37 this view becomes already clear in Samnium and the Samnites, which is actually, in large part, a meticulous description of Samnite-Roman military conflicts rather than a description of the land, its people, and its customs. Here, the strategic rationale of colonies comes clearly to the fore. In the introductory description of the land, it is precisely the sites that will later be occupied by latin colonists that are described as exceedingly strategic in geographic and topographical terms. thus, “clearly Sora, controlling so sensitive a region, is a very strategic site”, whereas fregellae is also “an important nodal position”, and Venusia is “the key to the whole valley of the upper Ofanto”, just as luceria would have been for its surroundings.38 It is a fact that in the index to the book, which lists numerous towns and sites, Roman and indigenous alike, the subheading “strategic site (of)” is used exclusively for colonies.39 Salmon takes this argument so far as to claim that every single colony is strategically located and that this fact in itself proves their primarily military function.40 this automatically raises the question of what exactly constitutes a good strategic location. Of course, ‘strategic’ is not an objective and static quality but a subjective qualification that depends on a variety of factors that may differ in importance according to specific historical and local conditions. for instance, the suitability of a given location in the physical landscape and its relationship with existing infrastructure and possible routes depend entirely on the type of conflict or interaction, the character and size of the parties involved, current technology, and conceptions of landscape, safety, and hegemony. Among other things, this means that a given site may be very strategic at one time in history but not in different circumstances; the ‘strategic value’ of a particular site can hardly be considered constant and clear-cut. It would be impossible to tease out all of Salmon’s specific criteria, but it seems feasible to outline a set of recurrent features in his analysis. He specifically mentions “enemy frontiers”, “axes of advance”, “roads and tracks”, “river crossings”, “mountain passes”, and “sea ports” as strategic points of interest.41 examples of these features are seen in the quotations cited above. Salmon does not consider all of them equal in strategic importance but rather considers some more decisive than others. In his study of The strategy of the Second Punic War he argues that “all history proves that rivers are obstacles, but by no means insuperable barriers, to an attacking force. Mountains, on the other hand, are a very different proposition. they have always proved to be much more than mere temporary hindrances to military operations”.42 Apart from such considerations, the relationship of a site with specific regions and ethnic groups seems to be paramount in Salmon’s general assessment of its strategic value. With respect to the character of the conflict or the relationship between colonies and the area where they are located, Salmon generally emphasizes defensive rather than offensive considerations: colonies served as “defensive bastions, rather than as offensive springboards, although presumably they were very useful in either capacity”.43 However, in his typical

36 37 38

Salmon 1969, 43. Ibid., 57. Id. 1967, 19, 20. In id. 1969, the colony of Cales would have ‘dominated’ a communication route (id. 1969, 55), just as fregellae ‘controlled’ different routes (ibid., 57; cf. id. 1967, 212). luceria is described as a ‘bastion’ and a “powerful site on a hill-top […} the key to the eastern approaches to Samnium” (id. 1969, 58). Saticula (313 B.C.) is a ‘border fortress’ and Suessa a “stronghold controlling one of the roads to

39

40

41 42 43

Capua”; Interamna is located at “an important river crossing” (Salmon 1969 58-59). Aesernia: ibid., 417; luceria: ibid., 433; Sora: ibid., 443; Venusia: ibid., 446. Id. 1955, 64, in this case with reference to early colonies. Ibid., 64; id. 1969, 43. Id. 1960, 137. e.g. id. 1936, 54; id. 1955, 64; of course following a long line of similar statements, e.g. Abbott 1915, 366.




argumentation, it is clear that ‘defensive’ usually denotes controlling conquered areas and protecting them against local uprisings and enemy attacks; this is, in other words, a very particular view of defense. More generally, Salmon’s interpretation of the strategic role of colonies closely reflects Cicero’s famous statement about the location of colonies and their appearance as “bulwarks of empire, rather than towns of Italy”.44 Indeed, Cicero occupies a prominent position in supporting Salmon’s statement.45 there is, however, the risk of circularity in Salmon’s assessment of the strategic value of specific sites. for instance, Saturnia, founded in 183 B.C., was supposedly colonized because even though “it was not situated on the coast, it nevertheless did control an easy road to and from the littoral along the valley of the Albinia and was admirably placed to prevent raids on central etruria from the sea”.46 this argument may not in itself necessarily be untrue but it is too easily transposed to other sites, including non-colonial ones, to have much explanatory value. this risk of circularity applies even more so to cases in which Salmon’s assumptions about the purpose of colonies influences his interpretation of their precise whereabouts. for instance, when discussing the possible location of the early colony of Vitellia in his brief article on The Latin colonies of Vitellia and Circeii, Salmon proposes the site of the modern village of Valmontone: “A colony placed here, owing to its control of important roads, would be admirably placed to perform the task of any colony situated in Aequis, viz. protect what was the big bone of contention between Romans and Aequi, the pass of Algidus”; he then argues that, since this was disputed ground, “we should expect to find them [the Romans] sending out a colony to block the path […] the Aequi would use”.47 Similarly, in his work on Rome and the Latins, part I, Salmon argues that Suessa Pometia should be located at modern Cisterna di latina on the grounds that it occupies the kind of strategic site “one would expect of colonies”.48 thus far we have primarily considered Salmon’s arguments with respect to individual colonies. yet Salmon also held that the location of individual colonial sites might coincide with strategic considerations on a large scale. A prime example of this is the idea that Rome encircled the Samnites during the Samnite Wars by deliberately establishing latin colonies to form ‘an iron ring’ around the enemy.49 A similar notion based on the location of several sites is the supposed connection between maritime colonies and latin colonies. Salmon proposes that, before the Punic Wars, every citizen colony on the coast was ‘backed’ by a latin colony, citing the pairs of Antium and Norba, Ostia and Ardea, Sena and Ariminum, and Castrum Novum and Cosa.50 In similar fashion, and more frequently, Salmon envisages colonies as protecting major routes or roads as part of long-term strategic considerations, for instance in his assessment of the vulnerability of the via Appia before the founding of the colonies of Minturnae, Sinuessa, and Suessa Aurunca to ‘cover’ the route (cf. Bradley in this volume for a critical view). Incidentally, the importance Salmon attributes to roads is illustrated by his assessment of the via latina, about which he states “the side that firmly controlled and dominated it could menace the territory of the other in the most deadly fashion. It is like a doubleheaded arrow aimed directly at latium in the one direction and at Samnium in the other” (fig. 6).51

Roman colonization and WWII It is important to emphasize here that Salmon assumes both that the strategic qualities of specific sites and routes are fairly objective and static facts of history and that the Romans clearly and instantly recognized

44



Cic., Leg. agr. 2, 23, 73: “In that class of places, as in other parts of the republic, it is worth while to remember the carefulness of our ancestors, who established colonies in suitable places in such a manner that guarded them against all suspicion of danger, so that they appeared to be not so much towns of Italy as bulwarks of empire” ([…] idoneis in locis contra suspicionem periculi collocarunt, ut esse non oppida Italiae, sed propugnacula imperii viderentur; loeb trans. J. H. freese 1930).

45

46 47 48 49 50 51

After stating that every single colony would be strategically located (quoted above), Salmon cites only Cicero to back up his argument. Salmon 1936, 53. Id. 1937, 112. Id. 1953a, 101 n. 25. Id. 1967, 270. Id. 1955, 67. Id. 1956, 99.


fIG. 6. Map of Central Italy showing the consular Roman roads (after Salmon 1982, 42-43).

these qualities. As to the first assumption, it is striking how often Salmon refers to vastly different time periods and historical situations to prove his point. for instance, in discussing the importance of the via latina, Salmon adduces the argument that the via latina “was the route used by the allied forces for their advance up the tyrrhenian side of Italy in WWII”, just as it was used by Hannibal.52 Similarly explicit is Salmon’s inference that the strategic importance of the sites of Valmontone and Cisterna di latina are proven by the fact that they were strategically important in WWII. In this way, Winston Churchill becomes a main authority for Salmon to cite regarding the suitability of the locations proposed for these colonies.53 this frame of mind, oriented strongly toward a comparative historical-strategic approach, is highly significant. Most of Salmon’s explicit references to strategic examples from his own time, which abound in many of his original articles, have been omitted in his synthetic and frequently cited monograph. Salmon, a political commentator for the Canadian radio station CKOC during WWII, reflected every evening on the daily developments in the war, and was a keen observer of contemporary geopolitical and military developments. this interest clearly emerges in his work on the Roman world. Salmon explicitly emphasizes the potential of comparative history in his 1960 article on The Strategy of the Second Punic War. Somewhat ironically – since he blatantly contradicts Churchill’s dictum that “history may teach no other lesson than that men are

52

Salmon 1956, 99 n. 2.

53

Id. 1953a, 101 n. 25, citing Churchill 1952, 424-37, 528-39.




unteachable” – Salmon adduces various examples from contemporary and modern military history to support his interpretation of Roman strategic considerations during the Punic Wars. this ranges from comparing Carthaginian naval ambitions to the ill-fated German Operation Sea-lion against Britain in WWII, to comparing Rome’s failure to help besieged Saguntum to Britain’s refusal to help Poland in 1939.54 Moreover, the fact that the Romans are compared to the Allies in WWII and their enemies to the Germans is probably not coincidental. Salmon’s assessment of Roman conduct over their history is overall extremely positive. even his alleged and today almost proverbial pro-Samnite and anti-Roman position in Samnium and the Samnites needs important qualification: it is always clear the Romans will eventually prevail on account of their innate strategic insight, ambition, and above all character, or indeed instinct.55 there is no deliberate Grand Strategy in Salmon’s vision of Roman imperialism, although he imagines the end result as both inevitable and ultimately desirable.56 empire is rather the result of their “peculiar virtuosity” and instinctive seizing of opportunities.57 One conclusion from this overview must therefore be that the military-strategic and pragmatic character of Roman colonization as argued by Salmon, even if it is partly based on particular readings of the ancient sources, represents a deliberate scholarly position that is clearly influenced by contemporary geopolitical and military events.

Perspectives Salmon did not pursue I: colonies and Romanization One particularly interesting aspect of Salmon’s scholarly position is his – at least initial – rejection of the Romanizing role of the colonies of the Republican period in Italy. this view went decidedly against current opinions about the long-term role of colonies in the ascent and maintenance of the Roman empire. Here too Salmon deliberately engages with and attacks accepted scholarly discourse. One of the clearest statements appears in his 1936 article on Roman colonization: Nor does it seem […] probable that the colonies of this period were sent out either as romanising or as punitive agents. under the empire the task of the colonies may have been to romanise. under the Republic they were not intended to do so. […] the latin colonies in Italy were independent members of the Roman confederation possessing a very high degree of local autonomy. Sometimes they did not even use the latin language extensively. It is impossible to visualise them imposing Roman customs and institutions on the inhabitants of the surrounding countryside. It is, of course, true, and cannot be too much emphasised, that the colonies accustomed the Italians to regard Rome as the controlling and political centre of Italy. Herein their moral effect was very great. But beyond this they can scarcely be regarded as romanising agents. (Salmon 1936, 52).58

Before Salmon, it had been commonplace to mention both military and cultural impact – equally – as the rationale for colonization. this notion can be found, for instance, in frank Abbott’s 1915 article on “the colonizing policy of the Romans” according to which “the latin colonies served as military outposts and as centers of Roman influence”. Abbott adds that this policy was particularly and exclusively used in Italy.59 this view appears widely in works spanning the entire historiography of Roman colonization. Montesquieu (1689-1755), for instance, in

54 55



Salmon 1960, 132, 135. ultimately, the Samnites’ role in history is essentially twofold: to protect Rome from corrupting Greek influences (Salmon 1967, 400: “to keep the national character of the Romans true to itself”), and to act as a sparring partner in order to ‘bludgeon’ the Romans into shape for their great future: the Samnites “roused and sped them on the road to empire”. Id. 1967, 401;

56 57 58 59

on the ‘anti-Roman’ aspect of Samnium and the Samnites cf. frederiksen’s review (1968); for qualifications, Dench 1995; ead. 2004. e.g., Salmon 1967, 346; id. 1982, 1. Cf. e.g. Wilson 1971. e.g., Salmon 1955; quote from id. 1982, 2. Cf. id. 1955, 75. Abbott 1915, 366.


his Considérations sur les causes de la grandeur des Romains et de leur décadence (1734), following Jacques-Bénigne Bussouet (1627-1704), states that, besides relieving poverty in the Urbs, the goal of colonization was to “garder les postes principaux, et d’accoutumer peu a eu les peuples étrangers aux mœurs romaines” (fig. 7).60 the alleged civilizing mission of colonies in general is of course ubiquitous in Roman historiography.61 theodor Mommsen (18171903) discusses Roman colonies under the subheading “die Einigung Italiens” and emphasizes their cultural affinity to the mother city, Rome.62 Also, ettore De Ruggiero (1839-1926) describes the role of Republican colonies in 1900 as being founded “con lo scopo della difesa che della propagazione del romanesimo nelle regioni italiche”.63 for James Reid (1846-1926), in The municipalities of the Roman empire (1913), the cultural impact of Republican colonies was clearly a very important factor,64 and Plinio fraccaro (1883-1959) defined colonies famously as “il vero strumento della romanizzazione dell’Italia” in 1931.65 fIG. 7. Charles-louis de Secondat, baron de la Brède et Salmon could not disagree more. On the contrary, de Montesquieu (after laboulaye 1876, frontispiece). according to him, outside Roman territory Italy remained largely untouched by Roman culture until the Social War.66 If colonization had any cultural effect at all, it was, in his view, decidedly secondary and unintentional.67 In his 1953 article, summarizing the main reasons behind colonization – protection against enemy attacks, control over the conquered population, and ‘partly’ to relieve the urban poor – he does not even mention potential romanizing or civilizing effects, let alone goals, of colonization.68 Such a position is in accord with his pragmatic, topographic-strategic view of colonies but it should be emphasized that the possibility of an additional ‘cultural’ or acculturative role – also in an explicitly strategic sense, as a hegemonic method – was commonplace in previous scholarship. It is also surprising in light of Salmon’s political sympathies that he occasionally accepts very dubious authorities to support his point. Josef Göhler’s (1911-2001) study Rom und Italien, ominously published in 1939, is cited in support of the statement that colonization never was intended to promote romanization.69

60 61 62

63

64

Montesquieu 1734, 24; Bossuet 1681, 553. See, e.g., lipsius 1598, lib. 1, cap. 6.3. Mommsen 1912, 420-21, pointing out that “diese neuen Stadtgemeinden römischen Ursprungs, aber latinischen Rechts immer mehr die eigentlichen Stützen der römischen Herrschaft über Italien [wurden]”, and “durch Sprach-, Rechts- und Sittengemeinschaft an Rom geknüpft waren”. Diz. Epigr 2 (1900) 415-64, s.v. “Colonia” (e. De Ruggiero), 427. e.g., Reid 1913, 64: “Small, numerically, as the number of latin and Roman settlers in these colonies was, their influence on the regions around them was immense. the

65

66 67

68 69

local dialects everywhere gave way before latin, and the populations were in course of time prepared, by subtle changes of culture and sentiment, to accept and even to welcome complete absorption into the Roman state.” Enciclopedia Italiana di scienze, lettere ed arti (1931) 843, s.v. “la colonizzazione greco-romana” (P. fraccaro). e.g., Salmon 1962, 107. e.g., id. 1936; id. 1955, 64 (where, in considering the use of colonies to ‘overawe’ the local population, the emphasis is on control, not on acculturation), and 75 with esp. n. 49. Cf. below. Id. 1953a, 93. Id. 1955, 75 citing Göhler 1939, 156.




Göhler’s theoretical position on the point,70 however, is grounded in a curious and confused mixture of fascist, racist, nationalist, and primitivistic ideas with an occasional dose of common sense, viewing “Kulturpolitik als ein Kampfmittel” but also questioning the ability of lesser peoples to acquire high culture.71 from this perspective, colonies would have been “römische Sprachinseln” that were “Hüter des römischen Wesens und Garanten gegen eine Überfremdung” from Greek and etruscan influences.72 Apart from the conclusion that colonies were not meant to promote romanization, there is little in his work that seems to agree with Salmon’s ideas. Interestingly, Salmon seems to have modified his bold position over time. the key to explaining this change is twofold. first, Salmon’s view of what a colony was and, consequently, what impact it could have seems to have shifted. Second, and related to the first point, new archaeological studies appear to have influenced his thinking. from both the content of his arguments and his wording, it is clear that in fIG. 8. ernst Kornemann as student in 1889 (courtesy of most of his early studies Salmon envisaged colonies the Corpsarchiv Teutonia Gießen). as groups of men or garrisons sent to strategic sites, rather than as the founding ex novo of monumental new towns.73 Indeed, according to Salmon, colonists were sent to pre-existing towns previously built by the conquered Italic populations, even at sites for which there is, or was at the time, no specific evidence that they were inhabited before colonization.74 there is thus surprisingly little emphasis on, or even mention of, the founding of new towns, the built environment in general in the process of colonization, or the similarity of colonies to Rome. this is surprising, especially in light of the common idea that the essence of Roman colonies was epitomized by their institutional layout, reflecting the mother city of Rome. this had been a central tenet of Roman historiography for centuries. the image of colonies as closely echoing the societal organization and layout of Rome was widespread in scholarship from, as we have seen, as early as lipsius75 and appears prominently in some of the most influential studies.76 Besides Mommsen and others, Beloch for instance stated that “Die innere Organisation der Colonien natürlich nichts anderes [ist] als ein Abbild

70 71

72 73



As laid out in Göhler 1939, 23-30. Cf. ibid., 30: “Die Stämme Italiens […] wussten, dass ihre völkische Kraft dem grossen, ruhmreichen Rom diente und damit einem gewaltigen Reich, dessen Stütze sie sein dürften. Für Rom kämpft man und kämpft gerne”. Ibid., 147: Gaius Gracchus is described as “ein politischer Führer echt nordischer Prägung”. Ibid., 25. for a more detailed discussion of this conceptualization of colonies and its problems see the contributions by Pelgrom and Stek in this volume.

74

75

76

e.g., Salmon 1929, 13. the general notion that “the Romans […] avoided establishing colonies in new places” can also be found in Stephenson 1891, 20. Similar statements in Niebuhr 1873, 48-59. lipsius 1598, lib. 1, cap. 6, lines 1-3. See also McCulloch 1853, 419. See Sewell in this volume for a detailed discussion of Gellius’ notorious text (NA 16.13).


der Zustände der Mutterstadt”; ernst Kornemann likewise described colonies as “der Mutterstadt als Tochtergemeinde nachgebildete und von ihr abhängige[n] Ortschaft[en]” that would have copied the social organization of Rome (fig. 8).77 Rather than endorse the idea of the founding of new, both institutionally and culturally, Roman towns, Salmon emphasizes self-government exclusively. A recurring theme in his description of colonies is indeed their autonomy.78 Roman colonization, for Salmon, is the strategic positioning of self-governing communities on nodal sites.79 to some extent, this emphasis may also explain his aversion to the cultural unity of colonies imagined by previous scholars. Indeed, Salmon explicitly highlights the local diversity of colonies rather than their similarity before the Social War.80 On several occasions, Salmon, not noted for his use of archaeology, even adduces archaeological and epigraphic evidence to prove the diversity of Roman colonial realities, for instance pointing to colonial-period terracottas from Cosa that supposedly appear to be “of non-Roman rather than Roman types” and to the variability of colonial institutions and cults.81 Salmon explicitly rejects the idea that Gellius’ image of Roman colonies as small copies of Rome can be applied to the period before the Social War or to the military colonies of Caesar.82

A dubious role for archaeology? It may have been precisely archaeological studies, however, that reminded Salmon of Gellius’ image of Roman colonies. especially the discoveries at Alba fucens and Cosa in the late 1940s and 1950s, respectively by Belgian and American teams, generated much discussion about the process of founding cities and the apparent regularity and Romanness of these towns.83 this was undoubtedly an important point in the conception of the relationship between Roman colonization, romanization, and urbanization and also in the rise of the conception of a colony as synonymous with a town or city-state. Salmon’s pupil, Paul MacKendrick, whose studies explicitly draw attention to the physical aspects of Roman colonization and the archaeological evidence for them, may have been inspired by his involvement in these discoveries. MacKendrick participated in his first excavations at Cosa with frank Brown’s team and was very impressed both by the results and by the actual process of archaeological excavation.84 the regularity and supposed similarity to Rome of the colonial urban layout suggested, in his view,

77

78 79

80

Beloch 1880, 154; RE 4 (1901) 510-88, s.v. “Colonia,” (e. Kornemann), pp. 512, 584: they “copierten auch im inneren Aufbau ihrer Gemeinwesen die Mutterstadt Rom”. Niebuhr 1873, 51, acknowledges the Gellian image only for the earliest periods of colonization. e.g., Salmon 1936, 52; id. 1953a, 94; id. 1953b, 123. Contrast, for instance, Reid 1913, 60, who acknowledges that ‘sometimes’ an existing town was colonized, but emphasizes the “new Roman or latin municipal body, with its defensive walls and its autonomous institutions and its ‘territorium’.” e.g., Salmon 1936, 55: “the latin colonists belonged to municipal commonwealths (res publicae) enjoying a high degree of local autonomy. the colony could issue its own coins, had its own magistrates and its own constitution. Nor is there any need to suppose with Beloch that the constitution of a latin colony was a copy of the constitution of Rome. On the contrary, the epigraphic evidence reveals great diversity.” With, tellingly, n. 43: “After the Social War uniformity became the rule, and much later Aulus Gellius (NA 16.13)

81

82 83

84

could say of colonies that in relation to Rome “quasi effigies parvae simulacraque esse quaedam videntur.” Salmon 1955, 75 n. 47, citing an oral presentation at the 1954 AIA Meeting by l. Richardson. Id. 1936, 55 n. 43; id. 1955, 75. Brown, Richardson and Richardson jr. 1951; Mertens 1953. MacKendrick 1952, 139: “these planned communities, with their walls, their neat crisscross of streets, their fora and basilicas and temples, and their pattern of allotments […] [testified] already to the might and the majesty of the Roman name”; and on p. 140 “the actual process of the founding of a colony shows that orderliness and respect for legality which we associate with the Roman mind.” Id. 1960, 98-107, offers a wonderful insight into his personal experiences during the excavations at Cosa, which is enthusiastically characterized as follows: “A healthy site, an orderly plan, a water supply, strong walls, housing, provision for political and religious needs: the basic necessities are all there, at Cosa, and all as early as the founding of the colony”.




“a master plan made in the censor’s office in Rome”.85 As a result, MacKendrick incidentally questioned the solely strategic rationale behind Roman colonization advanced so forcefully by his tutor.86 the inclusion of an appendix to the introduction to Salmon’s Roman colonization under the Republic, titled “Cosa: a typical latin colony”, and its relative detachment from the rest of the book, can thus be explained. In many ways, this appendix summarizes the recent studies by Brown, even reflecting their textual structure. It is undoubtedly Cosa that led Salmon to conclude that “the Romans must have had a master plan or ‘blue print’” for founding colonies.87 In this study, Salmon is also markedly less reluctant than before to use Gellius to support the statement that latin and citizen colonies were, “throughout the half millennium and more during which these settlements were founded, […] as a rule typically Roman in appearance”.88 In later works, Salmon developed a much greater interest in the culturally unifying aspects and effects of Roman colonization and conquest in general. His 1982 book The making of Roman Italy is notorious in this respect, with its unabashed glorification of the ‘great achievement’ of Rome in “weld[ing] all the disparate elements into a single, unified nation”, described as Italy’s “ultimate destiny”.89 Salmon similarly places greater emphasis on the urban aspect of colonial communities. the new studies of the orthogonal town plan of Norba are adduced here, showing Salmon’s engagement with works on urbanism and town planning, such as those by ferdinando Castagnoli and John Ward Perkins that had appeared in the meantime.90 the conclusion drawn from this overall picture may therefore seem somewhat paradoxical, especially to some contemporary archaeologists. In the postcolonial age, archaeology actually contributed to a more starkly colonialist view of Roman colonization than the earlier view based on literary sources. It is also clear that Salmon’s modest statement in the introduction to his 1969 book, that he only sought to synthesize what was known about Roman colonies, is untrue.91 His is a very particular view of Roman history and the role of colonies in it that cannot be explained solely by the nature of the sources he used.92 In Salmon’s view, neither cultural hegemony nor a deliberate Grand Strategy were behind Roman imperial success. Rather, the key was in the Roman character or instinct in matters of military strategy and statecraft, formed and developed further by the practice of establishing self-governing communities. Not even the ancient Roman historians themselves, according to Salmon, would have entirely appreciated the ingenuity of the Romans’ strategic instincts. Instead, and tellingly, Salmon claims Machiavelli93 (the Machiavelli of Il Principe, that is) had rightly gauged the importance of Roman colonies.94

Perspectives Salmon did not pursue II: colonies and the Struggle of the Orders A second perspective marginalized by Salmon regards the socio-economic role of the Roman colonization program in maintaining internal stability in Rome. this omission is all the more striking because this perspective was gaining considerable popularity in the international scholarly community when Salmon entered the academic stage. Salmon began his academic career and developed most of his ideas in the first

85

86 87 88

89



MacKendrick 1960, 98, where Alba fucens is described as “a smaller and more orderly replica of Rome”. Id. 1952; id. 1960, 92. Salmon 1969, 38. Ibid., 27, with n. 33 citing Gellius. the local diversity is downplayed at p. 28, where “such details as terracotta ornaments, cornice mouldings and the like […] might show the effects of local influences. But in general the colonies, latin and later Citizen alike, must have displayed a consistently Roman aspect”. Salmon 1982, 39 and 56.

90

91 92

93 94

Castagnoli 1972 (=english translation of id. 1956b); Ward Perkins, 1974. Both cited in Salmon 1982, 52. Id. 1969, 11. livy’s descriptions of the founding of colonies are often embedded in non-military discourses, such as socio-economic (agrarian) issues (cf. below); even Cicero’s famous description of colonies as ‘bulwarks of empire’ is, it should be noted, part of his De lege agraria. Machiavelli 1532. Salmon 1955, 74-75.


half of the 20th century. At that time, the field of colonization studies was dominated by Italian scholars, who during the interbellum period, and after the famous German school had lost some of its influence, had moved the epicenter of Roman Republican studies to Italy and especially to Rome and Pavia.95 In Roman colonial history, the key figures were ettore De Ruggiero (18391926), ettore Pais (1856-1939), and Plinio fraccaro (1883-1959).96 De Ruggiero and Pais had studied with Mommsen in Berlin and on their return had introduced the fact-orientated, multidisciplinary German approach to Italy. Although they followed the critical juridical-historical methodology of Barthold Georg Niebuhr (1776-1831) and Mommsen and in many ways continued lines of research initiated by these famous scholars,97 they are often credited with adding historical sensibility to the discipline.98 As an American newspaper in 1904 described it: “[ettore Pais] blends Italian geniality with Germanic severity: it is the spirit of Mommsen relieved by the breadth of latin scholarship”; and some lines further on, “Italian scholarship fIG. 9. ettore Pais (courtesy is the broad tolerance, the free spirit, the untrammeled attitude of mind which of the photographic archive allows absolute objectivity of judgment.”99 Apart from these poetic qualities, a of the university of Pisa). notable distinction of these Italian scholars was their interest in Roman colonial history, which they discussed at length and as a separate topic, in contrast to their German teachers, or their english-speaking colleagues for that matter.100 this is not the place to speculate on what exactly might have sparked this particular interest, but the strong focus on the role of colonies in the unification of Italy and the recurrent emphasis on their quality as national instruments that strengthened the State clearly resonated with the political concerns of the Risorgimento and the early colonial ambitions of the new Italian state.101 Although we can easily recognize similarities between the works of these Italian scholars and Salmon’s studies, especially in their attempts to produce complete and detailed lists of colony foundations according to the established tradition, they differ in their interest in the role colonies played in Roman society. While in the earlier works of the most influential Italian historians the military-strategic rationale of colonies was also analyzed, there is a clear trend towards investigating and appreciating the role that Roman colonies played in domestic socio-economic and political developments. for example, Pais initially emphasized the strategic role of colonies – also discussing at length their strategic positioning – in his early studies (fig. 9).102 In his later work, however, he shifted his attention to their role in Roman internal politics.103 to the Italian scholars, the role colonies had played in the turbulent socio-economic transformation of Mid- and late Republican Roman

95

96

97

On Italian dominance in this field of Roman Republican philology see for example McDonald 1960. the great German scholars of the early 20th century like Beloch (1854-1929) and Weber (1864-1920) were more interested in sociological perspectives on ancient history than the Mommsenian critical-philological approach, a tradition the Italians continued with success. On De Ruggiero see Diz. Biogr. Ital. 39 (1991) 24448, s.v. “De Ruggiero, ettore” (M. elefante); on Pais: Polverini 2002, esp. 8-19; Gabba 2003; on fraccaro: Diz. Biogr. Ital. 40 (1997) 552-56, s.v. “fraccaro, Plinio” (e. Gabba). See Pais 1931, V-VI in which he expresses his gratitude to his teacher Mommsen, but at the same time tries to counter criticism that he is a “ripetitore del grande critico

98

99 100

101

102

103

Tedesco.” Diz. Biogr. Ital. 39 (1991) 244-48, s.v. “De Ruggiero, ettore” (M. elefante). Boston Evening Transcript, November 12, 1904, p. 23. this tradition of discussing Roman colonial history as a separate research topic goes back at least to Ruperti 1838 and is continued, for example, in De Ruggiero 1896 and especially Pais 1923; id. 1924; id. 1925. De Ruggiero 1896, 7. Also See Pais 1931, III-VIII, for clear references to nationalistic sentiments. On early Italian colonial ambitions, see finaldi 2009. See esp. chapter 13 on the use of classical history in the modern colonial discourse. e.g. Pais 1920, following the established tradition of for example De Ruggiero 1896, 38-39. e.g. Pais 1931.




society was much more interesting and challenging than their strategic role, which they took for granted. Since public lands were the crucial source of discord and strife in the Roman Republic it is actually not surprising that these Italian scholars often discuss Roman colonial practices in the context of the wider debate on the Struggle of the Orders and the related Roman Agrargeschichte. especially the students of fraccaro, namely, Gianfranco tibiletti, Aurelio Bernardi, and emilio Gabba took this tradition to its peak.104 this was the line of research from which Salmon particularly sought to distance himself and which he seems to oppose outright. He does so largely by marginalizing the socio-economic debate and focusing only on the strategic and topographic characteristics of colonial sites along the lines described above. Whereas an important strategic factor was generally accepted for the period between the end of the latin War to the Second Punic War, Salmon argues that the strategic rationale was paramount also in the prelatin War period and even in the 2nd c. B.C. up to the Gracchi. At the time, and especially for the latter period, economic rather than strategic motives were universally accepted as the main factor. As seen, Salmon does only rarely engage in open polemics.105 He makes an exception to this rule, however, in his 1936 article on Roman colonization from the Second Punic War to the Gracchi. Here, he argues at length that before the Social War the Roman colonization program was never intended as a means to provide for impoverished and destitute citizens and thus did not play an important role in the emancipation of the plebs and the Struggle of the Orders.106 One of Salmon’s most important arguments is that Rome had another, more effective tool for this purpose, namely viritane land-distribution programs, which he considers an entirely different matter.107 Also imperative, at least within his narrative framework, is the argument that it would be unrealistic to assume that Rome would send out urban paupers with no military experience to important military strongholds in enemy territory.108 Among the minor supplementary arguments he adduces, Salmon also holds that the rather small size of the colonies is at odds with the hypothetical objective of providing for the poor, and that Rome’s insistence on (re)colonizing unattractive places, such as Buxentum and Sipontum, suggests that strategic considerations were paramount.109 Salmon clearly reacts to such contemporary views in a series of articles on the rationale for Roman colonization. While he is arguing that “the principal purpose of colonies was undoubtedly strategic”, Salmon accuses previous scholarship of anachronism for suggesting commercial or economic motives.110 In his later work, Salmon gives more weight to economic and demographic motives, but he remains convinced that these are only secondary.111 It is beyond the scope of this article to discuss the validity of Salmon’s arguments in detail, but it is important to point out that his attempt to marginalize this, at the time, influential scholarly tradition was not very effective; most scholars today accept the importance of the Roman

104 105

106

107



e.g. Bernardi 1946; tibiletti 1950; Gabba 1984. Subtle criticism is expressed especially on Pais 1924; id. 1928. See especially Salmon 1936, notes 10, 15 and 60. Ibid, 51. Contra Pais 1924, 353-54; id. 1931, 120-22; tibiletti 1950. Interestingly, Salmon aims his polemic not so much at Italian scholars, but rather at the Cambridge professor James Reid (Reid 1913), who had argued that, while a defensive purpose was a likely motive before the Second Punic War, economic motives dominated the colonization program after the Hannibalic War. See also Salmon 1933, 32-33 for a similar but less explicit position. e.g., id. 1969, 95-96. He does not, however, address the issue that viritane distributions might potentially disrupt the socio-economic equilibrium.

108

109 110 111

Also Stephenson, 1891, 21, who seems to have influenced Salmon in this respect, makes this point very clear: “Colonies never became the means of providing for the impoverished and degraded until the time of Gaius Gracchus. When new territory was conquered, there went the citizen soldier.” Salmon 1936, 51. Id. 1955, 64. Ibid., 64, but later in the same article (p. 69) he argues more strongly that colonization was important for turning paupers into soldiers. In id. 1969, 15, 95-96 we again find that colonies are “incidentally useful for accommodating urban lacklands’ and ‘the colonies were intended to support a military programme, not to make provisions for the needy”.


fIG. 10. Plinio fraccaro opens the new academic year of 1949 (courtesy archivio casa Visintin).

colonization program in maintaining the socio-economic and demographic equilibrium of Roman society.112 Perhaps the most crucial counterargument against Salmon is that there seems to be a clear, negative correlation between recorded instances of social unrest in Rome and the intensity of colonial foundations.

The peasant republic A related discourse dismissed by Salmon is the distinguished academic tradition that linked the Roman colonial land-division program with the much-admired social and moral structure of Roman Republican society. Besides Rome’s famously open citizenship policy and its mixed constitution, the most important characteristic of the successful Roman social model was the fact that it was grounded, at least initially, on a set of civic virtues that generated a sense of unity and willingness in the individual to sacrifice private interests for the good of the community. this, of course, was crucial in wartime and for the stability of the Roman Republic in general. In Salmon’s day, this ideological perspective was voiced most clearly in the various studies by fraccaro (fig. 10) that are collected in the three volumes of his Opuscula (esp. part 1).113

112 113

e.g., Crawford 1971, 253; Oakley 2002, 18-22. fraccaro 1956; see especially the first 3 chapters, pages 1-81. fraccaro revives the powerful doctrine

that considers the moral structure of Roman society the critical factor behind its imperial success.




the view that civic virtues and imperial success were closely related had already emerged in the Roman period. It can perhaps be recognized most clearly in late Republican and Augustan decline narratives, which lamented the disappearance of precisely these civic values in Roman society.114 What exactly constituted Roman Republican civic virtues remained rather nebulous, but since Antiquity they nonetheless were epitomized in the concept of the soldier-farmer and were closely associated with peasant culture. Cicero, for instance, informs us that the early Romans “diligently cultivated their own lands, they did not graspingly desire those of others; by which conduct they enlarged the republic, and this dominion, and the name of the Roman people, with lands and conquered cities, and subjected nations”.115 Besides physical strength and the capacity to endure hardship, it is above all virtues such as austerity, autarchy, self-sacrifice, comradeship, and discipline that were recognized and appreciated in peasants and in rural culture more generally. this paradigm has remained influential throughout history both among idealist and realist politicians. even Machiavelli accepted that successful warfare and civic virtues were fundamentally connected.116 Roman history, however, had proven that the soldier-farmer could not be taken for granted and was seriously threatened by the corrupting forces of greed, selfishness, and lethargy (epitomized in decadence) that accompany imperialism. An effective society needed to formulate policies to protect itself from these harmful influences. In his Considérations (1734) – an essay on the causes of the greatness of the Romans and their decline – Montesquieu expresses the increasingly popular opinion that the chief way to achieve greatness was with radical agrarian policies. He explicitly states that, “the founders of the ancient republics had made an equal partition of the lands. this alone produced a powerful people, that is, a well-regulated society. It also produced a good army, everyone having an equal, and very great, interest in defending his country.” further on in the same chapter: ‘It was the equal partition of lands that at first enabled Rome to rise from its lowly position; and this was obvious when it became corrupt.”117 too much land, however, as Montesquieu explains in his De l’esprit des lois (first published in 1748) would make farmers lazy: “It is not sufficient in a well-regulated democracy that the division of land be equal; they ought also to be small”.118 Clearly, in this view, economic inequality introduced negative forces such as jealousy and greed. equality of basic economic resources thus becomes a central element in this philosophy of state organization. this philosophy had already acknowledged the superiority of the political form of the Republic because it was rooted in the ideal of political parity. According to this view, strong civic commitment was not only connected to rural life, but was intimately associated with egalitarianism and frugality. It is not hard to see how Roman colonial territories closely mirrored the romantic ideal of the perfect peasant republic. the perceived division of the conquered territory into moderately sized parcels for peers reproduced and reinforced the values of egalitarianism, comradeship, and austerity.119 Moreover, the importance of colonial

114

115



On Roman decline narratives, see evans 2008; see also Nelson 1998, chap. 3; and a short summary of the traditional decline narrative and its relationship to farming in Dyson 1992, 26-27. Paradoxically, the erosion of these moral values was also believed to be connected to imperialism: especially the influx of riches (luxuria) supposedly corrupted Roman society. In these accounts, the farmer-soldier ingeniously operates both as the victor of the Roman empire and as its victim. Cic., Rosc. Am. 50, (transl. yonge 1903). Another good example can be found in Cato: “[…] it is from the farming class that the bravest men and the sturdiest soldiers come, their calling is most highly respected, their livelihood is most assured and is

116 117 118

119

looked on with the least hostility, and those who are engaged in that pursuit are least inclined to be disaffected.” (Cato, Agr. Orig. 1.1, trans. W. D. Hooper and H. B. Ash 1934). Cf. Hulliung 1976, 142-155. Montesquieu 1965, ch. 3. Id. 1766, 66. (Ch. 6: “In what manners the laws ought to maintain frugality in a Democracy”). Crucial civic virtues, without which republics fall according the sociological history of Montesquieu. Such autarchical ideals can be recognized in the small allotment size, which enforce frugality and puritanism; qualities that in turn served as an antidote to the corrupting influences of imperialism and the eventual decline of militarily successful societies.


territories as the true habitat of the peasant-soldier had grown considerably since Niebuhr’s fundamental study of early Rome.120 Before Niebuhr, it was debated whether Roman agrarian laws technically applied to all land or only to newly conquered public lands. the more radical, enlightened spirits championed the first option, recognizing these agrarian laws as the antecedents of their own policies of the confiscation and redistribution of aristocratic domains; more moderate individuals, however, argued that the equal division of land applied only to newly conquered lands. Niebuhr would settle the controversy in favor of the moderate faction: he firmly (re)established the orthodoxy that the agrarian laws applied only to public lands.121 Diminishing the scope and impact of Roman agrarian legislation undermined the view that Mid-Republican Roman society at large was structured as an egalitarian peasant society and thus potentially placed greater emphasis on the Roman colonization program for (re)producing landscapes that supported the vitally important soldier-citizens and related civic virtues. Notwithstanding these appealing possibilities, Salmon does not seem to have been attracted to these ideological and socio-political studies. He does not highlight the potential importance of the Roman colonization program in creating and supporting the crucially important peasant-soldier communities and connected civic virtues. He addresses Roman colonial land-division practices only briefly122 and makes no allusion to the importance of this socio-economic and political arrangement for Roman imperial success. this is striking because, at the time of his ongoing research, momentous progress was being made in the study of Roman land-division practices. the findings from the new discipline of aerial archaeology were especially spectacular, since they included a vast amount of new data on Roman Republican colonial landdivision systems.123 Several fundamental publications by now legendary pioneers of the discipline were available to Salmon, such as Bradford’s Ancient Landscapes. Studies in field archaeology (1957) and Castagnoli’s Le ricerche sui resti della centuriazione (1958).124 these impressively evenly divided landscapes fitted perfectly well with the ideal of the peasant republic as outlined by for example Montesquieu (cf. also Bradley in this volume, fig. 1). Moreover, the findings from aerial archaeology renewed admiration for the Romans’ colonial engineering and organizational skills and also seemed to corroborate, or at least made more comprehensible, the cryptic and controversial treatises of the Roman land surveyors. these treatises, which deal at length with the practical and religious organization of colonial territories, had remained understudied125 since their technical style made them difficult to interpret. Now that the Roman land-surveying practices they described had been attested empirically, interest in the writings of the land surveyors peaked. this surge of interest resulted, among other things, in the publication of studies such as Oswald Dilke’s The Roman Land Surveyors: An Introduction to the Agrimensores in 1971, which made these technical treatises accessible to a broader public.126

120 121

122

123

Niebuhr 1873 (first published in 1811-12). On this, see Momigliano 1982; Ridley 2000; and Rich 2008, 521-43 with further references. Salmon 1969, esp. 24-24 and notes 110 and 111. It is significant that Salmon discusses colonial allotments mostly in footnotes. the Danish naval officer Christian tuxen falbe (falbe 1833) is usually credited with the first detection of a Roman land-division system. falbe recognized a 20x20 actus grid in the territory of Carthage (cf. tozzi 1984, also for an overview of other early 19th century studies). In the early 20th century, important progress in this field was made by fraccaro (e.g., fraccaro 1939; id. 1940). A collection of his studies of Roman cadastral systems can be found in Opuscula III (fraccaro

124

125 126

1957). On his work, see also Attolini 1984. His work was continued by his student Castagnoli (cf. below). Bradford 1957; Castagnoli 1958. Other pioneering publications from this period include Bradford 1949; id. 1950, Castagnoli 1953-55; id. 1956a; and various studies by Chevallier (e.g., Chevallier 1960; id. 1961; id. 1962). to get an idea of the state of research at the time of Salmon’s works, centuriation grids had already been recognized in the territories of the following Mid-Republican colonies: tarracina, Minturnae, Antium, Puteoli, Salernum, Pisaurum, Parma, Mutina, Auximum, luni, lucca, luceria, Cales, Cosa, Paestum, Ariminum, Alba fucens, Isernia and Beneventum, and most of the colonies in the Po Valley. e.g. Dilke 1962, 170. Id. 1971.




Although Salmon was clearly aware of these studies, he bluntly dismissed their impact by briefly stating that the centuriation systems detected from the air could not be dated properly and might easily belong to later viritane land-division programs.127 He does not, however, back up this skepticism with an analysis of the arguments used to date the newly discovered centuriation grids to the early colonial period.128 this failure to engage seriously with an exciting new data set is striking. to many of Salmon’s contemporaries, these findings seemed to demonstrate more clearly than ever the impressive and extraordinarily rigid organization of colonial territories, which not only seemed to differ dramatically from the organization of the old ager Romanus but also unmistakably expressed the exceptional power of a society based on the principles of discipline, order, and equality. the two quotes given below clearly illustrate the impact these discoveries made on contemporary scholars:129 the forceful imprint of the elaborate gridded road-systems […] can still be traced across some thousands of square miles on both sides of the central Mediterranean. In origin, most of these systems were carved out of territories raw from conquest, and even now, in retrospect, their appearance deeply stirs the imagination, – so boldly artificial was the conception and drastic the creation as compared with any earlier man-made landscape in this region. (Bradford 1957, 145). Ces dernières [i.e. les centuriations] avaient en premier lieu valeur politique : à l‘origine au moins de la colonisation, Rome fit table rase du passé en imposant à ses conquêtes un cadre nouveau : soit par indifférence, soit par mépris, elle ignorait l’organisation administrative préexistante et marquait ses droits de propriété éminente en toisant sa conquête : la prise de possession est comme gravée dans le sol ; selon le principe politique éternel : ‘diviser pour régner’, on voit la centuriation isoler les zones de résistance, c’est-a-dire surtout les régions montagneuses, s’insinuer même dans les vallées et morceler les massifs dont elle ronge les premières pentes. (Chevallier 1961, 64).

Salmon’s marginalization of these qualities of colonial landscapes may perhaps be explained in part by his personal background. Salmon had strong anti-Soviet and more generally anti-Russian political views, which may go some way towards explaining why portraying colonies as more or less egalitarian societies did not appeal to him.130 then again, one might also interpret his practical, strategic outlook as a refreshing and deliberate reaction to the idealizing social theories of the Romantic era, which lay shattered after the horrors of the two World Wars. In any case, his attempt to undermine the importance of these new findings squares neatly with his hidden agenda to demonstrate that it was above all Roman military strategy, and especially the Roman instinct for knowing what areas are the most important to control, that had won Rome its empire. It is telling that in his brief discussion of the evidence for centuriation systems Salmon questions its association particularly with Mid-Republican Roman colonization. Since the Romans had already proven to be the strongest power in the Mediterranean by the late 3rd c. B.C., the secret of their success must be found in Roman society and military strategy before the Punic Wars. By questioning the existence of these impressively ordered and monumental colonial landscapes in the early colonial period, Salmon thus challenged the view that they played an important role in Roman imperial success as the habitat of the vitally important soldier-farmer and as a potent demonstration of Roman power and organizational skill to the conquered, thus dampening any aspirations to revolt.

127 128

129



Salmon 1969, 23-24. this is not to deny that the dating of early Roman land-division systems is notoriously difficult and controversial (cf. Pelgrom 2008, 358-67). More recently, cf. Purcell 1990, 15-17; Quilici 1994, 127, 130; Campbell 1996, 81.

130

In 1947, Salmon gave a paper at the empire Club of Canada, entitled “the Making of the Peace” in which he clearly expresses his anti-Russian political views (https:// speeches.empireclub.org/60827/data?n=18).


Aim and structure of this book: bringing a fragmented research field together this excursus on Salmon’s place in the wider scholarly debate on Roman colonization has revealed not only that Salmon adopted a very specific outlook, but also, and perhaps more importantly, that his monograph, despite his claim to the contrary, is part of just one of many different directions the study of Roman colonization has taken over time (fig. 11). the main trend that can be discerned from at least the late 19th and early 20th century is the ever-increasing specialization and fragmentation of colonization studies. Since the early comprehensive attempts of the Renaissance period to understand the role of colonies in Roman society, different aspects of Roman colonization have increasingly been discussed in separate discourses. this trend of increasing fragmentation, we would argue, was not reversed by Salmon. Rather, he pursued one particular perspective to the extreme. this fragmentation in colonization studies still dominates the research field today. Issues that are central to the character of Roman colonization continue to be studied in separate disciplines: from Roman historiography, urban archaeology, fIG. 11. edward togo Salmon in the McMaster Art Gallery architecture studies, landscape archaeology, Roman 131 religion studies, to Roman law. Notwithstanding coin room on March 28, 1988 (courtesy of McMaster Museum of Art, McMaster university, Hamilton, Ontario). the obvious importance of many new studies in these various realms the impact on the general understanding of Roman colonization and significance for understanding Roman imperialism has yet be fully appreciated. this is only possible by adopting a more holistic approach, and drawing together the historically diverged lines of enquiry in colonization studies. We realize that the complexity and specialized nature of the various discourses in which colonies are studied, together with the vast amount of new archaeological and, to a lesser extent, epigraphic data collected in the last decades, makes any attempt at an integrated, comprehensive treatment vastly more demanding than it was in Sigonius’ time. What for Salmon would have been an ambitious challenge, today

131

the historiography of Roman colonial studies after Salmon 1969 has been discussed in detail in several important and easily accessible studies and will therefore not be reproduced at length here. for historical issues and the contribution of archaeology, seminal publications include the proceedings of the Acquasparta conference in 1987 entitled La colonizzazione romana tra la Guerra latina e la Guerra annibalica (published in 1988 in DialArch 6); Crawford 1995; torelli 1999; Bradley and Wilson 2006;

and several contributions in De ligt and Northwood 2008. for the role of colonies in Roman rural history, land division, Roman law, and internal politics, see Settis 1984; Chouquer et al. 1987; Gargola 1995, Schubert 1996; Hermon 2001; Campbell 2000; laffi 2007; Roselaar 2010. for synthetic studies that address colonial urban topography, see Sommella 1988; fentress 2000; lackner 2008; Sewell 2010. See also Broadhead 2007; Stek 2013, 344-45, and its reading list on page 353.




is virtually impossible. Nevertheless, we believe that at this point a step in this direction is indispensable, and has potential. to this end, this volume brings together recent insights from a range of different academic traditions, thereby also lifting language and cultural barriers. It should be emphasized that this is only a first step in this direction and that it is very far from a conclusion. Rather, the aim is to outline a possible and promising agenda for future research and to explore the potentially productive interplay between different emerging research areas that are currently isolated. the structure of this volume is therefore arranged by theme, highlighting the relative strengths of different disciplinary approaches and methods for understanding the development and character of different aspects of Roman colonization and Roman imperialism. We have outlined four main thematic fields, which correspond with the four parts in which this volume is divided.

I. Contextualizing Roman Republican colonization. Backgrounds, definitions and comparanda the first part of the book explores the general character and function of Roman Republican colonization in relation to emerging ideas of the social structure of early Rome and the character of early Roman expansion and imperialism. Nicola terrenato’s contribution considers the relationship between Roman imperialism and colonization by focusing specifically on the structure of Central Italian society in the early and Mid-Republican periods and long-term continuities and changes therein. terrenato attacks the once popular doctrine of William Harris132 that elite competition in the city of Rome was the primary motor behind Roman imperialism. In contrast, terrenato notes that this argument fails to account for the sudden burst of expansion at the beginning of the 4th c. B.C. Importantly, however, terrenato accepts Harris’ emphasis on the role of individual agency. Rather, he challenges the claim that a single center, Rome - and the Roman Senate - drove the imperialistic process. In this important paper terrenato takes a closer look at some of the family groups involved, both in Rome and in incorporated communities, in the 4th and 3rd c. B.C. expanding on the particular history of the Plautii, he argues that expansion, at least initially, resulted from the sum total of several independent family agendas both from within Rome and from communities outside it, which aimed to benefit a specific factional group rather than a political abstraction like ‘the Roman empire’. Guy Bradley’s contribution also explores the impact of new insights in Roman imperialism on our understanding of Roman colonization in the Republican period. As we just saw, Roman imperialism cannot be understood merely in terms of internal factors as Harris proposed. Also Roman reactions to neighboring states were decisive and the pace and direction of conquest was shaped both by favorable and unfavorable factors. Scholars such as Arthur eckstein133 and John Rich134 have emphasized the genuine unpredictability of Roman behavior, and especially the element of fear. In particular, eckstein has stressed the anarchic nature of interstate relations in Italy itself: Roman control over Italy was less certain than hindsight might suggest. the picture now emerging would seem to leave less room for long-term planning and strategy – precisely the line pursued by Salmon and later emphasized, for instance, by filippo Coarelli,135 whereby a clear relationship between colonies and, for example, road building is seen as part of a master plan behind Roman colonizing movements. Bradley takes another approach in his paper, exploring how colonial foundations and related infrastructure projects such as roads depended on the initiative of individual politicians and generals and the Senate’s response to them. the uncertainty about the nature of the Senate before the lex Ovinia suggests that the late 4th c. B.C. is a critical period, when a more permanent Senate emerges with a highly competitive mixed aristocracy. this development had, Bradley argues, a decisive effect on the development of long-term strategy.

132 133 134 135



Harris 1979. eckstein 2006. Rich 1993. Coarelli 1992. Although it should be noted that Salmon actually believed more in Roman character,

intuition and destiny as main factors, than their having a masterplan, cf. above.


from imperialism we move to the issue of socio-political organization, with two papers investigating the axiom that Roman colonies of the latin right were city-states. Reviewing the historical evidence, Jeremia Pelgrom argues that the conventional understanding of colonies as territorial states with clearly defined topographic borders is not as self-evident as is often believed. New legal studies have reopened the door for considering different forms of colonial organization in which social or other criteria, rather than territorial boundaries, were decisive for determining colonial membership. In this model, colonial jurisdiction is restricted to the members of the colony and the fields they farmed. A colony may have possessed a unified territory, but this is no longer strictly necessary, and different, patchier scenarios are possible. tesse Stek proposes one such different and patchier scenario for Roman colonization by analyzing the evidence of cult places and rituals for the traditional city-state model of Roman Republican colonies of the latin right. exploring the value of rural cults and cult places for our understanding of colonial socio-political organization, he argues that there is no evidence from Mid-Republican colonies that colonial cult places served to demarcate territory, as has habitually been envisaged on the basis of the model of Rome and other (etruscan, Greek, etc.) city-states. the supposed parallel to the situation of Rome itself cannot withstand close scrutiny. Stek proposes instead that the pattern of sacred and ritual sites in colonial territories points rather to a correlation between cult places and rural settlement nucleation than to the territorial demarcation of the hinterland of the colonial urban center. the socio-cultural significance of this rural nucleated pattern – normally not associated with Roman culture but rather with indigenous, Italic society – is illustrated by showing both the local significance of rural cult places and their relationship to the colonial town center. luuk de ligt discusses the so-called maritime colonies, citizen colonies located in coastal areas. these colonial communities differ in many ways from other forms of Roman colonies. for one, they were on average ten to twenty times smaller than contemporary latin colonies. In their outward appearance, they presumably resembled military forts more closely than monumental cities. Interestingly, members of these colonies were exempt from legionary service, the so-called vacatio militiae, which has been a pivotal notion in explaining the function of this type of colonization, as well as the purpose of colonization in general. De ligt’s discusses this atypical situation and offers an original explanation for the peculiar legal and military position of these colonial communities.

II. Colonial landscapes. Colonists and natives shaping the urban, natural and social environment the second part of this volume is dedicated to colonial urbanism in relation to the organization of the rural areas of Roman colonies. It opens with an important discussion by Jamie Sewell. Sewell considers the vexed model-replica theory of Roman colonization studies, in which Rome is seen as a model for her colonies in their urban lay-out, and proposes a different way of understanding the adoption and adaptation of Roman, Greek, and other traditions of city planning by investigating the built environment of colonial towns. As noted above, Gellius’ statement about colonies as ‘small images of Rome’ is often cited in discussions of the urban and institutional character of latin colonies, yet the belief that colonial towns slavishly replicated metropolitan political institutions and cults has, for good reasons, recently been questioned or refuted.136 Sewell rightly states, however, that to argue whether the colonies were or were not replicas, likenesses, or images of Rome does not help us understand colonial towns. Drawing on the concept of creative adaptation, he investigates the built environment of Mid-Republican colonies’ primary settlements by examining the various Roman and foreign influences that shaped their physical forms. Notably, Sewell explores the similarity and contemporaneity of the expansion strategies attributed to Rome and to Philip II of Macedonia and proposes a historical link between them. frank Vermeulen gives an overview of new archaeological evidence and its significance for understanding Roman strategies in incorporating the Picene area (modern Marche) and related

136

See esp. Bispham 2006.




urbanization developments in that region. In Picenum, cities developed unevenly and often appeared relatively late, seldom before the 3rd c. B.C. According to Vermeulen, however, Roman intentions regarding the urbanisation of late Republican central-Adriatic Italy are clearly recognizable: a mostly linear network of new towns in strategic positions along the coastline is connected to an evenly spread system of incorporated and expanded indigenous centers in the interior. Drawing on new data from both surveys and excavations in four urban centers of northern Picenum (the colony of Potentia and the municipia of Ricina, trea, and Septempeda), Vermeulen shows that even if these towns occupied a relatively small area, their basic infrastructure is comparable to most important cities in Italy. Working from a very different, more legal-institutional angle, Michel tarpin reconsiders the relationship between native inhabitants and colonists and the ways in which this relationship may have shaped the legal and administrative colonial landscape. tarpin starts from the hypothesis that interaction between colonists and natives was very diverse and was influenced by a wide variety of factors, including both local conditions and traditions and also by changing Roman requirements and strategies in different regions and time periods. In tarpin’s view, modern representations of colonial realities have often overlooked such specific local and temporal developments and instead have tended to reduce them to a timeless ideal type of Roman colonial realities. taking a different approach, tarpin attempts to reconstruct the historical superimposition of different layers of legal, administrative, and social significance, shedding light on changing attitudes to colonists, natives, and others in colonial landscapes. In particular, he demonstrates how the diversity of the Roman territorial lexicon reflects different arrangements between Roman colonists and conquered or incorporated communities and the establishment of new communities within Roman colonial territories. Adopting a socio-economic approach, ella Hermon investigates the relationship between Roman colonization and the management of natural resources and the environment at large for agricultural and safety purposes. In discussions of the colonial hinterland, focus has usually been on land-division systems, often in relation to the social structure of colonial communities. On the basis of epigraphic and literary evidence, however, Hermon argues that much of the colonists’ energy was used for land improvement strategies and for securing flood zones. this interest in maximizing and controlling natural resources, resulting in more complex and irregular physical and legal-administrative landscapes than usually imagined, can be connected to rising demographic pressure on the environment during the late Republican period. from a landscape archaeological perspective, Peter Attema, tymon de Haas and Marleen termeer investigate the character of Roman colonization in the Pontine plain, a lowland area very close to Rome, in the early and Mid-Republican periods. Historical sources relate that various colonies, such as Norba, Setia, and Circeii, were created in the area in the early Republican period. these settlements occupied strategic positions dominating the entire plain. long-term intensive field-survey projects have revealed traces of rural settlements dating to the pre-colonial and colonial phases. these findings make it possible to view the establishment of these early colonies in the context of long-term urban and rural developments in the region. the authors argue that the Pontine region can be regarded as a laboratory of early Roman colonization in the sense that different strategies of colonization and incorporation were first tested in this area during this historically formative period. using a similar approach in South Italy, Maria luisa Marchi analyzes the development of the latin colonies of Venusia and luceria against their indigenous background. Basing her research on years of intensive field surveys in these areas, Marchi demonstrates how diversely colonial landscapes could develop in similar chronological and geographical frameworks. At the same time, she recognizes the strong impact Roman settlers had on the landscape, sometimes visible in clearly new settlement organization forms. the paper by Giovanna R. Bellini, Alessandro launaro and Martin Millett offers a detailed examination of earlier, partly unpublished studies of the latin colony of Interamna lirenas and explores how this data affects our general understanding of Roman colonization and the Roman economy.




the authors argue that although these previous studies have made very important contributions, essential lacunas in our information remain. It is of crucial importance to conduct new fieldwork and to improve our knowledge of local material culture.

III. The religious dimension of Roman colonization the third part of this volume is dedicated to colonial cults and religious practices. It focuses on the role of colonial cults in promoting social cohesion within the newly established colony and in defining relationships with other communities and on the role of magistrates in the decision-making process. Marion Bolder-Boos explores the role of tutelary or patron deities for colonial communities in both newly established colonies and pre-Roman cities where colonies would have been established later on. Although, again as a consequence of the Gellian model, most scholars studying colonial cults have focused their attention on the Capitoline triad, colonies often had tutelary deities of their own. Bolder-Boos describes the variety of these tutelary deities and their respective backgrounds in the context of individual colonial communities. In contrast, Andrea Carini focuses on the role of a single god in different Roman colonial contexts: Apollo. traditionally, Apollo has attracted the most attention in Greek colonial projects and contexts, whereas for Italy the association between Apollo and colonization has been studied far less systematically. Carini therefore takes up the evidence for Apollo in colonial settlements in Italy, including evidence that is often rejected or ignored, to conclude that Apollo, frequently together with Diana, also played a crucial role in the religious formulation of colonial communities in Italy. Daniela liberatore presents fascinating new evidence from the sanctuary of Hercules in the forum of the colony of Alba fucens (founded 303 B.C.). Recent excavations have revealed an exceptionally rich and well-preserved set of ceramics, votives, statuettes, altars, and inscriptions. this evidence sheds important new light on the relationships between natives and colonists and on the role played by cult places in negotiating these relationships. Many small votives and statuettes clearly refer to older ritual traditions from the region, while some ceramic forms and altars appear to have been introduced after the colonization of the area. Notably, the inscriptions on some altars are in latin but mention local family names. liberatore gauges the implication that this cult place functioned as a node in the new colonial reality by putting it into the broader perspective of regional cult places.

IV. The creation of Roman centrality the fourth and final part of the volume consists of two papers that discuss the importance of early Roman colonial models in slightly later periods. this section reopens discussion of how, and in what historical contexts expressions of Romanness were constructed, thus counterbalancing recent postcolonial endeavors that are arguably excessively deconstructive. Mario torelli explores the flexible character of the religious and ideological ties of colonies to Rome by discussing the adoption and adaptation of the Capitoline model in imperial cities. In A.D. 100 lepcis Magna, a flavian city with latin status, became a Roman colony, an event that apparently was not celebrated with special emphasis. Although a Capitolium temple, the building par excellence to signal colonial status in imperial times, was expected to be found at the site, archaeologists have not been able to identify any of the numerous temples located in the central area as the lepcitan Capitolium. Ingeniously, torelli argues that an official cult place for the Capitoline triad indeed existed, but in a non-traditional form. Building on this new identification, torelli discusses a series of Capitolia with exceptional layouts and their implications for our understanding of colonial religion. lastly, Simone Sisani discusses the intricate relationship between the historiography of early Rome and Roman colonization by detangling the traditional connection between the boundary of the pomerium and the course of the city walls traced by a plough at the founding of the city (sulcus primigenius). In literary sources on ancient Rome, the pomerium and city walls are often closely associated and even

ďœłďœˇ


equated with one another. from other ancient Italic towns, however, we know that they are two fundamentally different types of boundaries. In this chapter, Sisani proposes that the confusion in ancient and modern texts alike is due to colonial practices in the late Republican period, when no separate pomerium was marked out, but rather coincided with the city walls. the ‘Romulean’ colonial model practiced in later times was thus projected back onto the ancient city of Rome itself. together these essays provide a wealth of new perspectives on Roman colonization and offer important new insights on this wide-ranging topic. We are confident that the multi-disciplinary collection of recent theories and evidence presented in this volume will contribute to a more comprehensive understanding of Roman colonization and imperialism.

References Abbott, f. f. 1915. “the colonizing policy of the Romans from 123 to 31 B. C.,” CP 10-4, 365-80. Attolini, I. 1984. “Gli studiosi della centuriazione: fraccaro,” in S. Settis (ed.), Misurare la terra: centuriazione e coloni nel mondo romano (Modena) 174-78. Beloch, K. J. 1880. Der italische Bund unter Roms Hegemonie. Staatsrechtliche und statistische Forschungen (leipzig). Bernardi, A. 1946. “Incremento demografico di Roma e colonizzazione latina dal 338 a.C. all’età dei Gracchi,” Nuova Rivista Storica 30, 272-89. Bernardi, A. 1973. Nomen Latinum (Pavia). Bispham, e. 2006. “Coloniam deducere: how Roman was Roman colonization during the Middle Republic?” in G. Bradley and J.-P. Wilson (edd.), Greek and Roman colonization: origins, ideologies and interactions (Swansea) 73-160. Bossuet, J. B. 1681. Discours sur l’histoire universelle (Paris). Bradford, J. 1949. “Buried landscapes in Southern Italy,” Antiquity 23, 58-72. Bradford, J. 1950. “the Apulia expedition: an interim report,” Antiquity 24, 84-95. Bradford, J. 1957. Ancient Landscapes. Studies in field archaeology (london). Bradley, G. and J.-P. Wilson (edd.), Greek and Roman colonisation. Origins, ideology and interactions (Swansea). Broadhead, W. 2007. “Colonization, land distribution, and veteran settlement,” in P. erdkamp (ed.), A companion to the Roman army (Malden, MA/Oxford), 148-63. Brown, f. e., e. H. Richardson and l. Richardson jr.



1951. “Cosa I, History and topography,” MAAR 20, 5-113. Campbell, B. 1996. “Shaping the rural environment: surveyors in ancient Rome,” JRS 86, 74-99. Campbell, B. 2000. The writings of the Roman land surveyors. Introduction, text, translation and commentary (london). Càssola, f. 1988. “Aspetti sociali e politici della colonizzazione,” DialArch 6-2, 5-17. Castagnoli, f. 1953-55. “I più antichi esempi conservati di divisioni agrarie romane,” BullCom 75, 3-10. Castagnoli, f. 1956a. “la centuriazione di Cosa,” MAAR 24, 149-65. Castagnoli, f. 1956b. Ippodamo di Mileto e l’urbanistica a pianta ortogonale (Rome). Castagnoli, f. 1958. Le ricerche sui resti della centuriazione (Rome). Castagnoli, f. 1972. Orthogonal town planning in antiquity (Cambridge, MA) Chevallier, R. 1960. “la centuriazione e la colonizzazione romana nell’ottava regione augustea emilia Romagna,” L’Universo 40, 1077104. Chevallier, R. 1961. “la centuriation et les problèmes de la colonisation romaine,” Études rurales 3, 5480. Chevallier, R. 1962. “Notes sur trois centuriations romaines: Bononia, Ammaedara, Vienna,” in M. Renard (ed.), Hommages à A. Grenier. Latomus 58 (Brussels) 404-17. Chouquer, G., et al. 1987. Structures Agraires en Italie Centro-Méridionale. Cadastres et Paysages Ruraux (Rome).


Churchill, W. S. 1952. The Second World War (london). Coarelli, f. 1992. “Colonizzazione e municipalizzazione: tempi e modi,” DialArch 310, 21-30. Crawford, M. H. 1971. “Republican colonization. Review of e. t. Salmon: Roman colonization under the Republic,” CR 21-2, 250-53. Crawford, M. H. 1995. “la storia della colonizzazione romana secondo i romani,” in A. Storchi Marino (ed.), L’incidenza dell’antico. Studi in memoria di Ettore Lepore (Naples) 187-92. Dench, e. 1995. From barbarians to new men. Greek, Roman and modern perceptions of peoples of the Central Apennines (Oxford). Dench, e. 2004. “Samnites in english. the legacy of e. togo Salmon in the english-speaking world,” in H. Jones (ed.), Samnium. Settlement and cultural change (Providence, RI) 7-22. De ligt, l. and S. J. Northwood (edd.) 2008. People, Land and Politics. Demographic Developments and the Transformation of Roman Italy 300 BCAD 14 (leiden). De Ruggiero, e. 1896. Le colonie dei romani (Spoleto). Dilke, O. A. W. 1962. “the Roman surveyors,” Greece & Rome 9-2, 170-80. Dilke, O. A. W. 1971. The Roman Land Surveyors: An Introduction to the Agrimensores (Newton Abbott). Dyson, S. l. 1992. Community and society in Roman Italy (Baltimore/london). eckstein, A. M. 2006. Mediterranean anarchy, interstate war, and the rise of Rome (Berkeley). evans, R. 2008. Utopia Antiqua. Readings of the Golden Age and decline at Rome (london/New york). evens J. A. S. (ed.) 1974, Polis and imperium: studies in honour of Edward Togo Salmon (toronto). falbe, C. t. 1833. Recherches sur l’emplacement de Carthage (Paris). fentress, e. 2000. “frank Brown, Cosa, and the idea of a Roman city,” in e. fentress (ed.), Romanization and the city: creation, transformations and failures (Portsmouth, RI) 11-24. finaldi, G. M. 2009. Italian national identity in the scramble for Africa: Italy’s African Wars in the

era of nation-building, 1870-1900 (Bern). fraccaro, P. 1939. “la centuriazione romana dell’agro Pisano” SE 13, 221-29. fraccaro, P. 1940. “Intorno ai confini e alla centuriazione degli agri di Patavium e di Acelum,” in Studi di antichità classica offerti da colleghi e discepoli a Emanuele Ciaceri (Rome) 100-23. fraccaro, P. 1956. Opuscula I, scritti di carattere generale, studi catoniani, i processi degli Scipioni (Pavia). fraccaro, P. 1957. Opuscula III, scritti di topografia e di epigrafia (Pavia). frank, t. 1959. An economic survey of ancient Rome, vol. 1: Rome and Italy of the Republic (reprint 1933 edn, Paterson, NJ). frederiksen, M. W. 1968. Review of Samnium and the Samnites by e. t. Salmon, JRS 58, 224-29. Gabba, e. 2003. “Aspetti della storiografia di ettore Pais,” RSI 115, 1015-1020. Gabba, e. 1984. “Per un’interpretazione storica della centuriazione romana,” in S. Settis (ed.), Misurare la terra: centuriazione e coloni nel mondo romano (Modena) 20-27. Gargola, J. D. 1995. Land, laws, and gods. Magistrates and ceremony in the regulation of public lands in Republican Rome (Chapel Hill, NC/london). Göhler, J. 1939. Rom und Italien: Die römische Bundesgenossenpolitik von den Anfängen bis zum Bundesgenossenkrieg (Breslau). Harris, W. V. 1979. War and imperialism in Republican Rome, 327-70 BC (Oxford). Hermon, e. 2001. Habiter et partager les terres avant les Gracques (Rome). Hulliung, M. 1976. Montesquieu and the old regime (Berkeley/los Angeles). Hurst, H., and S. Owen (edd.) 2005. Ancient colonizations. Analogy, similarity and difference (london). Huskinson, J. (ed.) 2000. Experiencing Rome: culture, identity and power in the Roman Empire (london). laboulaye, e. (ed.) 1876. Oeuvres complètes de Montesquieu (Paris). lackner, e.-M. 2008. Republikanische Fora (Munich). laffi, u. 2007. Colonie e municipi nello stato romano (Rome).




lipsius, J. 1598. Admiranda, sive, de magnitudine Romana (Antwerp). lipsius, J. 1615. ‘L. Annaei Senecae philosophi opera (Antwerp). Macchiavelli, N. 1531. Discorsi sopra la prima deca di Tito Livio (Rome). Macchiavelli, N. 1532. Il principe (Rome). MacKendrick, P. l. 1952. “Roman Colonization,” Phoenix 6-4, 139-46. MacKendrick, P. l. 1960. The mute stones speak: the story of archaeology in Italy (New york). McCulloch, J. R. 1853. Treatises and essays on subjects connected with economical policy (edinburgh). McDonald, A. H. 1960. “fifty years of republican history,” JRS 50, 135-48. Momigliano, A. 1982. “Niebuhr and the agrarian problems of Rome,” History and Theory 21-4, 3-15. Mertens, J. 1953. “l’urbanizzazione del centro di Alba fucens,” MonAnt 8-5, 169-94. Mommsen, t. 1912. Römische Geschichte, vol. 1: Bis zur Schlacht von Pydna (Berlin). Montesquieu, C. de Secondat, Baron de 1734. Considérations sur les causes de la grandeur des Romains et de leur decadence (Amsterdam). Montesquieu, C. de Secondat, Baron de 1766. The spirit of laws (trans. t. Nugent, london). Montesquieu, C. de Secondat, Baron de 1965. Considerations on the causes of the greatness of the Romans and their decline (trans. D. lowenthal, New york/london). Niebuhr, B. G. 1811-12. Römische Geschichte, 2 vols (Berlin). Niebuhr, B. G. 1873. Römische Geschichte, vol. 2 (new edn of M. Isler, Berlin). Nelson, S. 1998. God and the land. The metaphysics of farming in Hesiod and Vergil (Oxford). Northwood, S. J. 2008. “Asconius’ fifty-three Roman colonies: a regal solution,” CQ 58, 353-56. Oakley, S. 2002. “the Roman conquest of Italy,” in J. Rich and G. Shipley (edd.), War and society in the Roman world (london/New york) 9-38. Pais, e. 1920. Storia critica di Roma durante i primi cinque secoli (Rome). Pais, e. 1923. Storia della colonizzazione di Roma antica (Rome). Pais, e. 1924. “Serie cronologica delle colonie



romane e latine dalla età regia fino all’impero. Parte I,” MAL 17 (s. 5), 311-55. Pais, e. 1925. “Serie cronologica delle colonie romane e latine dalla età regia fino all’impero. Parte II, Dall’età dei Gracchi a quella di Augusto,” MAL 1 (s. 6), 345-412. Pais, e. 1928. Storia di Roma: Dall’invasione dei galli all’intervento dei romani nella Campania, vol. 4 (turin). Pais, e. 1931. Storia interna di Roma e governo d’Italia e delle provincie. Dalle guerre puniche alla rivoluzione graccana (turin). Patterson, J. R. 2006. “Colonization and historiography: the Roman Republic,” in G. J. Bradley and J.-P. Wilson (edd.), Greek and Roman Colonisation: Origins, Ideologies and Interactions (Swansea) 189-219. Pelgrom, J. 2008. “Settlement Organization and land Distribution in latin Colonies before the Second Punic War,” in l. de ligt and S. J. Northwood (edd.), People, Land and Politics. Demographic Developments and the Transformation of Roman Italy 300 BC-AD 14 (leiden) 333-72. Polverini, l. (ed.) 2002. Aspetti della storiografia di Ettore Pais (Naples). Purcell, N. 1990. “the creation of provincial landscape: the Roman impact on Cisalpine Gaul,” in t. e. C. Blagg and M. Millett (edd.), The Early Roman Empire in the West (Oxford) 6-29. Quilici, l. 1994. “Centuriazione e paesaggio agrario nell’Italia centrale,” in J. Carlsen, P. Ørsted and J. e. Skydsgaard (edd.), Land use in the Roman Empire (Rome) 127-33. Rhenanus, B. 1520. P. Vellei Paterculi Historiae Romanae (Basel). Reid, J. S. 1913. The municipalities of the Roman empire (Cambridge). Rich, J. 1993. “fear, greed and glory: the causes of Roman war-making in the Middle Republic,” in J. Rich and G. Shipley (edd.), War and society in the Roman world (london) 38-68. Rich, J. W. 2008. “Lex Licinia, Lex Sempronia: B. G. Niebuhr and the limitation of landholding in the Roman Republic,” in l. de ligt and S. J. Northwood (edd.), People, Land and Politics. Demographic Developments and the Transformation of Roman Italy 300 BC-AD 14 (leiden) 519-72.


Ridley, R. t. 2000. “Leges agrariae: myth ancient and modern,” CPh 95, 459-67. Roselaar, S. 2010. Public land in the Roman Republic. A social and economic history of ager publicus in Italy, 396-89 BC (Oxford). Ruperti, f. 1838. De coloniis Romanorum tempore liberae reipublicae deductis commentatio quam themate proposito elucubravit (Rome). Salmon, e. t. 1929. “the Pax Caudina” JRS 19, 12-18. Salmon, e. t. 1933. “the last latin colony,” CQ 27, 30-35. Salmon, e. t. 1936. “Roman Colonisation from the Second Punic War to the Gracchi,” JRS 26, 47-67. Salmon, e. t. 1937. “the latin colonies at Vitellia and Circeii,” CQ 31, no. 2, 111-113. Salmon, e. t. 1953a. “Rome and the latins: I,” Phoenix 7-3, 93-104. Salmon, e. t. 1953b. “Rome and the latins: II,” Phoenix 7-4, 123-35. Salmon, e. t. 1955. “Roman expansion and Roman colonization in Italy,” Phoenix 9-2, 63-75. Salmon, e. t. 1956. “the resumption of hostilities after the Caudine forks,” TAPhS 87, 98-108. Salmon, e. t. 1960. “the Strategy of the Second Punic War,” Greece & Rome 7-2, 131-42. Salmon, e. t. 1962. “the cause of the Social War,” Phoenix 16-2, 107-19. Salmon, e. t. 1967. Samnium and the Samnites (Cambridge). Salmon, e. t. 1969. Roman colonization under the Republic (london). Salmon, e. t. 1982. The making of Roman Italy (london). Settis, S. (ed.) 1984. Misurare la terra: centuriazione e coloni nel mondo romano (Modena). Sewell, J. 2010. The formation of Roman urbanism, 338-200 B.C.: between contemporary foreign influence and Roman tradition (Portsmouth, RI). Sigonius, C. 1560. De antiquo jure Italiae (Venice). Schubert, C. 1996. Land und Raum in der römischen Republik. Die Kunst des Teilens (Darmstadt). Smith, W. 1842. Dictionary of Greek and Roman Antiquities (london). Smith, C. 2006. The Roman clan: the gens from ancient ideology to modern anthropology (Cambridge). Sommella, P. 1988. Italia antica. L’urbanistica romana (Rome).

Stephenson, A. 1891. Public lands and agrarian laws of the Roman republic (Baltimore, MD). Stek, t. D. 2009. Cult places and cultural change in Republican Italy. A contextual approach to religious aspects of rural society after the Roman conquest (Amsterdam). Stek, t. D. 2013. “Material culture, Italic identities and the romanization of Italy,” in J. De Rose evans (ed.), A companion to the archaeology of the Roman Republic (Malden, MA/Oxford) 337-53. Stek, t. D. 2014. “Roman imperialism, globalization and Romanization in early Roman Italy. Research questions in archaeology and ancient history,” Archaeological Dialogues 21-1, 30-40. terrenato, N. 2005. “the deceptive archetype. Roman colonialism in Italy and postcolonial thought,” in H. Hurst and S. Owen (edd.), Ancient colonizations. Analogy, similarity and difference (london) 59-72. tibiletti, G. 1950. “Ricerche di storia agraria romana,” Athenaeum 28, 183-266. torelli, M. 1999. Tota Italia: essays in the cultural formation of Roman Italy (Oxford). tozzi, P. 1984. “la riscoperta dal passato nell’Ottocento. Ricerche sulle divisioni agrarie romane dell’Italia Settentrionale,” in S. Settis (ed.), Misurare la terra: centuriazione e coloni nel mondo romano (Modena) 33-39. Van Dommelen, P. 1998. On colonial grounds. A comparative study of colonialism and rural settlement in the 1st millennium B.C. West Central Sardinia (leiden). Vedriani, l. 1665. Dottori modonesi di teologia, filosofia, legge canonica, e civile (Modena). Wallace-Hadrill, A. 2008. Rome’s cultural revolution (Cambridge). Wilson, A. t. N. 1971. “Review of Roman Colonization under the Republic by e. t. Salmon,” Gnomon 43-46, 583-89. Ward-Perkins, J. B. 1974. Cities of ancient Greece and Italy: planning in classical antiquity (New york). Webster, J. and N. Cooper 1996. Roman imperialism: post-colonial perspectives (leicester).




PART I CONTEXTUALIZING ROMAN REPUBLICAN COLONIZATION. BACKGROUNDS, DEFINITIONS AND COMPARANDA


Private Vis, Public Virtus. Family agendas during the early Roman expansion Nicola terrenato

Introduction A renewed and more sophisticated interest in the colonization process in peninsular Italy is a welcome recent trend in Roman Republican studies. Many of the contributors to this volume are responsible for important advances in this area, often with the result of turning accepted ideas, such as the military character of the colonies or their architectural or ritual imitation of Rome, on their heads (e.g. in the contributions of Jamie Sewell, Guy Bradley, Mario torelli). In the context of this debate, it might be useful to zoom out from the specifics of colonial foundations to look at the broader historical picture offered by the first couple of centuries of Roman expansion, to see whether the revisionist approach to colonies which is so well represented here finds echoes in new ideas about the imperialism of Rome in general. More specifically, now that colonies, especially latin ones, are seen more as hybrid spaces where Roman and non-Romans interacted in complex ways, it is appropriate to ask whether similar phenomena that blurred the basic ethnic boundaries were also present in the expansion process as a whole. this is of course not the place to review the debate on Roman imperialism in general. It can be noted, however, that the classic interpretation and explanations1 have come under a fair amount of deconstruction and criticism from a wide variety of different quarters (cf. Bradley in this volume).2 While it is impossible to generalize, it is clear that there are a number of scholars today who are attempting to redefine the causative chain that leads to the emergence of the Roman empire. Many, for instance, do keep the focus on Roman agency but attempt to explain it in new ways. At a recent Roman Archaeology Conference, for instance, Post-traumatic Stress Disorder, Roman rage and selective female infant exposure were all seriously considered as possible contributing factors to the expansionist behavior in three different papers by leading Roman historians. While perhaps surprising in their choices, these contributions clearly show that Roman imperialism is no longer simply seen as an obvious response to the opportunities, challenges or threats Rome faced. Instead, an underlying prime mover needs to be postulated to explain an otherwise perplexing behaviour. A similar tack is taken in those studies which emphasize the political instability around Rome,3 except that in them the hidden cause is sought, perhaps more convincingly, not in the internal workings of the expanding state, but rather on the wider Mediterranean scene. In all this, what has not yet changed much is our view of how the actual process took place. Scholars have tended to focus on the motivations of Roman aggression or on the conditions that made it successful. But it has so far remained a given that the Romans went out from their city and forced everybody else into submission in one way or another. In short, without coercion and threat there would have been no Roman empire. this means that the non-Romans involved in the process of empire-building are assumed to have had a really limited set of options in the face of Rome’s onslaught. they could give in early in the game or fight on to the bitter end, but their decisions would not have a significant effect on the final outcome of the process. even post-colonial thought, which has in many ways revolutionized our

1 2

Well summarized for instance in Woolf 1993. e.g. Hingley 2005; Mattingly 1997.

3

Such as eckstein 2006.




understanding of modern empires, has had only a limited impact in this specific historical instance of colonialism. the few scholars who have applied it in recent works4 have not got much further than highlighting the possibility that the conquered communities might have deployed forms of overt or covert resistance against Rome.5 this has, perhaps importantly, cast Roman expansion in a negative ethical light, downplaying the ‘civilizing mission’ and presenting the process in stark terms of exploitation and cultural genocide.6 In any case, the actual evidence for Italian resistance to Rome is on the whole very limited, so these attempts have definitely not resulted in a redefinition of the early stages of the Roman conquest. It is only very recently that greater attention has been paid to the impact of decision-making by nonRomans.7 these works suggest that the Roman empire would have been very different (or perhaps would have failed altogether), if the peoples who were absorbed into it had made radically different choices than they did. If this is indeed true, then it follows that any explanation of Roman expansion must also account for the agency of the non-Romans involved in the process. there should be, in other words, more to the phenomenon than just Roman military and political interventionism, whatever its cause might have been. this paper aims at pushing the deconstruction even a little further, attacking the traditional view of the very process of military expansion. the provocative argument advanced here is that, at least in some cases, the annual consular campaigns of the Mid-Republic were geared to serve interests which were other than those of the political abstraction called Rome. It is argued that private groups could temporarily hijack the imperial machinery and use it to further their own goals, rather than those of the empire as a whole (cf. on such a model for early colonization also Attema et al. in this volume).8 the existence of factions and even actual political parties has, of course, been contemplated in a number of discussions of Roman politics,9 so the concept of conflicting agendas within the empire is certainly not new. It has generally been agreed, however, that these groups might have temporarily influenced the foreign policies of Rome in one direction or the other, but that they did not substantially alter the nature of the expansion. Moreover, the factions have typically been conceptualized as completely internal to the political scene in Rome, with little role played in them by other Italians. Is it possible instead to question even what has always been taken for granted: namely, that the conquest was indeed Roman, by suggesting instead that inter-ethnic factions could, at least occasionally, rig the process and divert its rewards away from the actual sovereignty of Rome as a state? It goes without saying that such an approach is informed by the body of recent political thinking which looks at the agency of individuals and small groups within broader polities.10 this is a bottom-up perspective which has been successfully adopted to look at other early states and empires,11 but only rarely to the Romans, perhaps as a result of an entrenched perception of them as a near-perfect instance of devotion to common goods and public virtues. Interestingly, comparable ideas are being tentatively applied to the period of state formation in Rome12 and elsewhere.13 these studies argue that, from the beginning, elite groups did not entirely relinquish the power and prerogatives that they had held before the city was born. they always kept open the option of operating outside state rules and ideology, for instance, waging private wars or suddenly switching their civic loyalty.14 long considered a dying vestige

4 5 6

7 8



e.g. Van Dommelen 1998; Webster and Cooper 1996. terrenato 2005. Bénabou 1976, but the notion of Rome as a Räuberstaat had already appeared in Schwegler 1853, which in turn was based on Salvian. Keay and terrenato 2001; terrenato 1998. Here and later the word empire is used to refer to the dominion of Republican Rome and not to its political form.

9

10 11 12 13 14

Already in Münzer 1920, although the idea has been strongly denied by others, e.g. Develin 1985. e.g. Blanton and fargher 2008. e.g. yoffee 2005. Motta and terrenato 2006; terrenato 2011. for instance emberling 2012; Small 2009. Rawlings 1999.


fIG. 1A. Inscription on the cista ficoroni (courtesy of the Soprintendenza Speciale per il Patrimonio Storico, Artistico ed etnoantropologico e per il Polo Museale della Città di Roma).

fIG. 1B. Drawing of the cista ficoroni (after ficoroni 1745, 73).

of a vanished era, these phenomena might instead have been an integral and functional part of state dialectics in central Italy. In this case, the sorts of behavior reconstructed below could perhaps be seen as an adaptation of deeply rooted archaic mentalities to the brave new world of the Mid-Republic. Necessarily, in what follows, elites, both Roman and non-Roman, will receive the lion’s share in the narrative. this emphatically does not mean a return to the 19th-century idealist focus on great men and great deeds. It is primarily a question of historical visibility and of who was leading the groups which can be seen operating on the political scene. It should be noted in passing, however, that in those vertically integrated social formations the actions at the top were more tightly connected with the consensus of the whole group than in horizontally layered citizenships. furthermore, what is presented here is only a small part of a much larger and more complex argument about Mid-Republican imperialism, so necessarily the broader context of a comprehensive reconstruction is missing. It must be made clear that no one is arguing that the processes illustrated here are representative of the whole early expansion, and even less of the entire span of the Roman conquest. they are a contribution to an embryology of imperialism which can perhaps provide deeper insights into the true nature of the phenomenon than the senile phase on which many interpreters, especially in the english-speaking literature, have tended to concentrate on as a paradigm of the expansion process. there could have been, and there was for quite a while, a Roman empire without the conquest of Dacia or Britain, but there could be no empire before having successfully dealt with the main city-states of western Central Italy.

The Plautii in the Fourth century B.C. to illustrate the ideas advanced above, the specific example of an elite Mid-Republican clan is taken into consideration. the Plautii provide an extremely interesting instance, in which it is virtually unavoidable to attribute ulterior motives to them in their actions on behalf of the Roman state: namely, to see private agendas lurking under their official senatorial writ. they are, however, not at all noted for this reason in any of the surviving literary accounts.15 Indeed, their actions are considered highly

15

As is the case of other contemporary figures, like Manlius Capitolinus to name just one.




fIG. 2. Inscription from the tomb of the Reliefs (after Cristofani 1966, 227).

unremarkable and in line with the prevailing ethos. While this does of course not imply that what they did was entirely representative of an entire social group in Rome, it should at the very least alert us to the possibility that behaviors of this kind could have shaped the early part of the conquest. there is also a kind of poetic justice in choosing the Plautii to investigate the workings of Roman imperialism, when one considers that Aulus Plautius figured so prominently in the Claudian invasion of Britain, an event which in turn has played such a big role in the modern historiography and in the popular perception of the Roman conquest. the Plautii suddenly leap into our historical field of vision early in the 4th c. B.C. they were certainly an extended family with connections to the main latin centres of tibur and Praeneste, as well as the neighboring, less well-known center of trebula Suffenas,16 variously ascribed to the Sabine or Aequian ethnic groups. this is largely based on Mid- to late Republican inscriptions and dedications,17 which nevertheless do seem to present a fairly consistent picture. One contemporary piece of evidence which is clearly important, if difficult to interpret, is provided by the 4th c. B.C. ficoroni Cista, which notoriously declares in its inscription that Novios Plautios med Romai fecid (Novius Plautius made me in Rome) (figs. 1a and b). the prevailing opinion of scholarship is that Novius Plautius was the craftsman who made the object, but his status is less clear. Because Novius is a Campanian name (and not a known praenomen among the Plautii), theodor Mommsen first advanced the hypothesis that Novius might have been a slave or a freedman of the Plautii,18 although he later changed his mind. A different, and perhaps less likely, interpretation of the inscription is that a Novios Plautios commissioned the piece.19 the connection with Rome (where there were certainly Plautii by the 340s B.C., around the time when to which the cista is dated) is intriguing but far from clear, since the object was found in Praeneste and cista manufacture is otherwise unattested in Rome.20 It seems safe, in any case, to derive proof of the connection of the Plautii with other Praenestine aristocrats from this piece of evidence. to the early 4th c. B.C. dates another important written testimony, the name plavti carved in one of the niches of the splendid tomb of the Reliefs at Caere (fig. 2). this is generally accepted to be a female Plautia who had married into the prominent Caeretan clan of the Matuna, to whom the tomb belonged.21 A match of this kind confirms that the Plautii moved in the most exalted central Italian circles and were involved in aristocratic alliances which effortlessly crossed political and ethnic lines. Such wide-ranging

16 17 18 19



Ross taylor 1956. esp. CIL xIV 3212-15. Mommsen 1845, 72. Massa-Pairault 1992.

20 21

Wallace 1990, 283-84. Blanck and Proietti 1986.


horizontal elite mobility had characterized Italy (and more widely the central Mediterranean) at least since the Archaic period,22 although its political significance has still not been fully recognized, perhaps because it goes against the grain of the ethnocentric framework which lies at the core of most narratives about the conquest. It is essential to remember that, in sharp contrast to what happened during the european conquest of the Americas or of Africa, when Rome began its expansion its leaders had known, interacted and intermarried with their peers in the conquered communities for centuries and this cannot have failed to influence the process in a number of ways. Whatever the case, sometime before 358 B.C. at least a branch of the Plautii must have relocated to Rome and acquired full Roman citizenship, since this is when their first magistrature is recorded. In this year, C. Plautius Proculus was consul, not having held any prior recorded office, together with C. fabius Ambustus, while C. Sulpicius Peticus was made dictator to deal with a Gaulish threat. the consular armies were sent to avenge raids by the tarquinians and by the Privernates. Although livy says that lots were drawn (evenere; liv. 8.12.6), fabius ‘happened’ to receive the etruscan provincia23 whilst the latin Plautius received the Hernican mandate.24 While fabius suffered one of several fabian defeats in etruria (and an egregious one, resulting in hundreds of Roman prisoners being sacrificed), Plautius obtained a victory over an ethnic group which was located close to the area of provenance of his gens. Privernum also suffered a defeat in the following year and would appear to have entered in an alliance with Rome at this time. It is, of course, remarkable for a member of a group which had only recently entered the political fray in Rome to have risen all the way to the top of his adopted state, but it was not at all a unique occurrence. Many other prominent families from Central Italy attained memberships of the senatorial class between the 4th and the 3rd c. B.C. they represent a substantial part of the growing plebeian elite in Rome, to the point that it is debatable whether the evident broadening of the power structure which happened after the licinian-Sextian laws (367 B.C.) was more the result of vertical mobility inside Rome or of horizontal mobility in Central Italy. Once again, the latter kind of mobility, both as individuals and as whole family groups, was a long-term characteristic in this region. the narrative about Claudii from Sabina being admitted not only in the Roman nobility, but even in the patriciate places the event only a little more than a century earlier than the similar promotion of the Plautii. As it happens, the primary allegiance of the Plautii seems to have been precisely with the Claudii,25 who, like the Plautii, favored expansion to the south of Rome.26 In 356 B.C., the same C. Plautius was appointed magister equitum to the first plebeian dictator, C. Marcius Rutulus. this was a politically charged event, vehemently opposed by the patriciate, and signals clearly that the Plautii were at the forefront of the plebeian push for more power. Rutulus himself had been consul the previous year, during elections held by Plautius, so it is not hard to see a political partnership between the two.27 the next member of the family to be consul was C. Plautius Venox in 347 B.C., who was involved in cutting interest rates, another key plebeian priority. It has in fact been hypothesized28 that he had previously been involved in the institution of the quinqueviri mensarii in 352 B.C., based on the fact that most of the commissioners were known friends of his.29 they acted as sort of public mediators and bankers who were trying to compose debt issues.

22 23

24

Ampolo 1976-77. As befitted a member of a family which had a lifelong connection with etruria; Ogilvie 1976, 115. Specifically to tangle with the Privernates, it would seem from liv. 8.15.11; it should, however, be remembered that Privernum was Volscian according to Vergil (Verg., Aen. 7.685).

25 26 27

28 29

Syme 1939, 422. Massa-Pairault 2001. thus RE 21 (1951) 2-25, s.v. “C. Plautius Proculus” (M. Hofmann). Ibid. Münzer 1920, 38.




fIG. 3. Map with Rome, tibur, trebula, Praeneste, Privernum, fregellae (B. Brouwenstijn).

C. Plautius Venox was consul again, together with l. Aemilius Mamercinus, in 341.30 In that year, news reached Rome that Privernum had broken the alliance and attacked the neighboring colonies of Norba and Setia. Again, a blind sors (liv. 8.1.3) assigned the war to Plautius, as well as a mandate to deal with a rampaging Volscian army. He made a beeline for Privernum, took it, gave it back to its inhabitants after having installed a garrison and confiscated two-thirds of the territory. About a decade later, the Plautii pulled off a rare hat-trick, providing three consuls in a row in 330-28 B.C., a feat which was not easy even for the most distinguished and well-connected of the autochthonous patrician families.31 Once more, these years were clearly characterized by Roman activism in the area of Privernum (fig. 3). In 330 B.C., the Privernates, joined by the inhabitants of nearby fundi, were on the warpath, again laying waste to Norba and Setia, this time also attacking Cora. Both consuls, l. Plautius Venox and l. Papirius Crassus, were sent to sort things out and apparently caught the rebellious army in a pincer maneuver, forcing it to retreat into Privernum, but could not take the city. elections were hastily conducted and one of the new consuls, C. Plautius, was sent to take over operations with the army raised by his clansman, while his colleague, l. Aemilius Mamercinus, scraped together a ragtag army to repel a Gaulish raid. However, when the latter failed to materialize, he too descended upon Privernum. the city was finally taken, either by storm or by betrayal, without a fight.32

30 31

ďœľďœ°

At least according to Broughton 1952, 602. MĂźnzer 1920, 39-47.

32

Oakley 1997.


the evidence reviewed above seems to show that between 358 and 329 B.C., in four of the five years in which there was a Plautian consul, Rome was tangling with Privernum and the Senate randomly happened to assign the problem to the latin Plautius, rather than to his ‘old Roman’, patrician colleague. Such an impossible coincidence is hard to explain within the traditional interpretive frameworks for Roman imperialism. either the very election of a Plautius in Rome was enough to make tempers run high in 100km distant Privernum, or there must have been a ‘Plautian’ agenda at work which had to do with that region (and had inflammatory effects there), rather than a generic and randomized defensive mandate from the Senate. Obviously, the Plautii had a specific and overriding interest in controlling the city, which significantly lay right outside their area of origin, under the guise of Roman expansion. As suggested by later events, their aim must have been the addition of the city to the power base of their clan, through the links of patronage which always tied surrendering communities to their captors.33 Needless to say, clans in Rome fought the political game primarily by maintaining complex networks of long-term ties with their political constituencies in the city and around the empire. It is quite possible that the involvement of the Plautii with Privernum dated to before the events in question, but the new mechanism of Roman expansion gave them (and other groups like them) an unprecedented opportunity to achieve their goals. they must have been able to stir things up in the region to the point of precipitating war and rebellion, perhaps giving us an important clue to the many times in which alliances in Mid-Republican Italy were allegedly reneged by the partners of Rome without apparent reason and with disastrous consequences for the treaty-breakers. It is hard to escape the conclusion that, at least on this occasion, the Plautii were able to harness the power of the Roman alliance to expedite their power play over Privernum. Before exploring this fascinating story any further, however, it is worthwhile to make a short digression to look at the adversary of the Plautii in the events of 330-29 B.C. the rebellion of Privernum was instigated by a Vitruvius Vaccus, who also commanded the army. far from being a freedom fighter steeped in a traditional highland society (as the stock image of the highlanders from Southern latium would lead us to believe), he was a well-connected, rich, urban politician. He was a prominent figure in his native fundi (another nearby city), and very notoriously had a luxurious house on the Palatine hill.34 later, after his disgrace, the house was razed, and the vacant site became known as the Prata Vacci,35 analogously to what had happened to the houses of other famous Republican agitators, like Sp. Cassius, Sp. Melius or Manlius Capitolinus.36 the only reason a latin aristocrat would have had a highly visible house on the Palatine is that he wanted, just as the Plautii, to become a part of the Roman political game. But how could this possibly be reconciled with his becoming an enemy of Rome in the end? Clearly, if the struggles attending the early Roman conquest are understood in terms of clashes between self-exclusive ethnic groups and/or citizen bodies, the behavior of Vitruvius is inexplicable (but so, incidentally, is that of Coriolanus). However, if he indeed had a rival political agenda to that of the Plautii, which the latter proceeded to disrupt as soon as they came to power, on the global Central Italian scene, things suddenly become a little more understandable. What Vitruvius’ plan might have been is very difficult to imagine, given that the reasons of the losing side are hardly ever preserved fairly and adequately. It is on the whole unlikely that he originally intended to break fundi and Privernum from the Roman alliance, after the latin war of 338 B.C. had made it very clear that independence in southern latium was no longer an option. Why would he have invested in an atrium house near the Roman forum if that was the case?37 He must instead have been

33 34

Deniaux, Morstein-Marx and Martz 2006. He is in fact the earliest private resident of the hill known to us, unless one counts Romulus; Carandini and Papi 2005; Royo 1987.

35

36 37

LTUR 2 (1995) 215, s.v. “domus: Vitruvius Vaccus” (e. Papi). Coarelli 2007. Wiseman 1971.




trying to parley his influence in a strategic region of Italy for some kind of political or diplomatic advancement in Rome. It is fairly clear that, as it might have been expected, there was a pro-Vitruvius and an anti-Vitruvius faction in both fundi and Privernum. When Plautius descended on the former, the senate of the fundani was quick to disavow Vitruvius, claiming that his real base of operations was Privernum, where he was holed up at that point. And if Privernum was indeed delivered to Plautius by treason, as one source claims, it would suggest that there was an anti-Vitruvius (and presumably proPlautius) group in that city too.38 In light of this reconstruction, it would not be presumptuous to propose that there was much more to the ‘conquest’ of Privernum than simply the abstractions called Rome and Privernum clashing over submission or independence, as most modern narratives would have it. Beneath the surface, it is not hard to see the struggle between two factional aristocratic networks, with tentacles in various cities around the alliance as well as in Rome. It is highly doubtful that anyone was fighting for the chimeric freedom and independence of Privernum; rather, at issue was the way in which this particular community was to be integrated in the emerging new political entity. Most probably, the real bone of contention was the valuable privilege of being the patron clan of two important communities in the southern expansion of Rome. Both competing factions easily spanned ethnic lines, Roman, latin, Hernican, and neither appears to have been more culturally ‘Roman’ than the other. Had Vitruvius prevailed, it is highly unlikely that the cultural trajectory of Privernum would have been significantly different. the city would have simply belonged to a different faction within the emerging empire and would have had a different power broker in Rome. In short, at least at this particular juncture, the factional boundary between the Plautii and Vitruvius was far more meaningful in shaping the expansion process than the political one between Rome and Privernum, or the ethnic one between Romans and Hernicans. the aftermath of these events is also very instructive. livy (8.20.6-21.10) gives a highly fictionalized account featuring brave speeches about liberty and independence delivered in the Senate by envoys from Privernum. the consuls, touched, interceded for the rebellious, but proud, city, and punishment was not only waived, some citizenship rights (presumably civitas sine suffragio) were even granted. full Roman citizenship followed soon after, in 318 B.C., when the Oufentina tribe was created. It should not come as a surprise, by now, that a Plautius, l. Plautius Venox, was again consul in that year. this is a surprisingly benign treatment for a city which had rebelled twice in twelve years, but only if the event is understood as a desperate attempt to break free from the empire; indeed, the later sources can only explain such mildness by invoking a dubious Roman admiration for indomitability. If the latin Plautii did not like Hernican Privernum, why did they reward it so generously? If, however, we leave aside the idea of Privernum as a defeated rebel, it is not hard to see that the city, once subtracted to the influence of the rival Vitruvius faction, was very desirable for the Plautii and their allies as a client community. Granting Roman citizenship to it made perfect sense if the idea all along had been to increase that group’s power in the alliance, and to add another group of voters to its power base, especially in the yearly elections. the following year featured the third Plautius consulship in a row. Not much happened, with the exception of the foundation of the latin colony of fregellae, not far from Signia and only about 20 km from Privernum. If the finger of the Plautii can be intuited in this pie too, it would have resulted in a greater expansion of their political footprint in southern latium, seeing that colonies, just as cities which had surrendered with their victors, were automatically entered into the patronage sphere of the magistrates that had been in charge of founding the colony. While we cannot be completely sure of this, the suspicion is strengthened when the patterns of tribal gerrymandering are taken into account.

38



for very similar dynamics in Apulian communities: Gallone 2007.


Privernum was attributed to the recently established Oufentina tribe as the fate of fregellae was to be. the question of when colonies of latin right would be attributed is a complex one, and revolves around the issue of whether the magistrates in latin colonies received the full Roman citizenship or not.39 If they did, then it is even possible that the attribution of fregellae to the Oufentina might have dated to its foundation during the consulship of a Plautius. In any case, it is evident that the Plautii were busy stacking the Oufentina in their favor. In 318 B.C., Canusium in Apulia surrendered to yet another Plautius consul and was later placed in the Oufentina too. the political strategy was obvious and well-known throughout the Republic as a way of creating and maintaining political power. Concentrating client groups in new tribes guaranteed the control of that tribal vote, requiring far fewer votes than would have been necessary in one of the old tribes, maximizing the amount of electoral clout that could be obtained through one’s patronage.40 the vote in the electoral college of the tribes which was acquired in this way could be used to elect more family members, who in turn would expand the political reach of the clan to ever wider boundaries. even more often, these votes could represent extremely valuable bargaining chips within the larger factional syndicate which, in the Mid-Republic, revolved around the Claudii.41

The later life of the Plautii After their meteoric rise during the 4th c. B.C., the Plautii coasted along in Rome as a senatorial, but not particularly remarkable, plebeian family. there are no offices recorded for the 3rd c. B.C., but in the 2nd the Plautii Hypsaei, who claimed descent from the original family were consuls, praetors, moneyers and senators. Other offices were held in the 1st c. B.C. A continuing link to Central Italian elites can perhaps been inferred from the lex Plautia Papiria of 89 B.C., which extended citizenship to the communities which had actively rebelled during the Social War. Sponsored by the tribune M. Plautius Silvanus, in conjunction with the far more famous C. Papirius Carbo, it rewarded, as had happened in 329 B.C., Italians who had recently revolted with the concession of important political rights.42 two and a half centuries might seem today like a long time for family roots and old connections to be still relevant, but we actually know for a fact that the late Republican and early Imperial Plautii were still putting their past to good use. A coin type of the 50s B.C. celebrated the triumph over Privernum in 341 B.C. the choice of rehashing such a long-forgotten event by the moneyer, l. Plautius Plancus, clearly indicates that these connections were still vital to a Roman politician of that time. even more importantly, in this sense, is the construction, in the early 1st c. A.D., of a massive mausoleum of the Plautii in a prominent location on the via tiburtina, near tibur (tivoli).43 these mausolea were important forms of family aggrandizement and of self-promotion and the choice to build it in the ancestral homeland of the Plautii makes it extremely probable that, even in the changed political environment of the early empire, they still derived some measure of power and influence from the role they had played and continued to play in southern latium. At the same time, their center of gravity was in Rome and, as many contemporary families from elsewhere in Italy, they must have acted as mediators and brokers of their clients in the capital. In Rome, the Plautii had a fabled villa with horti on the lateran hill,44 from which the family branch of the Plautii laterani derived their cognomen, and which was probably torn down when the family

39 40 41 42

Mouritsen 1998. Ross taylor 1960. Massa-Pairault 2001. Mouritsen 1998; incidentally, but this must be a real co-

43 44

incidence, Plautius’ consular colleague who helped take Privernum in a pincer attack was another Papirius. Ross taylor 1956. Juv. 10.17; Manacorda 2007.




fIG. 4. the Corsini chair (courtesy of the Soprintendenza Speciale per il Patrimonio Storico, Artistico ed etnoantropologico e per il Polo Museale della città di Roma).

became entangled in an alleged anti-Neronian plot. A unique marble piece, the Corsini chair (fig. 4), was found nearby and it has been suggested that it might have been among the debris of the villa.45 If this hypothesis is correct, there would be an interesting connection attested between the Plautii and the etruscan world even at this late stage. the chair appears to be a marble copy of an Archaic etruscan bronze throne. Probably a product of the climate of nostalgia and etruscophilia of the Augustan and Julio-Claudian period, the piece certainly confirms that some aristocrats of this period were interested in the etruscan prestige display predating them by some seven centuries, presumably by means of heirlooms or depictions which were still circulating. In any case, the Plautii had an undeniable contemporary etruscan connection through Plautia urgulanilla, the first wife of the etruscophile emperor, Claudius, who also manifestly descended from the prominent etruscan urgulanii, who included the urgulania who was livia’s best friend. It is nothing less than striking to find the Plautii and the Claudii still partnered up more than four hundred years after

their first Roman adventure together. the partnership, as it is well known, was also active on the military side. there are several Plautii in the Julio-Claudian army, of whom the most famous is A. Plautius, to whom Claudius entrusts (together with another Central Italian, destined to much greater fortune, the young Vespasian) his one expansionistic foray in Britain.46 While, of course, it would not make sense to speak of a Claudian faction at this point, it is hard not to wonder whether the careers of the early Imperial Plautii were not favoured by the distant remnants of a long-standing factional allegiance with a family which had become the ruling one.47 the patronage tie between Plautius and Claudius in this period does not mean that Britain was conquered only for the benefit of one faction in the Roman government, but, besides their intrinsic interest, the old ways might provide some revealing background to the personalism which characterizes Roman imperial politics. the military commands of the Plautii neatly bookend the entire span of Roman imperialism and arguably illuminate an aspect of the process which has received less attention than it deserves and which can stimulate some unconventional reflections on the nature of the Roman empire.

Conclusions this brief review of the actions of one individual clan within the broader process of early Roman expansion is only exceptional because of the richness and variety of the information preserved. It is important to reiterate that none of our sources, which are as a rule interested in stigmatizing illegal or unseemly political actions, finds anything scandalous or even worthy of remark in what the Plautii did

45 46



torelli 1999, 150-64. Webster 1980.

47

As somewhat wittily suggested in Syme 1939; Birley 2005, 19-20.


with the authority and the decision-making power which came with their consulships. It is only the singlemindedness of the Plautii concerning one specific city, Privernum, which makes it visible for us, like dye on a microscope slide, an intricate filigree of interconnections between public politics and private agendas and strategies. Many other threads could be followed, maybe less clearly and over shorter time spans, in the historical tapestry of this period. their importance, overshadowed by later narratives and discourses propounding a monolithic concept of the Roman state, is such that it is not unthinkable to surmise that they played a very considerable role in what we call Roman expansion. It is a fact that clans which were not even strictly Roman in the first place could be allowed to commandeer the Roman imperial machinery for purposes which have every appearance of being their own. this does not, of course, mean that the Plautii or anyone else had tyrannical designs of any sort. It was well understood that, just as they were pursuing their vision in an area of latium, other clans would be attempting to do the same elsewhere at other times, using the same equipment. the competition between peer factions was fierce, but it was essential that none of them could control the system completely, even for a brief period. At the end of the day, they all benefited in some form from the sum total of the expansionist moves which were centered on the imperial capital and on the political entity which they had in common. the full import of processes of this kind remains to be evaluated, but it would appear that, when it came to the urban communities of Central and Southern Italy, it is not always necessarily appropriate to conceptualize the expansion process as a conflict between Rome and other communities. Military and diplomatic initiatives, peace treaties, surrenders, colonial deductions were all conducted by elites which had Roman citizenship but had come from disparate ethnic and cultural provenance, and they cannot all be automatically assumed to have had only the public benefit at heart. they often had evident ties with the communities which they were ‘conquering’48 and sometimes it is so evident that the process is remotecontrolled by a network of syndicated aristocrats in Rome and elsewhere that even the bewildered later historians cannot completely obfuscate it. At least in these cases, we have to contemplate the possibility that the primary aim of the expansionist action was not to assert the hegemony of the abstraction called Rome over other Italians, but instead to advance the agenda of a specific, successful inter-city elite network. Such agendas included dominance in the local community, maintenance of the established social order, control of the political brokerage between the community and the center of power, piloting tribal formation and composition, all resulting in increased network clout in Rome. this last element was, of course, essential in enabling the faction to hijack the imperial military machinery again in the future. this positive feedback could have been a powerful force adding a motivator which was fundamentally different to the traditional explanations of Roman imperialism. the Plautii do not fight to defend Rome, or to assail the Hernicans, or to acquire means of production or loot or simply to celebrate a triumph and boost their prestige in Rome. Something different and potentially more important was at stake. Who would be able to count the city of Privernum, which was arguably falling into the orbit of Rome anyway, among their clients and take out their enemies there? Who could found a colony at fregellae and gain a permanent group of citizen supporters there? the answers to these questions would help determine the political balance in Rome for generations to come. establishing patronage, stacking tribes and centuriae, eliminating local opposition had effects which would be felt centuries after the fact, and it is over these issues the struggle rages hardest. It is hard to escape the impression that, at least in some cases, the way in which the expansion happened was more important than whether or not the expansion itself took place. A symmetrical perspective can also be adopted in looking at the behavior of the elites in the

48

Illuminating in this sense are the cases of the fulvii and tusculum; eckstein 2006, 255-56.




incorporated communities. Just as Roman expansion can be described as a patchwork of factional projects, opposition to it should not necessarily be construed as unified ethnic or political groups desperately struggling for independence. While, as a rule, defeated ideas do not preserve well in the historical record, enough survives to gain glimpses of them, and rarely do they point in the direction of city-state independence. far more often, these alternative designs involved different versions or configurations of the same basic idea of a territorial empire, only with different people at the top. the Roman conquest was, in part, a violent and traumatic transformation, but one has to wonder how much of the conflict was not between those who wanted expansion and those who resisted it, but rather between different factions vying for the control of a newborn entity whose emergence was unavoidable in the changed world of the late Hellenistic central Mediterranean. Breaking down Roman expansion in Central Italy into its constituent elements makes room for a much wider range of responses on the part of non-Romans. If, instead of a single-minded, implacable onslaught, we envisage a series of multi-lateral interactions, resulting in ad hoc solutions, it is easier to accept that there were ways in which other Italians could have a say in how the process unfolded. Rather than being limited to the classic polar antithesis between acceptance and resistance (as it is often assumed in some provincial contexts), participants in the integration could deploy a wide variety of political behaviour which would have a measurable effect on its eventual outcomes. With the Plautii, we have seen one possible option: an early move to Rome, usually facilitated by existing ties with the local nobility. Building on the tradition of Archaic horizontal mobility, non-Roman clans in the Mid-Republic could resettle in Rome for good, with the transparent intent of claiming a larger share of the benefits being generated by the expansionist project centered there. the geographic distribution of their provenances is indicative of the scope of the vast elite networks which were coming together at Rome. latium is, of course, very well represented, as are etruria and Campania. the thickly urbanized western part of Central Italy was clearly an essential part of the new alliance. But a smaller number of families hailed from as far as Sabina, Samnium, umbria, and, slightly later, Picenum, Daunia and other places. there is practically no area of peninsular Italy which is not represented in this wave of early adopters. In Rome, they created complex and shifting clusters with the local nobility, who would much rather deal with them than with the lower classes of Rome itself. the convergence of Central Italian elites on Rome undoubtedly was a key requisite for an expansion centered there to be successful. But it does not automatically entail the marginalization of other civic communities. the centripetal elites were not immigrants in the modern sense of the word. they did not seek social promotion and, in fact, often had to accept a rank which was not as prominent as the one which they had held at home.49 they also kept a tight control over what happened in the ‘old country’ which was usually a central part of their political constituency. In fact, as the case of the Plautii illustrates, they often used the newly acquired central power to settle local scores, among other goals. Branches of these clans, perhaps cadet ones, remained locally based and ensured the functionality of the new power connection. Indeed, the incorporation of these communities into the alliance was often effected precisely thanks to the mediation of these groups which had branches in Rome and at home. Rather than a ‘betrayal’ of their cities of origin, the actions of these clans should be seen as an interested brokerage between the new capital and the communities which were being absorbed into the alliance. Making room for a variety of agency options open to Central Italian elites in the time of the conquest can help in making better sense of some apparently peculiar aspects of the conquest. therefore, if we provocatively describe Mid-Republican expansion as the result of a patchwork of private factional actions on the part of a few original Roman clans and scores of non-Roman ones, can we still speak of a Roman conquest? Or could we just as well speak of Rome as a vehicle for political clustering which was constantly

49



Cf. the royal Marcii or Cilnii.


hijacked and appropriated by an ever-growing number of trans-ethnic factions? Was Rome a city-state which spread its power in all directions, or was Rome simply a tool used by everybody else to effect their purposes, and which alone completely lost its identity in the process? there is hardly any need to stress that any model of Roman expansion can at best account for only a part of the actual phenomenon. And yet, if there is any value in the reconstruction presented so far, there might well be an important component which has not received the attention it deserves. In 1714, Bernard de Mandeville republished his allegorical poem about a beehive with the subtitle Private Vices, Public Benefits. A source of great scandal at the time, his proto-utilitarianist views revolved around the idea, later reworked by Adam Smith, that even the basest and most self-serving behaviors can benefit society as a whole.50 the expression has become commonplace and has been used to deride hypocrisy (as in the 1976 film Vizi privati, pubbliche virtĂš, to which the title of this paper alludes) and advocate pragmatism. In our case, it provides a sobering reminder that when historians underestimate the global effect of private agendas and motivations, they run the risk of creating an unrealistic narrative populated by animated political abstractions instead of real people. It is quite possible that the aggregate effect of a multitude of self-serving behaviors on the part of a wide array of Central Italian clans was the creation of a unified territorial empire which was Roman primarily in the choice of its capital. the emergence of this new geopolitical structure, if it did not exactly benefit the public in our sense of the word, certainly favored the landed elite class as a whole, in spite of the acrimonious factionalism which had always affected it, catapulting it, probably beyond its wildest imagination, to global domination.

50

Goldsmith 1985.

ďœľďœˇ


References Ampolo, C. 1976-77. “Demarato: osservazioni sulla mobilità sociale arcaica,” DialArch 9-10, 33345. Bénabou, M. 1976. La résistance africaine à la romanisation (Paris). Birley, A. R. 2005. The Roman government of Britain (Oxford/New york). Blanck, H. and G. Proietti 1986. La Tomba dei Rilievi di Cerveteri (Rome). Blanton, R. e. and l. fargher 2008. Collective action in the formation of Pre-Modern states (New york). Broughton, t. R. S. 1952. The magistrates of the Roman Republic, vol. 2: 99 B.C.-31 B.C. (New york). Carandini, A. and e. Papi 2005. “Palatium e Sacra Via, vol. 2: l’età tardo-repubblicana e la prima età imperiale (fine III secolo a.C. - 64 d.C.),” BA 59-60, 3-327. Coarelli, f. 2007. “Aree aperte e concezione dello spazio a Roma,” Fragmenta 1, 25-32. Cristofani, M. 1966. “le iscrizioni della tomba dei Rilievi di Cerveteri,” StEtr 34, 221-38. Deniaux, e., R. Morstein-Marx and R. Martz 2006. “Patronage,” in N. Rosenstein and R. MorsteinMarx (edd.), A companion to the Roman Republic (Malden) 401-20. Develin, R. 1985. The practice of politics at Rome, 366-167 B.C. (Brussels). eckstein, A. M. 2006. Mediterranean anarchy, interstate war, and the rise of Rome (Berkeley). emberling, G. 2012. Urban process and Mesopotamian cities 3500-1600 BC (Cambridge). ficoroni, f. 1745. Le memorie ritrovate nel territorio della prima, e seconda citta di Labico e i loro giusti siti (Roma). Gallone, A. 2007. “Agricultural landscapes in Hellenistic Messapia: socio-economic structures and cultural implications,” in P. A. R. Van Dommelen and N. terrenato (edd.), Articulating local cultures: power and identity under the expanding Roman Republic (Portsmouth) 23-31. Goldsmith, M. M. 1985. Private vices, public



benefits: Bernard Mandeville’s social and political thought. Ideas in context (Cambridge/New york). Hingley, R. 2005. Globalizing Roman culture: unity, diversity and Empire (london). Keay, S. J. and N. terrenato (edd.) 2001. Italy and the West: comparative issues in Romanization (Oxford). Manacorda, D. 2007. “Il laterano e la produzione ceramica a Roma: aspetti del paesaggio urbano,” in A. leone et al. (edd.), Res Bene Gestae: ricerche di storia urbana su Roma antica in onore di Eva Margareta Steinby (Rome) 195-204. Massa-Pairault, f.-H. 1992. “Aspetti e problemi della società prenestina tra IV-III sec. a.C.,” in P. Baglione (ed.), La necropoli di Praeneste: periodi orientalizzante e medio repubblicano (Palestrina) 109-45. Massa-Pairault, f.-H. 2001. “Relations d’Appius Claudius Caecus avec l’Étrurie et la Campanie,” in D. Briquel and J.-P. thuillier (edd.), Le censeur et les Samnites: sur Tite-Live, livre IX (Paris) 97-116. Mattingly, D. J. (ed.) 1997. Dialogues in Roman Imperialism (Portsmouth, RI). Mommsen, t. 1845. Oskische Studien (Berlin). Motta, l. and N. terrenato 2006. “the origins of the state par excellence: power and society in Iron Age Rome,” in C. C. Haselgrove and V. Guichard (edd.), Celtes et Gaulois, l’archéologie face à l’histoire, vol. 4: Les mutations de la fin de l’âge du Fer (Glux-enGlenne) 225-34. Mouritsen, H. 1998. Italian unification (london). Münzer, f. 1920. Römische Adelsparteien und Adelsfamilien (Stuttgart). Oakley, S. P. 1997. A commentary on Livy: books VI-X, vol. 1: Introduction and book VI (Oxford). Ogilvie, R. M. 1976. Early Rome and the Etruscans (Atlantic Highlands). Rawlings, l. 1999. “Condottieri and clansmen: early Italian raiding, warfare and the state,” in R. Alston and K. Hopwood (edd.), Organised crime in antiquity (london) 97-127.


Ross taylor, l. 1956. “trebula Suffenas and the Plautii Silvani,” MAAR 24. Ross taylor, l. 1960. The voting districts of the Roman Republic: the thirty-five urban and rural tribes (Rome). Royo, M. 1987. “le quartier républicain du Palatin: nouvelles hypothèses de localisation,” RÉL 65, 89-114. Schwegler, f. C. A. 1853. Römische Geschichte, vol. 1, parts 1-2: Romische Geschichte im Zeitalter der Konige (tübingen). Small, D. 2009. “the dual-processual model in ancient Greece: applying a postneoevolutionary model to a data-rich environment,” JAnthArch 28, 205-21. Syme, R. 1939. The Roman revolution (Oxford). terrenato, N. 1998. “tam firmum municipium: the Romanization of Volaterrae and its cultural implications,” JRS 88, 94-114. terrenato, N. 2005. “the deceptive archetype: Roman colonialism in Italy and postcolonial thought,” in H. Hurst and S. Owen (edd.), Ancient colonizations: analogy, similarity and difference (london) 59-72. terrenato, N. 2011. “the Versatile Clans: the nature of power in early Rome,” in N. terrenato and D. C. Haggis (edd.), State Formation in Italy

and Greece (Oxford) 231-44. torelli, M. 1999. Tota Italia: essays in the cultural formation of Roman Italy (Oxford/New york). Van Dommelen, P. A. R. 1998. On colonial grounds: a comparative study of colonialism and rural settlement in first millennium B.C. West Central Sardinia (leiden). Wallace, R. 1990. “Hellenization and Roman society in the late fourth century B.C.: a methodological critique,” in W. eder and C. Ampolo (edd.), Staat und Staatlichkeit in der frühen römischen Republik (Stuttgart) 278-92. Webster, G. 1980. The Roman invasion of Britain (totowa). Webster, J. and N. J. Cooper 1996. Roman imperialism: post-colonial perspectives (leicester). Wiseman, t. P. 1971. New men in the Roman senate, 139 B.C. - A.D. 14 (london). Woolf, G. 1993. “european social development and Roman imperialism,” in P. Brun, S. e. van der ernst and C. R. Whittaker (edd.), Frontières d’empire: nature et signification des frontieres romaines (Nemours) 13-20. yoffee, N. 2005. Myths of the archaic state: evolution of the earliest cities, states and civilizations (Cambridge).




fIG. 1. the ager Cosanus from the air. the city walls of the colony are clearly visible. Marked with arrows are the traces of centuriation Castagnoli identified. One can also see the via Aurelia. the photo was made in 1954 (after Castagnoli 1956a, tav. 24).


The nature of Roman strategy in Mid-Republican colonization and road building Guy Bradley

Introduction Scholars often observe that a ‘hegemonic landscape’ was characteristic of Italy by the time of the Social War.1 Roman power was manifested through the great number of colonies, roads and centuriated grids which had spread across the peninsula from the 4th to the 2nd c. B.C. the huge amount of Roman effort and resources expended on reordering the landscape has attracted the attention of many scholars. It has recently been estimated by tim Cornell, for instance, that over 70,000 adult males were settled in colonization schemes in the 4th and 3rd c. B.C.2 this concept of a hegemonic landscape raises some important questions. How was it created, and who was responsible? to what extent was it a deliberate creation, the result of Roman planning? What does its creation tell us about Roman imperialism? exploring the complexity of this issue is a useful way of demonstrating the limits of the concept of a hegemonic landscape and ultimately of improving our understanding of Roman imperialism (fig. 1).

Ancient and modern approaches to colonization fundamental to most modern accounts of Roman colonization is the idea that colonial foundations played a key military role in Roman history. Certainly, many sources do attest to a military purpose for colonies, particularly those writers looking back to traditional practice well before their own day. In two speeches Cicero famously describes colonies as propugnacula imperii, “bulwarks of empire”.3 for the agronomist Siculus flaccus, colonies were to “repel the attacks of enemies”.4 In the introduction to the Civil Wars, Appian describes how colonies “were the alternative the Romans devised to garrisons” (although his emphasis on the agricultural use of the confiscated land is also notable).5 there are plenty of specific examples of this type of reasoning, such as Asconius’ description of Placentia being set up against the Gauls,6 and livy’s of Narnia as established “against the umbrians”.7 this is the explanation emphasized by many scholars, who often seem uncomfortable with socio-economic motivations. take edward togo Salmon, for instance, who has claimed that “Rome was thus able to rid herself of some landless poor, but that was not the main aim. the chief purpose of colonies was strategic” (cf. the discussion on Salmon’s views by Pelgrom and Stek in this volume).8 livy often portrays debates preceding the foundation of colonies in the early Republic, generally with the plebeians favoring colonization and the Senate resisting. Salmon considered it likely that there are genuine records of early Republican foundations but not of the debates about their foundation; such discussions were unnecessary given that colonies were “so obviously needed on military grounds”.9 Salmon’s work set out a narrative of colonization in the Mid-Republic as integral to Roman survival and success. He pointed to several striking episodes in the Hannibalic War when he believed Rome was saved by the loyalty of the latin colonies.10 for instance, Ariminum blocked Hannibal’s route around

1 2 3 4 5

e.g. laurence 1999, Chapter 2. Cornell 1995, 381. Cic., Leg. agr. 73; id., Font. 5.13. Sic. flacc. 135l = 99 th. App. B Civ. 1.7.

6 7 8 9 10

Asc. Pis. 3C. liv. 10.10.5. Salmon 1969, 15. Ibid., 189, note 203. Ibid., Chapter 5.




the east of the Appennines, and forced him into the more treacherous area of the etruscan marshes. After trasimene in 217 B.C., Hannibal was discouraged from attacking Rome by the opposition he encountered from the latin colony of Spoletium, which blocked the route south. farther south, the colony of Venusia provided a vital base for Marcellus, in the fight back against Hannibal. Salmon also made use of archaeological evidence which was beginning to emerge at the time of his book (1969). He noted that the fortifications of latin colonies, such as Cosa, are almost universal and formidable.11 they provide good evidence of the military purpose of colonies (at least acting as refuges). there is a standardized use of polygonal walling (probably refined as a latin technique) in the late 4th and early 3rd c. B.C. Colonies were walled soon after their foundation, which can be postulated from Hannibal’s repulse by Spoletium in 217 B.C., some twenty-four years after its foundation. In some Mid-Republican colonies, such as Cosa, towers and portcullises demonstrate contemporary Hellenistic influence. While obviously defensive, these walls, like roads, also had a cultural function: they were a demonstration of the Graeco-Roman concept of a city, and of visible power over the territory and a deterrent to local threats.12 Hostility towards colonies is not surprising. Colonization often involved the confiscation of land from a defeated enemy and its redistribution to incoming Roman or latin settlers. A colony would commonly have been surrounded by recently defeated Italian states. Pelgrom has recently made the interesting suggestion that the territories of the earliest latin colonies might not have been initially centuriated; but that the settlers huddled in defendable villages in the hinterland instead.13 this suggests that both the settlement pattern in the territory and the fortification of the town had defense in mind. A basic military function of colonization is therefore well established, with colonies serving to protect Roman territory and provide fortresses against local populations. the main issue which I want to highlight, however, is that most studies of Roman colonization go well beyond the idea of attributing a simple military function to colonies. Scholars have commonly assumed that colonies were part of a broader Roman strategy for the conquest and control of Italy. this consensus goes back at least as far as Mommsen.14 Characteristic of modern views is the role of colonies in the ‘Samnite Wars’. the strategy of surrounding the Samnites is thought to have determined virtually all colonial foundations between 334 and at least 285 B.C. However, the gap between this interpretation and the literary evidence for the motivations behind the foundation of colonies is often considerable. Hence Narnia, founded in southern umbria in 298 B.C., is often envisaged as having helped to encircle the Samnites and protect the Via Salaria. Nevertheless, Narnia lies a long way from the northern boundaries of Samnium, and in fact all livy says about its motivation (as noted above) is that it was founded ‘against the umbrians’. In a similar vein, the colonization in 313 B.C. of the island of Pontiae, taken from the Volsci, has been seen as guaranteeing the sea lanes to Campania should the Samnites sever the land route.15 Most of these arguments about a long-term Roman strategy against the Samnites are therefore interpretative guesswork.16

11 12 13

14



Salmon 1969, 29-39. Cf. the contribution by Sewell in this volume. Pelgrom 2008, 348-54 for clustered settlements with Stek 2009, 135-170 on colonial vici, cf. also Stek in this volume. e.g. Mommsen 1869, 474-75, on 313-11 B.C.: “the designs of the Romans were more and more fully developed; their object was the subjugation of Italy, which was enveloped more closely from year to year in a network of Roman fortresses and roads.” for a modern restatement, see loreto 2007, 75-97.

15 16

Salmon 1969, 59, 60. Polybius is probably the best evidence for broader Roman ambitions, but he does not (even with hindsight) say that Rome aims at the conquest of all Italy before the 280s B.C.: “the Romans […] now for the first time attacked the rest of Italy not as if it were a foreign country, but as if it rightfully belonged to them” (1.6.6).


It might be useful to compare these views to the idealizing images of Romanization in the edwardian era, as characterized by Chris Gosden: “the unfolding of the empire was guided by a rational plan on the part of the emperor and senate, giving the whole a unity of conception and practice.”17 this is also a view with the advantage of the hindsight of knowing that the Romans did go on to conquer Italy. the basic assumption is that colonies originally created for the Samnite Wars served a renewed defensive function in the Hannibalic War. In Salmon’s account, it is almost as if Rome had foreseen that it would face a bellicose invader some hundred years later. thereafter, however, Rome goes on to found colonies in the early 2nd c. B.C., ending in the 170s, despite facing no threat within Italy apart from that of the Gauls in the north, and henceforth fighting most major wars far distant from Italian soil. Once again highlighting the apparent prescience of Rome, the old importance of the latin colonies was briefly revived in the Social War, in which they were seen to act as Roman strongholds and hence became the focus of allied attacks. Hindsight is therefore critical to this view of colonies as planned bulwarks from which to thwart the enemies of Rome. Strategy implies a long-term vision and an overall geographical perspective. However, this idea of a long-term strategy in relation to Roman imperialism remains very controversial, even in much better attested periods of Roman history: for instance, in the debate generated by edward luttwak’s book The Grand Strategy of the Roman Empire.18 luttwak has been heavily criticized by fergus Millar who alleges that he had misunderstood the reactive nature of most imperial business.19 Benjamin Isaac argued that luttwak misconceived the potential threats, and that the Roman army in the east was more preoccupied with internal disorder.20 Susan Mattern states that critical decisions about where to place armies, whether to withdraw, whether to fight barbarians were discussed at a high level.21 However, this discussion was not conducted in the modernizing rational sense relating to maps and resources. It tended instead to concern the image of the emperor, the honor of the empire and the maintenance of terror and fear in the enemy. the most recent studies of imperialism provide a variety of views on the extent of a Roman strategy in the Mid-Republican period (cf. terrenato in this volume). William Harris’ work in War and Imperialism is well known. It emphasizes that war was driven by competition among the Roman elite and exceptional Roman militarism.22 Roman wars were not defensive, were largely fought abroad and were undertaken annually. Recent studies have adapted and modified Harris’ views. for example, in an influential article John Rich has argued that Harris (and his critics) had not appreciated that structural factors could cut both ways.23 Attempts by members of the Roman elite to restrict their rivals’ opportunities are well known: these included denying them triumphs or the environment in which they could have found an enemy to fight. Rich also suggests that Harris underplays the effect of fear on Rome which, though not always rational, was genuine. In another important contribution, tim Cornell develops Harris’ arguments about the unrestrained and haphazard nature of Roman belligerence. the Samnite Wars were not a ‘struggle for supremacy in Italy’, but essentially the result of unremitting Roman aggression.24 the Samnites were largely on the receiving end, and there is little genuine evidence of their united organization against Rome or that they posed a real threat to the city. In contrast, Arthur eckstein has portrayed Rome as struggling for survival in a competitive geo-political environment.25 He argues that Roman imperialism should not be interpreted solely in terms of internal Roman features: context matters too. Roman exceptionality is exaggerated by Harris: most of the imperialistic features of Roman society which he highlights can be paralleled in other Hellenistic societies.26 the process of imperialism must be understood in terms of the environment. from eckstein’s so-called ‘Realist’ perspective,

17 18 19 20 21

Gosden 2004, 105. luttwak 1976. Millar 1982. Isaac 1990. Mattern 1999.

22 23 24 25 26

Harris 1979. Rich 1993. Cornell 2004. eckstein 2006. Ibid., 238.




fIG. 2. Map of Roman Republican roads as reconstructed by Wiseman (after Wiseman 1970, 127 and 132).

this anarchic environment encouraged imperialistic aggression on all fronts, by all parties. He sees the Samnites and other peoples of Italy as active agents in the conflict. they were equally aggressive participants in the wars of the period, and genuine or at least perceived threats to Roman survival. It is critical to avoid too much recourse to hindsight. eckstein implies that there was little Roman control over the outcome. this, he claims, is borne out by modern political science comparisons, which show that military conflicts can go contrary to the plans of even the most powerful modern states (witness modern uS involvement in Vietnam or Afghanistan), and that planning for eventualities more than a decade in advance is rare. In summary, this recent work on imperialism by eckstein and others has significant implications for the understanding of colonization, particularly in terms of the difficulty of long-term planning in such an unpredictable environment, and the importance of understanding Roman imperialism in its broader context. As has been seen, it has been claimed that colonization is one of the best examples of Roman strategy at work and it is worth analyzing it in more detail. therefore the theme of this chapter will be to try to apply some of the insights in recent work on imperialism to Roman Republican colonization. Its purpose is not to deny that there was some planning or strategy involved in colonization. Nevertheless, I shall try to situate Roman colonization in its contemporary context, rather than viewing it from hindsight, and point out evidence suggesting individual and ad hoc initiatives rather than overarching organization.

ďœśďœ´


the main questions addressed are: I) Who directed the ‘strategy’ behind colonization, and in what interests were they acting? II) to what extent does the inter-linkage of colonies and roads suggest forward planning? this is certainly by no means an exhaustive approach, and there are many other topics germane to this issue which are beyond the scope of this paper. these include the extent of the similarities in the physical make-up of colonies and what this suggests about unified planning of them in Rome; what the cults of colonies reveal about their Roman roots and their relationship with local populations could also be considered.27

Decision making in the foundation of Latin colonies When the sources discuss the foundation of colonies, they describe the Senate as being the ultimate decision-making body. It is clear that Velleius and livy thought the Senate took the lead role.28 Particularly notable is livy’s report (4.49.6) that a tribune proposed a colony at both Bolae and labici, but was restrained by his colleagues who would not allow his proposal to be passed without the approval of the Senate. this is irrefutable evidence that livy considered the Senate to be the most important decisionmaking body. Independently of livy, festus explicitly records the passing of a senatus consultum for the colonization of Saticula in 313, and provides the names of its triumviri.29 this much is clear but it is more difficult to assume that these decisions were part of a senatorial strategy, sustained over many years. Another passage in festus suggests that it was not until the late 4th c. B.C. that membership of the Senate was formalized in the way it is known from late Republican sources: before then, the consuls used to appoint their close associates, hence there was no disgrace in being passed over. Presumably therefore, before the late 4th c. B.C., the composition of the Senate might have changed year to year.30 Without wishing to push this argument too far given the ambiguities in the passage, it does seem to raise questions about the capability of the earlier Senate to formulate long-term strategy. Hence it might be anachronistic to regard

27

28

29 30

for works addressing these issues see Sewell 2010; torelli 1999; Bispham 2006; Glinister 2009; Boos in this volume; and Stek and Burgers forthcoming. Vell. Pat. 1.14.1: “I have decided to separate the first part of this work from the second by a useful summary, and to insert in this place an account, with the date, of each colony founded by order of the senate since the capture of Rome by the Gauls”. for discussion of this topic see Wiegel 1983; laffi 1988; Gargola 1995, 51-8; and Oakley 1997, 572. festus, Gloss. Lat. 458. festus, Gloss. Lat. 290 on the composition of the senate before the late 4th c. B.C.: “Praeteriti senatores quondam in opprobrio non erant, quod, ut reges sibi legebant, sublegebantque, quos in consilio publico haberent, ita post exactos eos consules quoque et tribuni militum consulari potestate coniunctissimos sibi quosque patriciorum, et deinde plebeiorum legebant; donec Ovinia tribunicia intervenit, qua sanctum est, ut

censores ex omni ordine optimum quemque curiatim in senatum legerent. Quo factum est, ut qui praeteriti essent et loco moti, haberentur ignominiosi.” (“there was once a time when it was not considered disgraceful for senators to be passed over, because, just as the kings by themselves used to choose (or to choose as replacements) men who would serve them as public advisers, so under the Republic the consuls or military tribunes with consular power used to choose for themselves their closest friends from among the patricians and then from among the plebeians. this practice continued until the law of the tribune Ovinius put an end to it. Ovinius’ law (between 339 and 318 B.C.) bound the censors to enrol in the Senate the best men from all ranks by curia. the enforcement of this law had the consequence that senators who were passed over, and thus lost their place, were held in dishonour.”) See Cornell 2000 (from where the translation is taken); reservations are expressed by Richard 2005, 121-22.




it as having much identity as a body before 300 B.C., and sources are perhaps simply retrojecting the situation they knew in the later Republic. Also revealing is what happened to Postumius Megellus in 291 B.C. Successful conquerors as he was could expect to occupy a leading position in the triumvirate distributing the land won. Postumius was denied this usual privilege on account of his insufferable arrogance, which had led him to employ 2,000 of his soldiers on his estates (Dion. Hall., Ant. Rom. 17-18, 4; cf. Marchi in this volume). this episode may be a sign that the Senate was beginning to assert its right to decide on such matters, turning its back on traditional precedent. Just such a principle is outlined in Antium in 467 B.C., where titus Quinctius was appointed head of the triumvirate because he had conquered the land (livy 3.1.5-7; cf. Attema et al. in this volume). Siculus flaccus does say that members of the commission for viritane land distribution schemes could reserve land for themselves and their friends, and it is plausible that the same construction could also have applied to colonial schemes.31 they would also have been in a position to win over clients, as they decided who should be enrolled on the list for land allotments. So it seems that it was customary for a successful general to take the lead in distributing conquered land, allowing him to benefit from this privilege in a variety of ways by ensuring that land he had conquered was colonized.32 Plutarch explicitly records that titus flamininus had benefited politically from the support of the colonists he had settled in Narnia and Cosa.33 However, there are problems with this story: latin colonists could not vote for flamininus, and could only have provided support in other, less tangible, ways; Plutarch might also have been mistaken, given that livy places flamininus as a triumvir in Venusia. Nevertheless, Plutarch must be right in principle, namely: serving on a colonial commission to distribute land allotments was a source of a powerful client base.34 therefore, in the earliest phases of colonization, an important role was played by the initiative taken by the conquering general, even if the formal decision was taken by the Senate. It was only after the formalization of the Senate in the late 4th century that the preconditions for the development of longer term strategic thinking developed.35

Road building and colonization My second area of investigation is the relationship between colonies and roads (fig. 2). It should be noted that there is considerable difficulty in dating major consular roads. the sources for roads are much less systematic than those for colonies, which were recorded with their year of foundation in the documentary records which lie behind the accounts of livy and Velleius Paterculus. this is an interesting fact in itself. the close links between colonization and road building had already been appreciated by theodor Mommsen, and have been highlighted more recently by filippo Coarelli, stressing the early dates of many road projects.36 for example, the Via Valeria was probably created by the censor M. Valerius Maximus in 307 B.C. the road just predates the extermination of the Aequi (304 B.C.) and also the foundation of colonies at Alba (303 B.C.) and Carseoli (298 B.C.) by a few more years. therefore it does seem that this road did presage the conquest, annexation and definitive organization of this territory. Another example is the Via Aurelia, best dated to the censorship of C. Aurelius

31

32

33



Sic. flacc., Cond. agr. 157.7-8l; see Coles 2009, 73, note 85. See also Diod. Sic. 19.101. for a further discussion of this principle, see Bradley 2006. Cf. also terrenato in this volume. Plut., Vit. Flam. 1.4-2.1 (transl. Perrin, 1921): “for this reason he was also chosen director-in-chief of the colonists sent out to the two cities of Narnia and Cosa. this success more than anything else so exalted his ambition that he ignored the intervening of-

34 35 36

fices which young men generally sought, the offices of tribune, praetor, and aedile, and thought himself worthy at once of a consulship; so he became a candidate for that office, with the eager support of his colonists.” Cf. torelli 1999, 93. Hölkeskamp 1993, 34-35. Coarelli 1988. Cf. Mommsen 1869, 84. Cf. more recently De Cazanove 2005, 116.


Cotta in 241 B.C., though other dates have been suggested. the road followed a series of interventions in this district: the definitive conquest of coastal etruria (280-73 B.C.); an alliance with Pisa (at the terminus of the route) soon after; the foundation of latin Cosa at the farthest point of Roman territory in 273 B.C.; then the foundation of citizen colonies at Castrum Novum (264 B.C.), Pyrgi (264 B.C. or just after), Alsium (247 B.C.) and fregenae (245 B.C.). Coarelli sees the intention as defence against the Carthaginians on Sardinia and elsewhere, and planning for the campaigns in liguria and perhaps even farther afield.37 Roads are frequently associated with colony foundation and other measures such as viritane distribution and the foundation of fora. the overall result of this process by the time of the Gracchi had given much of Italy an appearance of a hegemonic, conquered landscape.38 Nevertheless, it is important to note that, once again, hindsight makes the whole operation appear much more planned than it probably was, and the evidence for a systematized program of colonization and road building needs to be weighed up against other considerations. unquestionably, the picture varied greatly between regions, depending on the circumstances of the conquest and pre-existing trends towards urbanism in a particular area: etruria does not seem to have required the same level of colonization for political control to work effectively there.39 Roads did not automatically precede the colonization of an area, indeed often considerably post-date it. A clear case is the Via flaminia, created in 220 B.C., some forty-eight years after the foundation of its terminus at Ariminum (and longer after the foundation of Sena Gallica). Moreover, the foundation of colonies on road routes appears to have been rather haphazard. for instance, Spoletium was founded in 241 B.C., therefore well after Ariminum in 268 B.C., yet the latter was situated in much more unsettled country. Its creation might have been a response to the revolt of falerii in the same year (241 B.C.), although no linkage can be attested beyond the date; the settlement of first Punic War veterans is also a possibility.40 Spoletium was also, curiously, left off the western branch of the Via flaminia. Strabo’s description of umbria (5.2.10), and polygonal remains of a bridge on this branch suggest that this was the earlier branch.41 flaminius’ road was perhaps designed for military purposes, the conquest of Gallia Cisalpina in particular, and to benefit his settlers in the ager Gallicus, rather than for providing Spoletium with links to the capital (cf. Vermeulen in this volume). Another example is Carseoli, founded in 298 B.C. some five years after the more distant colony of Alba (founded in 303 B.C.). In a more rational system, the foundation of colonies would have proceeded outwards from Rome. Although it is always possible to try to rationalize such oddities, there was certainly an ad hoc element in the decision making. A similar conclusion might be drawn from the colonies which were not connected to Rome by major consular roads. A few examples can be listed, although the gaps in the evidence mean that proving a negative irrefutably is difficult. I) Paestum never lay on a major consular road and, when the Via Popilia was built from Capua to Rhegium, it bypassed the colony. II) As was Alba fucens, luceria was on a major drove road, but luceria was not on a consular road. III) the Adriatic colonies of firmum, Sena Gallica, Auximum and Potentia were connected only by diverticula of the Via flaminia and Via Salaria. IV) As already mentioned, Spoletium might only have become part of the flaminia later, when an eastern branch was built. this is not to say that these colonies were left isolated from all communication. they must have been linked to surrounding towns and ultimately Rome by pre-existing routes, and eventually diverticula of the

37 38

Coarelli 1988, 47. Dench 2005, 164, emphasising that this is the ‘end-result’.

39 40 41

Harris 1979, 98, 156. Salmon 1969, 65. Bradley 2013.




major consular roads, as noted above. It is known that etruscan cities were connected by a road system which predated the Roman conquest and this does raise questions about the nature of any central Roman co-ordination of roads and colonies. It is also odd from this perspective that, in some cases, allied cities rather than colonies were the termini of some roads: Ameria for the Via Amerina; Asculum for the Via Salaria, until it was extended to Castrum truentum; and probably also Pisa, Coarelli’s suggestion for the Via Aurelia. the overall pattern might therefore be better characterized as the product of convergent pressures, rather than as the result of a coherent program for linking colonies to Rome. Another important issue is the extent to which the Senate, as a body, can be said to have played a controlling role in the building of roads. One of the most intriguing features of road building, recently highlighted by Ray laurence, is that it frequently emerges as populist in intent, even demagogic in character.42 Building roads involved spending public funds on projects which created employment, and it was also often linked with land distributions beneficial to the plebs. the building of the Via Appia by Appius Claudius, and his other activities after being elected to the censorship in 312 B.C., clearly aroused a storm of controversy, judging by the accounts in livy and Diodorus.43 livy notes the positive effect the road had on Appius’ later reputation. But it seems clear from Diodorus that this project was also very unpopular with the majority of the Roman elite. this was the first Roman road to have been named after the magistrate responsible for its construction. the names of earlier roads reflected their purpose, such as the Via Salaria, or the region they traversed, such as the Via latina and the Via Campana, or the town to which they led, such as the Nomentana, Collatina, tiburtina, Gabina, Praenestina and Ardeatina. the only previous naming of a monument after an individual is the columna Maenia of the 330s. Significantly, unlike Maenius’ column, this road took Appius’ praenomen, as did the Aqua Appia and forum Appi, also created in his censorship.44 A clear contemporary model is the early Hellenistic colonization and road building of Philip and Alexander, who bestowed their own given names on the cities they founded, and connected new foundations like Philippi to the capital by Macedonian royal roads (Cf. Sewell this volume).45 Appius’ ambitions are clearly exposed in Diodorus, and the road should best be weighed up in the context of his other populist actions. linked to this popularist interpretation was the continuing work on the Via Appia after the end of Claudius’ censorship. the curule

42 43



laurence 1999, Chapter 4. liv. 9.29.5-9 (312): “[…] et censura clara eo anno Ap. Claudi et C. Plauti fuit; memoriae tamen felicioris ad posteros nomen Appi, quod uiam muniuit et aquam in urbem duxit; eaque unus perfecit quia ob infamem atque inuidiosam senatus lectionem uerecundia uictus collega magistratu se abdicauerat, Appius iam inde antiquitus insitam pertinaciam familiae gerendo solus censuram obtinuit.” (“Noteworthy, too, in that year was the censorship of Appius Claudius and Gaius Plautus; but the name of Appius was of happier memory with succeeding generations, because he built a road, and conveyed water into the city. these undertakings he carried out by himself, since his colleague had resigned, overcome with shame at the disgraceful and invidious manner in which Appius revised the list of senators; and Appius, exhibiting the obstinancy which had marked his family from the earliest days, exercised the censorship alone.”). Diod. Sic. 20.36 (transl. Oldfather, 1954): “In the first place he

44

45

built the Appian Aqueduct, as it is called, from a distance of eighty stades to Rome, and spent a large sum of public money for this construction without a decree of the Senate. Next he paved with solid stone the greater part of the Appian Way, which was named for him, from Rome to Capua, the distance being more than a thousand stades. And since he dug through elevated places and levelled with noteworthy fills the ravines and valleys, he expended the entire revenue of the state but left behind a deathless monument to himself, having been ambitious in the public interest.” It is possible that Appius was used as a nomen, as appears the case in CIL VI 1280 Appios consol (see note to frontin. 5.1 s.v. “aqua Appia”, in Rodgers 2004, 144). Humm 1996. Note also the parallel with the ‘royal road’ created in the Persian empire (Dench 2005, 164), suggesting that Appius’ road could have evoked monarchic power as well as the secure control of Rome over its territory.


aediles paved parts of the Appia in the vicinity of Rome in 295 and 292 B.C. using fines levied on usurers and graziers, who must have been drawn from the ranks of the wealthy elite of Mid-Republican society.46 Another interesting example is the creation of the Via flaminia by Gaius flaminius when he was censor in 220 B.C. In 232 B.C. he had enacted the viritane distribution of the ager Gallicus as tribune, a move which was very unpopular with the Senate. Polybius must be reflecting the contemporary rhetoric of the senatorial opposition to flaminius when he claims that it provoked the Gauls to go to war: “Gaius flaminius was the originator of this popular policy, which we must pronounce to have been, one may say, the first step in the demoralization of the populace, as well as the cause of the war with the Gauls which followed”.47 further evidence of his poor relations with the rest of the Senate comes in 223 B.C. when he invaded Gaul and crossed the River Po, despite the attempts of the Senate to withdraw him on religious grounds.48 the negative portrayal of his character in the sources culminates in the description in livy of his defeat and death at trasimene in 217 B.C., brought on by his rash enthusiasm for battle. the building of the Via flaminia fits this radical profile, providing employment for the urban poor and a link to flaminius’ allotments in the ager Gallicus, as well as a route to the hostile areas of Northern Italy. He therefore cuts a controversial figure, who seems often to operate in conflict with the Senate, rather than acting according to its wishes. the association between roads and radical populism continued in the late Republic. Gaius Gracchus is described by Plutarch as a tremendous road builder, and Caesar is noted as having built up support by his massive expenditure as curator viarum.49 Significantly, Augustus picked the flaminia to repair, encouraging senators to restore the rest. this partly served a practical purpose: the flaminia was the main road north, and northern Italy and Illyricum were the subject of major military activity in his reign. It also allowed him to promote the image of tota Italia, as this was the main artery to the newly incorporated Po Valley. But another implicit point is Augustus’ ostensible concern for the plebs, as he might have had the idea of appropriating the legacy of flaminius as a people’s champion at the back of his mind. In short, it is difficult to see how it can be meaningful to ascribe a senatorial strategy to such roads and associated schemes, given that they were often created by individuals hampered and attacked by the majority of the Senate. the sense is often rather that the role of the Senate was to restrain the potential power which such figures as Postumius Megellus and Gaius flaminius could acquire. It did so by regulating the competitive forces encouraging individuals to propose colonies and create roads.

Variety in Mid-Republican colonies At this point it is useful to consider some other types of evidence, mainly material in form. these are more contemporary than the literary sources and are largely uncontaminated by the hindsight of a later imperial perspective. this highlights the variety in Mid-Republican colonies. two main areas can be considered, coinage and colony names, which reveal something of the contemporary ideology of the early colonists and colony founders. It is well known that the coinage of latin colonies in the Mid-Republic is very diverse and shows a range of local influences. Many produced their own issues of bronze coinage from the first quarter of the 3rd c. B.C. down to the Hannibalic War. Most strikingly, silver coinage, symbolic of their political autonomy, was minted by five colonies founded before 268 B.C. (Cales, Suessa, Alba, Cosa and Paestum).50 A wide variety of weight standards is also revealed in the coinage of colonies. Hadria and Ariminum used an ‘Adriatic’ standard shared with the Vestini; in contrast nearby firmum had a Roman weight

46 47 48

liv. 10.23.12-13; 10.47.4; laurence 1999, 18. Polyb. 2.21; laurence 1999, 21 with further references. Plut., Vit. Marc. 4.

49

50

Plut., Vit. C. Gracch. 6-7; Plut., Vit. Caesar 5; cf. laurence 1999, 40. Salmon 1969, 85; Crawford 1985, 45-48.




standard. Alba, Norba and Signia produced coins on Greek lines, which as Michael Crawford has argued were borrowed independently of Rome, during the Pyrrhic War.51 large issues of bronze coinage were issued by the colonies around Campania (Cales, Aesernia and others), borrowing types from Naples. farther south, there was a locally patterned bronze coinage produced by luceria and Venusia in the Hannibalic War.52 the names of colonies are also diverse and symbolic, as Salmon and Mario torelli have shown.53 Some were named after values: firmum suggests strength and support, Potentia power, Copia (abundance) was the new name for colonized thurii and Hipponium became Vibo Valentia (power, health). Some adopted the names of gods and goddesses, including Saturnia, Neptunia for tarentum, Minervia for Scolacium, Junonia for Carthage and Narbo Martius.54 In a similar vein, Alba fucens was apparently named after the legendary latin city Alba longa. Other names sometimes replaced inauspicious local names on sites on which earlier settlements predated Roman colonization: Narnia instead of the local Nequinum and Beneventum (welcome) replaced Malventum (ill met).55 local topographic and ethnic names were also quite common: Narnia was apparently taken from the umbrian name of the River Nar; Cosa was perhaps from the etruscan Cusi or Cusia, Paestum from lucanian Paistom.56 What emerges is the variety of naming patterns, suggesting that individual choice was critical: its name would seem to have been up to the chief triumvir responsible for the foundation. It is interesting here to compare the Via Appia, named after its creator: there were no similar colony names until the triumviral and early Imperial period, when the monarchic implications of such a practice were more acceptable.

Conclusion In conclusion, the development of Mid-Republican colonization, beginning on a large scale from the foundation of Cales in 334 B.C., and the associated road system, was less the product of an overall guiding senatorial strategy, and more a case of structural pressures operating in certain directions. Colonies and roads enhanced the political support and social status of their founders, as land distributions and employment opportunities were popular with the plebs. Colonization schemes reduced civil strife in Rome, and also helped establish a Roman presence in conquered territory. this normally involved the confiscation of land from the conquered communities, and colonies were strongly fortified to provide for their defense. However, the hypothesis that there was a grand strategy for the conquest of Italy which guided Roman colonization in the Mid-Republic is a product of hindsight. Initially, this hypothesis seems quite attractive, looking back from the late Republic, particularly with the aid of modern maps, as an explanation of the dense network of roads across Italy which eventually linked most colonies directly to Rome. But the preconditions to support it were largely lacking in the Mid-Republic, namely: a permanent body to oversee it and the context in which something other than mere survival could be contemplated. Road-building initiatives were often enacted by populist and individualistic politicians, who seem to have been working against, rather than with, the wishes of the Senate. Colonial foundations also owed much to the initiative of the generals who conquered the area. the fiercely competitive ethos of the Roman elite, renewed by the admission of plebeians to the highest offices and priesthoods from the later 4th c. B.C., drove these projects forward without any need of a senatorial strategy, at least in its earliest phases. there is no doubt that Rome did develop an aspiration to control all of Italy, and in this sense a longer term strategy did develop in the 3rd c. B.C.57 In vying for control of the Mediterranean with the great

51 52 53 54

ďœˇďœ°

Crawford 1985, 47. Ibid., 65. Salmon 1969; torelli 1999. Cf. on the preservation of the indigenous name of Venusia, torelli 1999, 93-94.

55 56 57

liv. 10.10.5. Brown 1980, 8; torelli 1999, 43. Cf. HĂślkeskamp 1993.


imperialistic power of Carthage and the Hellenistic monarchies, Rome became a more consciously imperialistic state. However, it is notable that Polybius – from the hindsight of the second half of the 2nd c. B.C. – sees full control of Italy as the aim of Rome only on the eve of the war with Pyrrhus, in 280 B.C.58 the diverse names and coinage of latin colonies, and the variety of different statuses applied to defeated enemies speaks strongly of fragmented Roman aims in its settling conquered areas. this ad hoc and flexible element in Roman imperialism might be said to have been one of its key strengths, allowing it to survive such catastrophes as trasimene and Cannae, and marking it out from Hellenistic monarchies.

Acknowledgements this chapter has benefited from points raised by audiences in Cardiff, Glasgow, london and Nijmegen, and from discussions with Alun Williams during his Ph.D.

References Bispham, e. 2006. “Coloniam deducere: how Roman was Roman colonization during the Middle Republic?” in G. Bradley and J.-P. Wilson (edd.), Greek and Roman colonization: origins, ideologies and interactions (Swansea) 73-160. Bradley, G. 2006. “Colonization and identiy in Republican Italy,” in G. Bradley and J.-P. Wilson (edd.), Greek and Roman colonization: origins, ideologies and interactions (Swansea) 161-87. Bradley, G. 2013. “umbria and Picenum,” in C. J. Smith (ed.), The Cambridge Ancient History, vol. 7: Plates to Volumes VII, Part 2 and VIII (Cambridge) 82-97. Brown, f. 1980. Cosa: the making of a Roman town (Ann Arbor, MI). Castagnoli, f. 1956a. “la centuriazione di Cosa,” MAAR 24, 149-65. Coarelli, f. 1988. “Colonizzazione romana e viabilità,” DialArch 6, 35-48. Coles, A. J. 2009. Not effigies parvae populi romani: gods, agency, and landscape in MidRepublican colonization (unpublished Ph.D thesis, univ. of Pennsylvania). Cornell, t. 1995. The beginnings of Rome (london).

58

Cornell, t. 2000. “the lex Ovinia and the emancipation of the Senate,” in C. Bruun (ed.), The Roman Middle Republic: politics, religion, and historiography c. 400-133 (Rome) 69-89. Cornell, t. 2004. “Deconstructing the Samnite Wars,” in H. Jones (ed.), Samnium: settlement and cultural change. The proceedings of the third E. Togo Salmon conference on Roman studies (Providence, RI) 35-50. Crawford, M. H. 1985. Coinage and money under the Roman republic: Italy and the Mediterranean economy (london). De Cazanove, O. 2005. “les colonies latines et les frontières régionales de l’Italie. Venusia et Horace entre Apulie et lucanie: Satires, II, 1, 34,” Mélanges de la Casa de Velázquez 35-2, 107-24. Dench, e. 2005. Romulus’ Asylum. Roman identities from the age of Alexander to the age of Hadrian (Oxford). eckstein, A. M. 2006. Mediterranean anarchy, interstate war, and the rise of Rome (Berkeley). Gargola, J. D. 1995. Land, laws, and gods. Magistrates and ceremony in the regulation of

Polyb. 1.6.6, discussed above. Compare his characterization (1.20.1-2) of the opportunistic shift in Roman aims during the campaign of 261 B.C., from helping the Mamertines to expelling the Carthaginians from Sicily.




public lands in Republican Rome (Chapel Hill, NC/london). Glinister, f. 2009. “Burning boats and building bridges: women and cult in Roman colonisation,” in e. Herring and K. lomas (edd.), Gender identities in Italy in the First Millennium BC (Oxford) 117-26. Gosden, C. 2004. Archaeology and colonialism: cultural contact from 5000 BC to the present (Cambridge). Harris, W. V. 1979. War and imperialism in Republican Rome, 327-70 BC (Oxford). Hölkeskamp, K.-J. 1993. “Conquest, competition and consensus: Roman expansion in Italy and the rise of the nobilitas,” Historia 42, 12-39. Humm, M. 1996. “Appius Claudius Caecus et la construction de la Via Appia,” MÉFRA 108, 693746. Isaac, B. 1990. The limits of empire: the Roman army in the East (Oxford). laffi, u. 1988. “la colonizzazione romana tra la guerra latina e l’età dei Gracchi: aspetti istituzionali,” DialArch 6-2, 23-33. laurence, R. 1999. The roads of Roman Italy: mobility and cultural change (london). loreto, l. 2007. La grande strategia di Roma nell’età della prima guerra punica (ca. 273-ca. 229 a.C.): l’inizio di un paradosso (Naples). luttwak, e. 1976. The grand strategy of the Roman Empire from the first century A.D. to the third (Baltimore/london). Mattern, S. 1999. Rome and the enemy: imperial strategy in the principate (Berkeley/london). Millar, f. 1982. “emperors, frontiers and foreign relations, 31 B.C. to A.D. 378,” Britannia 13, 123. Mommsen, t. 1869. History of Rome (New york).



Oakley, S. P. 1997. A commentary on Livy: books VIX, vol. 1: Introduction and book VI (Oxford). Pelgrom, J. 2008. “Settlement organization and land distribution in latin colonies before the Second Punic War,” in l. de ligt and S. J. Northwood (edd.), People, land and politics: demographic developments and the transformation of Roman Italy 300 BC-AD 14 (leiden) 333-72. Rich, J. 1993. “fear, greed and glory: the causes of Roman war-making in the Middle Republic,” in J. Rich and G. Shipley (edd.), War and society in the Roman world (london) 38-68. Richard, J.-C. 2005. “Patricians and plebeians: the origins of a social dichotomy,” in K. A. Raaflaub (ed.), Struggles in Archaic Rome: New Perspectives on the Conflict of the Orders (Oxford) 107-27. Rodgers, R. H. (ed.) 2004. Frontinus: De aquaeductu urbis Romae (Cambridge). Salmon, e. t. 1969. Roman colonization under the Republic (london). Sewell, J. 2010. The formation of Roman urbanism, 338-200 B.C.: between contemporary foreign influence and Roman tradition (Portsmouth, RI). Stek, t. D. 2009. Cult places and cultural change in Republican Italy. A contextual approach to religious aspects of rural society after the Roman conquest (Amsterdam). Stek, t. D. and G.-J. Burgers (edd.) forthcoming. The impact of Rome on cult places and religion in Italy. torelli, M. 1999. Tota Italia: essays in the cultural formation of Roman Italy (Oxford/New york). Weigel, R. D. 1983. “Roman colonization and the tribal assembly,” PP 38, 191-96. Wiseman, t. P. 1970. “Roman Republican RoadBuilding,” PBSR 38, 122-153.


Roman colonization and the city-state model Jeremia Pelgrom “[…] a colonia was a city-state. It was not a geographical region or administrative subdivision of the Roman state but an urban commonwealth with its immediate surrounding territory.” (Salmon 1969, 14).

According to edward togo Salmon, the quintessence of a Roman colony was its organization as a citystate. this now conventional view was certainly not new at the time Salmon wrote his seminal book; it can be traced back at least to the great German historians of the late 19th and early 20th century (cf. Pelgrom and Stek in this volume).1 Salmon does not provide us with a detailed description of what he believes constitutes a city-state, but it is clear from his introduction that self-government and urbanism are its two main components.2 these elements are essential to him above all because they differentiate Roman colonization from other colonial strategies. Whereas political and economic independence separate Roman colonies from most modern colonial models, urbanism is the crucial variable that distinguishes Roman colonization from ‘primitive’ Italic forms of colonization, such as, for example, the migration of tribal groups from the Apennines to coastal areas.3 these two variables in combination are of course characteristic of Greek civic ideology; they thus would allow us to place Rome and her colonies at the same stage of sociopolitical evolution as the Greek colonial world.4 Indeed, if modern reconstructions of the territorial organization of Mid-Republican Roman colonies and Greek poleis are compared striking similarities can be seen immediately (compare figs. 1a and 1b). Both are conceptualized as clearly defined territorial units of moderate size, encompassing an urban administrative center, a regularly partitioned hinterland, and a transitional zone where the remaining indigenous population lived. these close physical similarities cannot be coincidental, but rather must be the result of a shared vision of societal organization.5 In fact, the proposed geopolitical organization of Mid-Republican Roman colonies corresponds very closely to the ideal community envisioned in utopian writings of Classical and early Hellenistic Greece.6 We can easily recognize the implementation of Hippodamic views of political and economic equality in the rigidly orthogonal division of space in both town and countryside. Hippodamus’ ideas of functional differentiation can also be identified in the

1 2

3

e.g., Marquardt 1881, 52-53, and p. 16 on territorium. this is clear from the quotation above, which categorizes a colony as an ‘urban commonwealth’. Although ‘commonwealth’ can have different meanings, it is clear that for Salmon it meant a political state or nation in which the people are sovereign (see also Salmon 1953, 93; id. 1982, 159). Id. 1969, 13-19. On several occasions, Salmon sets out to differentiate Roman colonization from modern colonialism (esp. p. 13 and 14). the crucial difference for him is that a Roman colonia, unlike modern colonies, is not an (overseas) subject territory in a large empire, but a new, self-governing political community. this emphasis on ur-

4

5

6

banism as a crucial criterion that distinguishes Roman from Italic-Apennine societies is clearly expressed in his monograph on the Samnites (id. 1967, 51-52, 78-79). See for example De Ruggiero 1896, 7-9 for an older view that explicitly highlights the differences between Greek and Roman colonization. See especially Sewell 2010, 21-54 and Sewell in this volume for a good discussion of Greek influences on Roman colonial urban organization. See Cahill 2002 for these theories. See also Sewell 2010, 64-67, 131-36, who likewise discusses the differences between Greek and Roman city planning in the 4th and 3rd c. B.C.




fIGS. 1A AND B. the territorial reconstruction of the Greek colony Metapontion (A) and of the latin colony of Cosa (B) (after Carter 2006, 118, fig. 3.32 and Carandini et al. 2002, 107, fig. 40). In dark grey the supposed living areas of the colonists (called the chora in Greek context and ager centuriatus in Roman colonial context). In light grey the undivided lands where the remaining indigenous population might have lived.

categorical division of urban space into separate areas for religious, private, and public activities.7 Apart from Hippodamic socio-geometric principles, we can also recognize the influence of Platonic political theory, such as the allocation of land according to property classes8 with allotment sizes just large enough to guarantee autarchy, yet small enough so as not to have a corrupting influence.9 Another element that echoes Plato’s utopian vision of Magnesia is the obsession with maintaining stable population levels, as witnessed by rules restricting mobility and the sale of allotments.10

7

8

9



for examples of latin colonies with clear functional differentiation in their city planning, see Cosa (Brown 1980, fentress 2000, with further references) and Alba fucens (Mertens 1969). Laws 744c. Cf. the Roman practice of assigning different, but proportional, plot sizes to people of different rank (see Pelgrom 2008, 338 for a list of Roman land allotments recorded in the sources). Laws 737d. the Roman ideal of the self-sufficient colonial soldier-farmer is expressed in colonial allotments that are just large enough for a family to live off of (see Dyson 1992, 26-27; Pelgrom and Stek in this volume). the numerous Roman anecdotes of good leaders who reject wealth (e.g., the stories about Cincinnatus, Dentatus and Atilius Regulus) tally nicely with the austere ideals expressed by Plato.

10

Laws 740b-e, 742b. Also Sewell 2010, 103. On geographical fixity in Roman colonial policies, see Broadhead 2001; id. 2008. Note especially the law that a Magnesian citizen should leave always one son behind (cf. liv. 41.8.9). even in a demographic sense, latin colonies seem to approach Platonic views of ideal community size and composition. Plato’s ideal size of 5,040 citizens is of the same order as the 2,500, 4,000 or 6,000 colonists that were sent to latin colonies. One might even recognize some similarity in Plato’s system of geographic subdivision of colonial territory into phyles and demes/villages with the pagi and vici known to have existed in latin colonial territories. for the Roman colonial world, see Stek 2009, 123-71 with further references.


Despite these apparent and real similarities in design and organization, this paper will argue that the implicit adoption of the ideal Greek city-state as a model for understanding Roman colonization has severely narrowed our understanding of Roman colonial practices. this does not deny that the Greek and Roman worlds strongly influenced one another, but that an exclusive comparative focus on Greece prevents us from recognizing other patterns and influences (cf. Sewell in this volume).11 the adoption of the static citystate model in particular has prevented us from seeing changes in Roman colonial practices over time.12 Rome and her colonies continued to be influenced by Hellenism throughout their history, and there is reason to believe this process intensified after the conquest of the Greek world. We should not, therefore, take the fact that some Roman colonies, eventually, in their monumental and archaeologically more easily recognizable forms, came to follow Hellenistic models as confirmation that they were designed as such from the start. Hellenism was not about directly copying Greek models, but about adapting, selecting, and reinterpreting within the existing cultural framework single features from the large pool of Hellenistic cultural elements.13 Careful diachronic analysis, therefore, is essential for understanding acculturation processes in the Hellenistic world, while also focusing on elements of change and difference. this is all the more important because a dynamic and varied understanding of Roman colonization has serious implications for our understanding of Roman imperial success. It raises the old Polybian question of whether the Roman conquest of culturally superior Greece was rooted in a different, militarily and politically more effective form of societal organization, or rather should be understood as the victory of one polis over others. We know Polybius’ answer – a view shared by Salmon14 – but there are good reasons to believe that Roman societal organization and colonial strategy in the formative age of imperial success differed significantly from the city-state model.15 In a previous article, attention was drawn to the fact that there actually is little firm evidence for the presumed existence of orderly and monumental Roman colonial landscapes in the period before the Punic Wars.16 the hypothetical city-state settlement model, consisting of a single urban core and an egalitarian and strictly organized hinterland, is not firmly attested until the late 3rd to early 2nd c. B.C. thus only after Rome conquered Italy, the Carthaginians, and the Greek city-states in Southern Italy, do we begin to see convincing archaeological proof that such landscapes were constructed. If correct, this evidence significantly undermines the view that the spread of ‘attractive’ city-state culture throughout Italy by means of founding colonies spurred the cultural unification of Italy, which in itself is considered an important explanation for the remarkable stability of the Roman commonwealth.17 Here the focus is on the second crucial component of the city-state model: the view that a latin colony is a commonwealth, an independent territorial state bound only to Rome by bilateral treaty. the view that a colony is a politically and economically sovereign territorial state is central to the city-state model.18 Moreover,

11

12

13

14

We should not forget that Rome founded at least nineteen colonies between the latin War and the first Punic War. On the traditional city-state view, that would imply the creation of one Olynthus- type city every three to four years (and that is excluding the creation of the smaller citizen colonies). Also Sherwin-White 1973, 86; Crawford 1995; and Bispham 2006 for the view that colonial practices changed under the Middle Republic. for a recent discussion, see Wallace-Hadrill 2008, 1728 (with further references) and Stek 2009, esp. chapter 3. Salmon 1982, 159: “Admittedly even in Italy the urban

15

16 17

18

commonwealth was not an original Roman idea. But the Roman community was the most successful specimen of a city-state”. See especially Ando 1999 on the historiography of ancient Greek authors trying to explain Roman rule and societal organization through typical Greek conceptual models. Pelgrom 2008. Salmon 1982, 158-60. See also Stek 2009, esp. 25-28 for a critical view. for a good discussion on the concepts of the city-state and territorial sovereignty, see Hansen 2000, 11-34. for Hansen, a city-state is necessarily also a territorial state.




fIG. 2. Map of Central Italy in 241 B.C. (after toynbee 1965, map 2).

it is crucial to the interpretation of the early Roman imperium as a conglomeration of individual states working together for the glory of Rome.19 Various studies have highlighted the limitations on the political autonomy of Roman and latin colonies, drawing attention to the fact that colonies lacked the right to an independent foreign policy and required Roman permission for such elementary decisions as the enrollment of new citizens.20 But the view that coloniae, especially those of the latin type, were independent territorial states seems to remain largely unquestioned. As is argued below, the partial political autonomy colonies enjoyed does not necessarily lead to the conclusion that they were organized as territorial states. this inference is a dangerous assumption drawn largely from the hypothetical analogy with the Greek city-state model, which itself may be a false interpretation of the ancient Greek societal organization.21

What is a colonia? We should start our analysis by considering the term colonia itself and asking whether it necessarily implies a territorial state in the modern sense of the word. for Salmon, this was clearly the case, but even he admits that etymologically the word colonia was a collective noun meaning a body of coloni. He goes on to explain, however, that it technically designated “a group of settlers established by the Roman state, collectively and with formal ceremony, in a specified locality to form a self-administering civic community”.22

19



On this, see Ando 1999, 22, who draws attention to the fact that the understanding of the Roman imperium as a sort of league is rooted in Greek conceptual frameworks that were incapable of understanding the unity of people in any form other than a league, i.e., as individual poleis working together toward a common goal. See also Coarelli 2005, 24-25, who rejects the idea that colonies were separated politically

20 21

22

from Rome (contra Mouritsen 2004) and explicitly argues that they were completely different from Greek colonies in this respect. Cf. erdkamp 2011. undoubtedly, modern nation-state ideology also inspired the territorial conceptualization of both Greek and Roman colonies. Salmon 1969, 15.


Again, this definition does not describe a colony as a territorial state, but it is clear from the rest of Salmon’s introductory chapter that self-government in this context entails territorial jurisdiction.23 this view is by no means exceptional. Most scholars think of colonies as places, towns, or territories. Perhaps the clearest examples of this view come from historical cartography. On most modern maps of Italy during the Middle Republic we find latin colonies depicted as territorial units confidently outlined by sharp boundaries surrounding a central dot indicating the colonial urban center (fig. 2).24 this tradition of cartographic representation goes back at least to Beloch who, we must not forget, needed colonies to have spatial dimensions because he wanted to calculate colonial population densities (fig. 3).25 However, the method used by historical cartographers to draw such maps is anachronistic and largely conjectural. this is clearly demonstrated by the fact that, more often than not, scholars disagree about the precise size and shape of colonies (fig. 4). this is not surprising considering the weak evidential basis for such territorial reconstructions, which consists of very little more than anachronistic data shored up by extensive guesswork.26 that in itself does not mean colonies definitely were not fIG. 3. Map of Central Italy in 298 B.C. (after Beloch 1926, map 2). territorial states; but the claim that they were should likewise be based on reliable data. One important argument used to support the territorial understanding of colonia is that the hypothetical territorial jurisdiction of a colonia crucially distinguishes it from other types of Roman settlement. for example, umberto laffi has argued that a colonia had territorial sovereignty by definition, since there is a specific legal term to describe a community of Roman citizens without territorial claims, namely conventus.27 this argument, however, is not as strong as it appears. Besides the general observation that it is impossible to say whether juridical categories that were definitively shaped in the late Republican and early Imperial period can be applied to the Mid-Republican period,28 this terminological argument is undermined especially

23

24

25

26

27

Salmon 1969, 14; his use of the term territorium is particularly important, because it implies that the colony controlled a single, coherent piece of land. examples are abundant, e.g., Humbert 1978; Cornell 1995, 381-82. Beloch 1880. this tradition is continued by Afzelius 1942; fraccaro 1956-57; toynbee 1965, Map 2. territorial boundaries are typically reconstructed by using medieval maps, identifying probable natural barriers, and finding traces of centuriation. laffi 1966, 111-14. A conventus civium Romanorum

28

was a permanent organization of Roman citizens in the provinces, under the aegis of a curator (Berger 1953). laffi also discusses the term consistentes (a term for persons who reside temporarily in a place that was neither their birthplace nor their domicile). even if we accept the juridical terminology of the late Republic, we can easily think of other criteria that separate a conventus and a colonia. Most importantly a colonia had an obvious military function and most probably a strong agrarian based economy: members of a colony were entitled to a piece of land; those of

ďœˇďœˇ


fIG. 4. Reconstructions of colonial territories in latium, based on: A) Beloch 1926, map 2; B) toynbee 1965, map 2; C) Cornell 1995, 381; D) Bouma and Van ‘t lindenhout 1996-97, 94.

by the fact that most scholars, for good reasons, now accept that the small colonia civium Romanorum of the Mid-Republican period was not a territorial sovereign state, because it lacked proper civil constitutional structures and moreover was located on Roman territory (cf. Sisani in this volume).29 It is also clear especially from livy’s and Dionysius’ description of Regal and early Republican colonies that, according to their

a conventus were not. Arguably, a colonia was also an official state foundation, whereas a conventus was an unofficial settlement (See RE 6 (1901) 1173-200, s.v. “conventus” (e. Kornemann) and also Sherwin-White 1973, 225-32). Similar criteria also easily differentiate coloniae from viritane settlements. the latter were not military in nature and, perhaps more importantly, did not have a fixed number of settlers.



29

Rudolph, 1965, 130, Sherwin-White 1973, 84-85. In the case of Roman colonies, Sherwin-White convincingly argues that they were primarily military settlements with very restricted local government, probably only in the military sphere, and that this changed only after 183 B.C., when the first steps were taken toward the assimilation of the colonia to the status of municipium.


conceptual framework, a colonia could easily be founded inside the territory of another community and that colonists and natives intermingled with one another (cf. tarpin in this volume).30 this view is supported also by definitions of Roman colonies in latter sources. Both Servius and Siculus flaccus define colonies as groups of people who are led out together to a certain place, which could well be an already existing and populated town.31 We must therefore conclude that territorial sovereignty cannot simply be deduced from the term colonia. But what then about latin colonies? Does latin status imply territorial sovereignty (on latin status see tarpin in this volume)? to many scholars it does, because the larger size of these colonial settlements required a larger apparatus for local government and consequently greater autonomy. Above all, the fact that latin colonists lacked Roman citizenship is thought to indicate they must have had a territory of their own. Within the conceptual framework of city-state organization, this indeed seems logical, but, as will be argued below, self-government is also possible without exclusive territorial sovereignty.32 While it is clear that coloni acquired land of their own,33 it is much less certain that the colonia itself was a territorial entity. the way livy describes colonial foundations actually argues against such an interpretation: coloniae are usually ‘brought’ or ‘sent’ to a place (deducere, mittere), which suggests that they themselves were not places but mobile entities.34 On this evidence, we should conclude that a colonia is in the first place a body of coloni, or to use the terminology of ulrich Kahrstedt a Personalgemeinde rather than a Territorialgemeinde.35

Colonial territories the crucial question thus is whether a colonia as a socio-political entity also possessed exclusive jurisdiction over a spatial extent.36 In other words, did it have a clearly defined territory of its own? As tesse Stek cogently argues in another chapter of this book, there is no convincing archaeological evidence in support of this view. there are no boundary stones known from colonial territories, and the spatial configuration of the sacred landscape of Mid-Republican Roman colonies reveals no effort to define and

30

31

e.g., Sherwin-White 1973, 80 note 4; levick 1967, 69. A good example of this is Circeii. According to a passage in Dionysius (8.14) which describes the siege of Circeii by Coriolanus, at that time leader of the Volscian forces, he “came to the city of Circeii, in which there were Roman colonists living intermingled with the native residents, with his army; and he took possession of the town as soon as he appeared before it”. Sic. flacc., Cond. agr. 99.1l; Serv., Aen. 1.12. Both sources give definitions of the meaning of colonia which are consistent with the view that the term refers to a gathering of men who are led to a specific and sometimes already settled place. the fact that Servius mentions that a colony is what is called ἀποικία in Greek, is perhaps closest to the conventional understanding of a colony as a city-state. However, most scholars of Greek colonial history now agree that ἀποικία can refer to different settlement realities, not necessarily city-states (cf. tsetskhladze 1998, 15-17). It is true that Varro (Ling. 5.143) sees colonies as urbes

32

33

34

35 36

since they are founded in the same way as Rome (thus, that the city limits had been drawn with the plough). See the contribution of Sisani in this volume for an excellent discussion of this passage and on colonies as urbes in general. Nothing refers in this context to these urban centres having territories. See on the ambiguous and changing definitions of colonia in Roman sources Bispham 2006, 83-85. According to modern state theory, sovereignty implies having exclusive jurisdiction over a defined territory (cf. Holland 2010, 450-51). Interestingly, however, the first attestation of latin colonists receiving plots of land dates to the early 2nd century (liv. 35.9 and 40). On this Pelgrom 2008, 358-61. eg. liv. 8.14.7-8; liv. 8.16.13; liv. 8.22.2. In a few exceptional cases more ambiguous terminology is used: imponere in liv. 8.23.6 or in coloniam missi in liv. 8.21.11. Kahrstedt 1959, 206. for the importance of exclusive jurisdiction for state sovereignty see Holland 2010.




protect territorial boundaries in the way postulated for Greek and etruscan communities. the hypothetical existence of latin colonial territories thus must be deduced from literary and epigraphic sources. An important argument in favor of the presumed existence of latin colonial territories derives from the fact that we occasionally encounter terms in the sources that seem to indicate them, such as ager Calenus.37 Such terminology, although known only from late sources, seems to tie in nicely with the view that latin colonies had territorial jurisdiction over a spatial area (the ager). However, a closer reading of the available texts suggests that the situation is more complex and that ager more likely refers to a conquered territory over which Rome retained some form of control.38 An important passage in this discussion is livy’s description of the fate of the disloyal Capuans after 211. According to livy, the plebiscite that decided the fate of the Capuan elite specified that those who had been deported beyond the tiber were forbidden to acquire or to hold either for themselves or their posterity landed property anywhere except in the territories of Veii, Sutrium and Nepet and in no instance was such a holding to exceed fifty iugera (liv. 26.34).39 the passage seems to suggest that there was land in the territories of Nepet and Sutrium (both latin colonies founded in 382 B.C.) that the Capuans could acquire. Although this passage mentions the existence of an ager of Sutrium and Nepet, we cannot conclude that the latin colonies sent to these former faliscan places had exclusive territorial control over these lands. the fact that the Roman senate directly decided over these lands and regulated the amount of land the Capuans could occupy seems rather to imply that the land in question was ager publicus populi Romani located within the areas originally inhabited by the (pre-Roman) communities of Nepet and Sutrium.40 the situation in thurii-Copia (founded in 193 B.C.) is perhaps comparable. livy mentions that more land was available than was distributed to the colonists in the confiscated territory of this former Greek polis.41 this might suggest that the colony had jurisdiction over a territory larger than the sum of land divided among the individual colonists, and thus that the colony had jurisdiction over the remaining land.42 However, the fact that the land was reserved for future distribution, on the contrary, makes it obvious that it remained the property of the Roman state. Although it is possible that the colony had some jurisdiction over these lands, this right cannot be deduced from the passage. What is more, the recorded enrollment procedure for supplements of new colonists also supports an alternative reading of the passage. In the period after the Second Punic War, several latin colonies received new supplements of colonists to counter the demographic blow the war had inflicted on their communities.43 Interestingly, the latin colonies suffering population decline apparently were forbidden to enroll new citizens on their own initiative, but had to ask Rome for permission. We know from inscriptions that these supplements were led by Roman magistrates, who probably also provided the new colonists with land (cf. Sewell in this volume, who discusses Roman interferences in the case of Aquileia). this strongly suggests that undistributed land in these colonized territories were in fact formally the property of Rome, which remained the competent authority to distribute it to new colonists. that the Roman people was the real owner of undistributed land in various colonized territories is

37

38 39 40



Cf. liv. 26, 13, 10. Cales was a latin colony founded in 334 B.C. See also Roselaar 2010, 142-44 (with further references). lat: in Ueiente Sutrino Nepesinoue agro. Arguably, the correct technical term for these lands would be ager occupatorius. On this type of land see Roselaar 2010, 89, who defines this category of land as: “land owned by the state, but which the state did not use for the moment. this land could be used by anyone

41 42

43

– Roman citizens […], latins and Italians as well – who wanted to work it. It could only be held without a legal title, and the state could, at least in theory, take the land away from the occupier whenever it was needed”. liv. 35.9. On this passage, see Gargola 1995, 89. Cf. also Roselaar 2010, 50. Recorded examples are: Venusia, Narnia, Cosa, and Cales.


further supported by several controversial passages in the Liber coloniarum that mention the existence of limites Gracchani in the territories of Cales and Venusia (colonized respectively in 334 and 291 B.C.).44 It is possible, as some scholars have argued, that these limites Gracchani were not really land-division schemes drawn up in the late 2nd c. B.C., but rather were later systems that adopted a model similar to that used by the Gracchi.45 However, not only have Gracchan boundary stones been found in several territories presumed to have belonged to latin colonies, but we also know of a land division ordered by a lex Sempronia in the territory of Suessa Aurunca and indirectly also for luceria.46 these references are much harder to explain as reflecting a Gracchan model rather than an actual land reform ordered by the Gracchi. line 31 of the lex agraria (dated 111 B.C.) may in fact also support the view that Roman public lands existed inside areas where latin colonies were founded. the passage reads: {—- to whichever colonies or} municipia, {or} any equivalent of municipia or colonies {(there may be), of Roman citizens} or of the latin name, land {has been} granted by the people or by a decree of the Senate to exploit, {which land those colonies or those municipia or any} equivalent of a colony or municipium or of municipia (there may be) shall exploit {…}.47

the statement that latin colonies were allowed to exploit ager publicus suggests that the inhabitants of some latin colonies had a right to make use of land that formally belonged to Rome, perhaps the undistributed land in the confiscated territories they were sent to. Based on this evidence it can be concluded that ager in a Mid-Republican Roman colonial context, most likely refers to an area conquered by Rome where colonists were allowed to settle, and not to a territory under the exclusive jurisdiction of a colony.48

A non-territorial understanding of Roman colonies If the notion of territorial sovereignty implies that a state has exclusive jurisdiction over the people, resources, and all authorities within the territory under its control, we must conclude that this definition is contradicted by what archaeological, literary, and epigraphic sources tell us about Mid-Republican Roman colonial organization. the available evidence seems to suggest that Rome had supreme jurisdiction over colonized land, at least the land that had not been assigned to individual colonists. Such an understanding of Roman colonial organization does not conform to the principles of the ideal city-state as an autonomous polity, but rather corresponds more closely to different models of territorial expansion and colonization known to the Hellenistic world.49 Apart from assumed analogies between Roman and Classical Greek forms of civic organization, the traditional view that a colony was a territorial sovereign state also relies on the argument that Rome could not have controlled such distant lands and accordingly solved this problem by cleverly creating new independent but loyal states to control these lands on its behalf. this argument would make some sense if we accept a primitivistic understanding of Roman imperialism. However, besides the fact that the primitivistic view has been challenged cogently,50 the assumption that Rome could not exert direct control over these

44

45 46 47 48

for example: ager Benusinus, Comsinus limitibus Graccanis (Lib. colon. (C. 164, 25)). e.g., Campbell 2000, 403. See Roselaar 2009 with further references. text: RS I, lex agraria, line 31; translation, RS I, p. 145. Also liv. 45.13 on the conflict between Pisa and luni can be explained in this way. Apparently, it was not entirely clear what lands could be settled by the colonists.

49

50

Rome was asked to define more clearly the area where colonists were allowed to settle. On the use of ager in the late Republican period see Hermon in this volume. See also Sewell in this volume for Macedonian influence on the Roman colonization program, and terrenato in this volume on alternative Italian models of expansion and colonization. e.g., Coarelli 1992.




remote areas does not necessarily lead to the conclusion that Rome renounced all its territorial claims and surrendered all its newly conquered lands into the full possession of a new territorial state of colonial farmers. One could easily think of different scenarios in which Rome could effectively control these territories indirectly, for example, by merely granting loyal communities the right to use sections of the land, which formally remained under Roman power. Obviously, coloniae were communities expected to be loyal, but we may also assume that Rome granted similar rights to indigenous communities living within the boundaries of conquered territory, which were expected to protect Roman interests in the region.51 for some time the conventional view among legal historians was that Roman law excluded such models of coexistence and that indigenous populations living on colonized territories fell under colonial jurisdiction as second-class citizens without many legal and political rights.52 this view is rooted in the more general conviction that in the Classical World, in contrast to Medieval societies for example, citizenship was closely related to a place.53 However, several scholars have recently reopened the debate and have argued that there is evidence, above all in the writings of the Roman land surveyors, of the existence of separate indigenous and colonial communities living in a single territory.54 for obvious reasons, the acceptance of such an arrangement, also known as the double-community scenario, further complicates the view that a colony is a territorial state. If colonists shared a territory with another community that legally and politically was not part of the colony, this would seriously undermine the territorial sovereignty of the colony. the debate on the status of indigenous populations in Roman colonial contexts is complex and highly dynamic and cannot be discussed exhaustively in the present chapter (cf. tarpin in this volume). the point to stress here is that recent legal studies propose that Roman law did not necessarily oppose a legal construction in which two politically separate communities, one consisting of colonists and the other of natives, shared a single territory; and, in consequence, the basis for membership in a colonial community may not have been defined principally according to territorial criteria. Such a legal situation would match the scenario sketched above in which Rome retained supreme control over colonized territories and in which the juridical position of the indigenous population was negotiated directly with Rome without the intervention of the colonists.55

51

52

53



there is ample literary, epigraphic, and archaeological evidence for indigenous presence in Roman colonies (e.g. Bradley 2006; Roselaar 2010, 71-84). e.g. laffi 1966, 111 and 76-83 on incolae also Brunt 1971, 254 and 538-44; lo Cascio 1996; Hermon in this volume. Before the 1970s, a double-community scenario was widely accepted (for Republican Italy, see especially Kahrstedt 1959). for the view that indigenous inhabitants fully integrated in the colony, see Bradley 2006. for an old discussion of this view, in a comparative perspective: see the lectures of Albert Dicey (Allison 2013, esp. 192-214) who observes that it is not so much living within the boundaries of a certain territory which was significant in Antiquity, but rather being a member of a particular town and ‘sharing the common labors’ that gave them the relevant criteria for citizenship. Cf. also tarpin in this volume, esp. the example of the complex juridical status of the captive colonists from Cremona and Placentia. However, the fact that one apparently lost citizenship of a colony if captured and brought to

54 55

the enemy’s praesidia does not necessarily leads to the conclusion that citizenship was defined by territorial criteria. the loss of citizenship could also be the result of a person’s inability to perform certain civic duties that require one’s presence in the colonial town (such as certain important rituals, or being present at the local census). for the importance of the colonial oppidum for membership in a colony in the late Republican and early Imperial periods, see Sisani in this volume. Gagliardi 2006, 160-75; Bispham 2007, 445-51. In late Republican times, some indigenous civitates were placed under the government of a neighboring colony. this arrangement was called a civitas adtributa. the adtributi had to pay the dominant community for the use of the land, which in a formal sense, was the property of the Roman people and not of the neighboring colony. No clear evidence of such a socio-political construction in the Mid-Republican period exists and most scholars agree that this system was introduced only in the late Republican period and was geograph-


the conclusion that a colony, in theory, was not legally defined primarily as a territorial entity obviously does not exclude the possibility that colonial settlers lived in close proximity to one another and claimed a coherent section of the conquered territory as their own. especially in the post-Hannibalic period, when the sources begin to report the distribution of equal-sized allotments and there is convincing evidence of the existence of rigid land-division systems, there is a strong possibility that colonies acquired clear territorial dimensions. On the other hand, there is little evidence to support the view that such an arrangement was standard practice in the period before the Punic Wars.56 Several literary sources clearly describe situations in which colonists and indigenous people intermingled with one another.57 Although it is difficult to assess the authenticity of these stories, it should at least be accepted the possibility that socio-political and ethnic landscapes were patchier than previously assumed; that there were potentially no clear-cut topographical boundaries that unambiguously indicated colonial and native living spaces. Whatever the precise situation might have been in the early colonial period, it is likely that with the passing of time ties between colonists and natives living closely together became stronger and interaction among them intensified. this could have blurred the original socio-political and juridical divisions that separated them. In some cases, this fusion could have resulted in a decision to unite the communities formally, as is recorded to have happened in the late case of taras-Neptunia in the late second or early first century.58 the reorganization of Italy into new municipal districts after the Social War offered an especially good opportunity to unite scattered colonial and peregrine communities that had lived alongside one another for generations. Perhaps it was only at this moment that the patchwork of territorially defined municipal communities that is so familiar to us now was created.

ically limited to the Alpine regions (cf. laffi 1966, 9091). A possible early example, dating to before the Social War, comes from the Sententia Minuciorum dated 117 B.C. However, according to laffi, strictu sensu this example is not a form of adtributio because, at the time, Genua was not a community with latin or Roman rights, but a civitas foederata (laffi 1966, 61, 90 and 95). Also Galsterer 1976, 53 n. 83, argues that the system

56 57 58

cannot be used to define relationship between natives in colonies in the Mid-Republican period. However Brunt (Brunt 1971, 541) states that the system was developed after the Social War and claims that it might well have had precedents in the south. Pelgrom 2008, 358-68. See discussion above. On this case, see laffi 1966, 109-17; id. 2004.




References Afzelius, A. 1942. Die römische Eroberung Italiens (340-264 v. Chr) (Copenhagen). Allison, J. W. f. (ed.) 2013. Comparative constitutionalism, by A. V. Dicey (Oxford). Ando, C. 1999. “Was Rome a Polis,” Classical Antiquity 18-1, 5-34. Beloch, K. J. 1880. Der italische Bund unter Roms Hegemonie. Staatsrechtliche und statistische Forschungen (leipzig). Beloch, K. J. 1926. Römische Geschichte bis zum Beginn der punischen Kriege (Berlin/leipzig). Berger, A. 1953. Encyclopedic dictionary of Roman law (Philadelphia). Bispham, e. 2006. “Coloniam Deducere: How Roman was Roman Colonization during the Middle Republic?,” in G. J. Bradley and J.-P. Wilson (edd.), Greek and Roman Colonisation: Origins, Ideologies and Interactions (Swansea) 73-160. Bispham, e. 2007. from Asculum to Actium: the municipalization of Italy from the Social War to Augustus (Oxford). Bradley, G. 2006. “Colonization and identity in Republican Italy,” in G. Bradley and J.-P. Wilson (edd.), Greek and Roman colonisation. Origins, ideology and interactions (Swansea) 161-87. Broadhead, W. 2001. “Rome’s migration policy and the so-called ius migrandi,” Cahiers Glotz 12, 69-89. Broadhead, W. 2008. “Migration and hegemony: fixity and mobility in second-century Italy,” in l. de ligt and S. J. Northwood (edd.), People, Land and Politics. Demographic Developments and the Transformation of Roman Italy 300 BC-AD 14 (leiden) 451-70. Bouma, J. W. and e. Van ‘t lindenhout 1996-97. “light in Dark Age latium. evidence from settlements and cult places,” Caeculus 3, 91-102. Brown, f. 1980. Cosa: the making of a Roman town (Ann Arbor, MI). Brunt, P. A. 1971. Italian manpower. 225 B.C.-A.D. 14 (Oxford). Cahill, N. 2002. Household and city organization at Olynthus (New Haven, Ct). Campbell, B. 2000. the writings of the Roman land surveyors. Introduction, text, translation and commentary (london). Carandini, A., f. Cambi, M.-G. Celuzza, and e. fen-



tress. 2002. Paesaggi d’Etruria tra la Fiora e l’Albegna, valle d’Oro, valle del Chiarone, valle del Tafone (Roma). Carter, J. C. 2006. Discovering the Greek countryside at Metaponto (Ann Arbor). Coarelli, f. 1992. “Colonizzazione e municipalizzazione: tempi e modi,” DiaArch 3-10, 21-30. Coarelli, f. 2005. “Pits and fora: a reply to Henrik Mouritsen,” PBSR 73, 23-31. Cornell, t. J. 1995. The beginnings of Rome. Italy and Rome from the Bronze Age to the Punic Wars (c. 1000-264 BC) (london). Crawford, M. H. 1995. “la storia della colonizzazione romana secondo i romani,” in A. Storchi Marino (ed.) L’incidenza dell’antico (Naples) 187-92. De Ruggiero, e. 1896. Le colonie dei romani (Spoleto). Dyson, S. l. 1992. Community and society in Roman Italy (Baltimore, MD/london). erdkamp, P. 2011. “Soldiers, Roman citizens, and latin colonists in mid-republican Italy,” Ancient Society 41, 109-146. fentress, e. 2000. “frank Brown, Cosa, and the idea of a Roman city,” in e. fentress (ed.), Romanization and the city: creation, transformations and failures (Portsmouth, RI) 11-24. fraccaro, P. 1956-57. Opuscula (Pavia). Gagliardi, l. 2006. Mobilità e integrazione delle persone nei centri cittadini romani. Aspetti giuridici. La classificazione degli incolae (Milan). Galsterer, H. 1976 Herrschaft und Verwaltung im republikanischen Italien: die Beziehungen Roms zu den italischen Gemeinden vom Latinerfrieden 338 v. Chr. bis zum Bundesgenossenkrieg 91 v. Chr. (Munich). Gargola, J. D. 1995. Land, laws, and gods. Magistrates and ceremony in the regulation of public lands in Republican Rome (Chapel Hill, NC/london). Hansen, M. H. (ed.). 2000. A comparative study of thirty city-state cultures: an investigation conducted by the Copenhagen Polis Centre (Copenhagen). Holland, B. 2010. “Sovereignty as Dominium? Reconstructing the constructivist Roman law thesis,” International Studies Quarterly 54, 449-80. Humbert, M. 1978. Municipium et civitas sine suffragio. L’organisation de la conquête jusqu’à la


Guerre Sociale (Paris). Kahrstedt, u. 1959. “Ager Publicus und Selbstverwaltung in lukanien und Bruttium,” Historia 8, 174206. laffi, u. 1966. Adtributio e contributio. Problemi del sistema politico-amministrativo dello stato romano (Pisa). laffi, u. 2004, “Osservazioni sulla lex Municipii tarentini,” RAL 9-15, 611-40. lo Cascio, e. 1996. “Pompei dalla città sannitica alla colonia sillana: le vicende istituzionali,” in M. Cébeillac-Gervasoni (ed.), Les élites municipales de l’Italie péninsulaire des Gracques à Néron (Naples) 111-23. levick, B. M. 1967. Roman colonies in Southern Asia Minor (Oxford). Marquardt, J. 1881. Römische Staatsverwaltung, vol. 1 (2nd edn., leipzig). Mertens, J. 1969. Alba Fucens, vol. 1: rapports et études (Brussels). Mouritsen, H. 2004. “Pits and politics: interpreting colonial fora in Republican Italy,” PBSR 72, 37-68. Pelgrom, J. 2008. “Settlement Organization and land Distribution in latin Colonies before the Second Punic War,” in l. de ligt and S. J. Northwood (edd.), People, Land and Politics. Demographic Developments and the Transformation of Roman Italy 300 BC-AD 14 (leiden) 333-72. Roselaar, S. 2009. “References to Gracchan Activity in the Liber Coloniarum,” Historia 58, 198-214. Roselaar, S. 2010. Public land in the Roman Republic. A social and economic history of ager publicus in

Italy, 396-89 BC. (Oxford). Rudolph, H. 1965. Stadt und Staat im Römischen Italien: Untersuchungen über die Entwicklung des Munizipalwesens in der Republikanischen Zeit (2nd edn., Göttingen). Salmon, e. t. 1953. “Rome and the latins: I,” Phoenix 7. 3, 93-104. Salmon, e. t. 1967. Samnium and the Samnites (Cambridge). Salmon, e. t. 1969. Roman colonization under the Republic (london). Salmon, e. t. 1982. The making of Roman Italy (london). Sewell, J. 2010. The formation of Roman urbanism, 338-200 B.C.: between contemporary foreign influence and Roman tradition (Portsmouth, RI). Sherwin White, A. N. 1973. The Roman citizenship (2nd edn., Oxford). Stek, t. D. 2009. Cult places and cultural change in Republican Italy. A contextual approach to religious aspects of rural society after the Roman conquest (Amsterdam). tsetskhladze, G. R. 1998. “Greek colonisation of the Black Sea area: stages, models, and native population,” in G. R. tsetskhladze (ed.) The Greek colonisation of the Black Sea area: Historical interpretation of Archaeology (Stuttgart) 9-69. toynbee, A. J. 1965. Hannibal’s legacy. The Hannibalic War’s effects on Roman life (london/New york). Wallace-Hadrill, A. 2008. Rome’s cultural revolution (Cambridge).




The city-state model and Roman Republican colonization: sacred landscapes as a proxy for colonial socio-political organization tesse D. Stek “In that world, of which the city-state was the basic unit and in which every important political act was dIivinely sanctioned, the birth of a new city-state was the most momentous.” (frank Brown 1980, 16) “extra-urban sanctuaries are particularly revealing.” (françois de Polignac 1995, 33)

Introduction What did a Roman colony look like? ‘like Rome’, has long been the favorite answer. A standard reply to this question would at any rate evoke the image of a colonial urban center (oppidum) boasting political, commercial, and religious services (forum, Capitolium) and a neatly partitioned hinterland producing food for the inhabitants of the colony (ager).1 the main tenet of this view is that colonies were organized as Rome was: that is, as a citystate, with a clearly defined center and a clearly defined territory (cf. Pelgrom in this volume). In light of exciting new work on Roman colonization in the early and Mid-Republican periods, however, it now seems imperative to review the evidence for this model and remain open to vastly differing scenarios when assessing the organization and appearance of Roman colonies of the latin right before the late Republican period. this paper aims to contribute to this reassessment by investigating the socio-political, and especially the territorial, make-up of Roman Republican colonial communities through an analysis of colonial cults and cult sites and their location in the colonial landscape. After reviewing the arguments in support of the standard thesis that colonies copied the city-state model supposedly set by Rome and examining the actual archaeological and epigraphic evidence for the earlier phases of Roman colonies of the latin right in Italy, it will be argued that the character and morphology of the sacred colonial landscape do not support traditional assumptions that Roman Republican colonies were city-states à la grecque, and that the available evidence instead points toward a different model of socio-political organization.

Beyond ‘Mimic’ the urgent need to put the evidence to the test springs from important new historiographical and archaeological work on Roman colonization. In particular, the ‘Gellian’ notion that colonies closely copied the political (infra)structure of the city of Rome may now confidently be considered anachronistic at least for the early phases of Roman colonization.2 Such findings remind us to be wary of the power of hindsight in historical and archaeological interpretations of earlier phases of Roman colonization and raise important

1 2

Cf. Hermon in this volume. Bispham 2000, to which the title of this paragraph alludes, and id. 2006 are seminal; cf. Pelgrom and Stek

and esp. Sewell and torelli in this volume with further bibliography.




questions regarding the political structure of Republican colonies. Cults and religious practices also played an important part in shaping Roman colonial society. According to tradition, the foundation of the city of Rome itself was deeply imbued with religious ritual. the location of the future city was supposedly divinely sanctioned by the flight of birds, whereas the outer limits of the city were ritually traced with a plough, creating the primordial furrow or sulcus primigenius (cf. Sisani in this volume). In light of this prominence of ritual and religion in our sources, it is not surprising that, just as with their political structure, it has generally been expected that the religious make-up of colonies would also closely reflect the situation of the city-state of Rome. the clearest example is the Capitoline cult and Capitolium, which scholars traditionally believed were replicated in the colonies as proper, small copies of Rome. An entire research project could easily be dedicated to the Capitoline issue alone,3 but it is important to note that the literary and archaeological evidence for Republican colonies is also in this case much messier than it has been presented in earlier scholarship and is problematic, to say the least, for the early phases of Mid-Republican colonies. On closer inspection, as edward Bispham has recently shown, it is difficult to find solid evidence for Capitolia erected at the foundation of the colony. In some cases they were built much later, mostly in the late Republican or Imperial period, and sometimes not at all.4 In the meantime, the potential importance of other pre-existing or new tutelary deities of colonial communities has been explored.5 this further qualifies, or at least contextualizes, the role of the Capitoline model for colonial communities. What is at stake in this new wave of studies is not so much the precise correspondence of text-based expectations and archaeological evidence, but rather the basic character of Roman colonization and, by extension, of Roman imperialism at large in the early and Mid-Republican periods. even if this goal is not always programmatically announced, several of these groundbreaking studies have seriously undermined the ‘statist’ view of Roman colonization by deconstructing the later historiographical logic imposed on a potentially more haphazard or at least messier reality. Instead, they emphasize the diversity of colonial communities and their appearance and have developed analogies that suggest different socio-political structures at work behind the colonization movements.6

Beyond the oppidum. The importance of the colonial territorial organization Recent archaeological critiques have primarily been engaged with urban aspects of colonization, that is, with what happens within a city’s walls. this is true of studies both of the sociopolitical and of the religious character of colonial communities, such as those that test the supposed Romanness of colonies in terms of their architectural layout. However, this new wave of studies also raises the question of how colonies were organized as a whole, that is, including their extra-urban territory. A colony consisted not (only) of a centre, but of the whole community of colonists living in a certain area. In fact, including the extra-urban ‘hinterland’ in our analysis may give us important insight into the physical structure and socio-political organization of the colonial community. the question, then, that we should ask ourselves is this: what types of landscapes would we expect to correspond to different societal organizational scenarios? On the standard, city-state view of Republican colonies, a clear division between center and hinterland is expected, with colonists’

3

4



A study on the subject is in preparation by Nicholas Purcell. Bispham 2006. Importantly, the provincial situation is also not clear-cut; cf., e.g., Beard, North, and Price 1998, 328-39, and recently Quinn and Wilson 2013 arguing for the locally-determined character of Capitolia in provincial North Africa.

5 6

Boos 2011; ead. in this volume. Most explicitly, esp. Càssola 1988; torelli 1999; Crawford 1995; Bispham 2006; Bradley 2006, see also the introduction to this volume.


fIG. 1. the standard view of a colony, with urban centre and centuriated hinterland, as transmitted by illustrations to the Corpus Agrimensorum Romanorum. Miniature of the manuscript Arcerianus A, Wolfenbüttel (l. Opgenhaffen).

fIG. 2. traditional conceptualization of the Roman colonizing process: before (left image) and after (right image), (after G. Moscara in Settis 1984, 73 fig. 9 and 150 fig. 129).

farms evenly distributed over the territory. One might think, for instance, of the illustrations in the Corpus Agrimensorum Romanorum (fig. 1).7 Such a model would be in line with the long-term view that strongly links Roman conquest and imperialism to the promotion of urbanism (fig. 2). A more clustered distribution of settlements, in contrast, with villages or fortified hamlets as important

7

e.g., in Campbell 2000. Cf. tarpin in this volume for the illustrations of the Corpus Agrimensorum Romanorum.




fIG. 3. ‘leopard skin’ clustered patterns in archaic urban landscapes. Veii (A), Caere (B), Vulci (C) (after Guidi 2006, 75 fig. 1).

habitation centres, would fit socio-political scenarios that emphasize more archaic or gens-oriented forms of control. According to this pattern, smaller nuclei are found both in the rural and the supposedly ‘urban’ landscape, since the latter often seems to consist of different clusters linked to different communities (the so-called ‘leopard-skin’ pattern). for a general picture, one might think of southern etruria in the Archaic period, as documented for Veii, Caere and Vulci, for instance (fig. 3). It is clear therefore that assessing the physical and socio-political organization of the colonial community as a whole, including both hypothetical centres and the rural hinterland, may provide valuable information with respect to the recent reconsideration of Rome as a state power and prototype in Roman colonization and imperialism. Possible approaches to this question, however, are limited. the above-mentioned problem of anachronism precludes using our almost invariably later literary sources directly for information about the development of settlement organization and the territorial definition of the colonies. the epigraphic record, as I have argued elsewhere, tentatively attests a socio-political organization that seems to deviate from standard expectations in its emphasis on separate rural communities rather than on the colonial urban center or the civic community.8 Archaeological evidence for settlement patterns and organization provides another dataset for the period under discussion. evidence from field surveys should give us a general idea of the settlement organization of Roman colonies of the latin right. the recent review of settlement patterns in colonial territories by Jeremia Pelgrom, drawing on existing field survey evidence, is important in this regard. Pelgrom’s findings may indeed point to less neatly organized landscapes than anticipated for the early colonial phases9 – a pattern that tallies well with the rural socio-political organization tentatively proposed on the basis of the epigraphic evidence. the diverse character and quality of the dataset, however, does not yet allow us to draw definitive conclusions.10 Also, it should be acknowledged upfront that the correspondence of settlement morphology and state organization is an intricate question – and indeed one of the most discussed issues in the discipline of archaeology from the past to the present day. So is, however, the relationship between architectural form and state organization, which has nevertheless attracted far more attention in the current debate on Roman colonization, a fact that can probably be explained in part by the particular history and aims of Classical/Roman archaeology.

8 9 10



Stek 2008 = id. 2009, esp. 123-70. Pelgrom 2008. the recently started project landscapes of early Roman Colonization, directed by tesse D. Stek and

Jeremia Pelgrom, seeks to collect and make available a methodologically sound dataset on this matter, with fieldwork in the latin colonies of Aesernia and Venusia.


In any case, it is beyond doubt that an analysis of the rural make-up of the colony, especially the specific relations between rural communities and the colonial centre, Rome, and other reference points, may help us better understand the character of latin colonies, providing clues as to their purpose and how they functioned in practice.

Colonial sacred landscapes: different scenarios An excellent point of departure for studying precisely these issues is evidence from rural or extra-urban – in the broadest sense of the word – cult sites and cults. the underlying and, I believe, safe assumption is that religion and ritual are central to the definition of any ancient community, and that the colonial sacred landscape thus reflects the way that the colonial community conceived of its organization.11 Not only can cults and cult places give insight into the social, cultural, economical, and political reality of the community of worshippers, they are also very sensitive to ideological and socio-political changes. they are able to escape the general drawback of morphology, mentioned above, in that they sometimes indicate relations between communities and inform us about different levels of socio-political organization. Most importantly, however, cults have often been assigned very specific roles in different models of state organization. At least in theory, these roles may provide us with a powerful tool to test different scenarios for the sociopolitical organization of latin colonies. Now, what patterns in rural cults and cult places would we expect in different socio-political scenarios and how would these cults and cult places be related to the colonial center? Or, to put the question differently, how might the different ways in which colonial settlements were organized have been expressed in the sacred landscape? Did the foundation of a colony coincide with any notable changes or reformulations of the sacred landscape? On the strong, standard view, Roman colonization entailed a substantial reorganization of rural space (cf. fig. 2), and it can be assumed that such a reorganization was also expressed in terms of ritual. following the city-state scenario, a sacred landscape of the type proposed for the poleis of mainland Greece and Magna Graecia would seem natural. As is well known, it has been assumed that the presence and development of extra-urban cult places played a crucial role in the formation and territorial definition of poleis in the first place. the idea was first developed for the Greek colonies and then most famously elaborated into a comprehensive model for both the Greek colonies and the mainland by françois de Polignac.12 Rural or extra-urban cult places, in his view, demarcated the territory of the civic community; to use de Polignac’s words, they would have marked “the city’s control over the terrain.”13 Related religious rituals, in particular identical cults and rites performed in both central and peripheral cult sites, thus defined and consolidated the structure of the city-state.14 for instance, centrifugal processions to rural cult sites at the boundaries of the civic community’s land could reinforce the claim of the power center to its territory. A similar scenario has been proposed for the etruscan city-states, where frontier or border sanctuaries would have visually and ideologically marked the territory of the community.15 Such a model has also been proposed for the city-state of Rome itself: cult places located at a set distance from the center would have collectively demarcated the ancient territory of Rome, the so-called ager

11

12 13

On the value of the ‘sacred landscape’ for understanding the socio-political organization of a community, see, e.g., Vallet 1968; Nenci 1979; De Polignac 1984 = id. 1995, 33; Guzzo 1987; Alcock 1993, esp. 172, for the Greek world; edlund-Berry 1987 and Stek 2009 for the diverse pre-Roman situation in Italy. e.g., Vallet 1968; De Polignac 1995 with bibliography. Ibid., 33.

14

15

the birth of the Greek city, in de Polignac’s words, “was established through, on the one hand, the religious definition of a new representation of space – the city territory – and on the other, the elaboration of new civic community, thanks to rites of social integration” (ibid., 9). e.g., zifferero 1995; Riva and Stoddart 1996; various contributions in Cifani and Stoddart 2012.




fIG. 4. the “perimetro suburbanus di Roma primitiva” as defined by lugli (after lugli 1966, 650 fig. 2).

Romanus antiquus (fig. 4).16 Specific territorial rituals and festivals, including processions such as the Ambarvalia, would have woven these cult sites together so as to constitute the sacralized territorial definition of the Roman city-state. With respect to the colonies Rome founded, the model that posits the replication of the sacred landscape of the city of Rome is indeed the one we normally encounter in modern scholarship on the subject. for instance Daniel Gargola, in his Lands, Laws and Politics, explicitly imagines the configuration at Rome, in particular the Capitolium at the center and boundary sanctuaries demarcating the ager Romanus antiquus, as the model for the colonies, the territory of which would likewise have been demarcated ritually by cult sites on the territorial border.17 this hypothesis, widely accepted in both Anglophone and Italian scholarship,18 logically also presumes regular rituals whereby, in Gargola’s words, the “colonial priests and magistrates confirmed possession of their territory”, analogously to Rome.19 It therefore seems worthwhile first to briefly review the evidence for Rome itself before discussing the evidence for the colonies. even if, as I will argue in greater detail further below, the situation in Rome in itself is not (necessarily) indicative of that in Roman colonies, it is important to understand the entire complex of arguments that underlies the current standard view.

16 17



e.g., Scheid 1987; Colonna 1991, cf. below. Gargola 1995, 82-87.

18 19

e.g., Guidobaldi 1995. Gargola 1995, 85.


fIG. 5. Presumed border sanctuaries outside Rome. the stars indicate the discussed cult sites (after ziolkowski 2009, 79, fig. 1).

Rome: a deceptive archetype? the idea that a ring of cult sites around the city of Rome, located on its major arteries at specific distances, demarcated the ancient border of the city-state was established long ago. It is important to realize, however, that it is not directly based on ancient sources that mention frontier sanctuaries and link them to the definition of the original ager Romanus, but is a modern reconstruction based on the interpretation of a variety of literary and epigraphic evidence that in some cases can be connected to archaeological remains.20 the point of departure is a literary reference by Strabo in his Geographica. In this passage, Strabo (5.3.2) describes the extent of ancient Rome, and writes that “between the fifth and sixth stones which indicate the miles from Rome there is a place called Festoi, and this, it is explained, is a boundary of what was then the Roman territory, and the priests celebrate there and at several other places a sacrifice called Ambarouia”. this text has been instrumental in extrapolating from different kinds of evidence the existence of a set of frontier sanctuaries encircling Rome. this set of sanctuaries would correspond to the original Roman territory before expansion in the Republican period: a ‘Romulean’, ‘Numan’, or at least an archaic chronological horizon. the core set consists of the sanctuary of terminus at the sixth milestone of the via laurentina,21 a shrine of Mars at some point on the via Appia,22 the temple of fortuna Muliebris at the fourth milestone of the via latina,23 the shrine of Robigus at the fifth milestone of the via Claudia,24 and the sanctuary of Dea Dia at the fifth milestone of the via Campana,25 as well as the “site called festoi” mentioned above, for which different locations have been proposed (fig. 5).

20

21

e.g. Beloch 1880, 43-44; id. 1926, 169-70; Ashby 1927, 29-39; most explicit in formulating the concept has been Andreas Alföldi (Alföldi 1962, esp. 194-201 [“Die Sakralgrenze des ager Romanus antiquus”]; id. 1965, 296-304); further, esp. lugli 1966; taylor 1960, 75; Gjerstad 1973, 107-12; Quilici Gigli 1978; Scheid 1987; Colonna 1991. Ov. Fast. 2.679-684.

22

23

24 25

Alföldi 1962, 199 = id. 1965, 302, on the basis of liv. 22.1.12; 10.23.12; 47.4; 38.28.3. liv. 2.40.12; Val. Max. 1.8.4; festus, Gloss. Lat. 282 l. Cf. LTUR Suburbium (2004) 272-73, s.v. “fortunae Muliebris Aedes” (R. egidi). Fasti Praenestini April 25 = Inscr.It xIII, 2, 131. Scheid 1987; Scheid and Broise 1989; Scheid 1990.




Although this set of Roman frontier sanctuaries has become practically axiomatic in Roman studies, a recent study by Adam ziolkowski has seriously undermined both the evidence for each hypothetical frontier sanctuary and their relationship to the original extent of the territory of the city of Rome.26 It goes beyond the aim of this paper to discuss his arguments in detail, so I will limit myself to some comments on what is most relevant to the argument here. first, one cannot but agree with ziolkowski that there is a fair amount of circular reasoning and wishful thinking in the modern creation of Rome’s border sanctuaries: at the least the shrine of Mars should be eliminated from the list, and most ‘territorial’ interpretations of specific cult characteristics seem to have been informed by the notion that the sanctuaries were frontier sanctuaries in the first place (e.g., the discussion of the meaning of limen in the carmen recited by the fratres Arvales in the lucus Deae Diae). Indeed, in terms of method, we can observe that out of the vast universe of cult sites outside the walls of Rome, any (potential) cult site that happens to be located within an imaginary zone five to six miles from Rome – as dictated by Strabo – has virtually been predetermined as a frontier sanctuary. Some of ziolkowski’s arguments are perhaps less persuasive, particularly with respect to the cult place of Robigus. there is considerable confusion about the location of Robigus’ cult site,27 but while ziolkowski considers this confusion grounds for rejecting Verrius flaccus’ evidence for a cult site at the fifth milestone entirely, his argument is not compelling.28 Normative assumptions about the appropriate character of divinities venerated in frontier sanctuaries may also bias ziolkowski’s assessment in some cases. for instance, fortuna Muliebris and Robigus supposedly do not fit a frontier function very well because they are not associated with any military/defensive role in religious-historical terms.29 It is true that previous scholarship has insisted on precisely such a function. As Andreas Alföldi puts it, the “magisch-religiösen Bräuche dienten vor allem zur Verhinderung der feindlichen Raubzüge in das römische Agrargebiet.”30 It is not self-evident, however, that frontier sanctuaries necessarily housed (only) deities we associate with warfare or clearly demarcating functions. It follows that deconstructing the alleged martial character of these cults and cult places does not necessarily disprove their potential territorial significance. In fact, territorial conceptions can relate to a much broader range of aspects of human society, including the creation of a sense of community in contrast to “the wild beyond” and/or other communities (not necessarily in defensive terms), and more generally to fertility.31 these reservations may attenuate the perhaps excessively drastic dismissal of the (at least theoretical) demarcating/defining role of some cults and cult sites, but ziolkowski’s main point cannot be overlooked: that it is extremely improbable that the sanctuaries in question demarcated such a thing as the ‘original’ territory of the city-state of Rome from time immemorial. As a matter of fact, if one thing is clear it is the consistently late date of the evidence for the territorial facet of the cults and festivals under discussion.32 even the cult site with arguably the strongest claim to a territorial association, the shrine of terminus, loses its authority if we consider that the association of terminus with state borders seems to be a secondary and definitely late development.33 We should also remember that we know the site itself, as well as its possible association with the extent of the Roman empire, only from Ovid’s Fasti. Moreover, the fundamental text on the Ambarvalia, on which a large part of the whole territorial edifice is built, is from the pen of Strabo and may well refer to the very particular historical conditions of the late Republican and/or Augustan periods. the suggestion that the Ambarvalia should actually be interpreted as the rural counterpart of the amburbiale sacrificium, a lustration around the city walls in Rome, is therefore very appealing. the amburbiale

26 27

28

29



ziolkowski 2009, 91-130. Fasti Praenestini April 25 = Inscr.It xIII, 2, 131 apparently contradicts Ov. Fast. 4.905-908. It cannot be excluded (as previously argued, for instance, by frazer 1929, 408-10) that the Robigalia were celebrated at different spots. ziolkowski 2009, 104, 106.

30 31

32

33

Alföldi 1962, 194, following Wissowa 1912, 143-44. On this subject, see esp. De Polignac 1995, Chapter 2 with 45-60 for a detailed account of the military aspects of frontier sanctuaries in the Greek world. As is also acknowledged by Scheid 1987, 384-85, for instance. ziolkowski 2009, 119-20


sacrificium was performed in 49 B.C.,34 when it was introduced ex novo on the recommendation of – nota bene – an etruscan haruspex as a measure against Caesar’s march on Rome.35 Although it has played a very minor role in the debate, even the archaeological evidence does not take us far back in time. the excavated cult place at the lucus Deae Diae, for instance, has yielded incredibly rich material for the Imperial period, but it has not produced clear evidence that dates to the Archaic or early and Mid-Republican periods.36 In the light of the state of the evidence, we may therefore conclude that I) the sanctuaries in question probably never demarcated something called the ager Romanus antiquus; and II) they potentially never served to demarcate territory at all or III) developed such a function only later in Roman history, probably in the late Republican or Augustan period. We might think of it then in terms of the reinvention of “Romulean” Rome, perhaps intertwined with contemporary Hellenistic Greek experiences and perceptions. thus, in scholarly efforts to understand the latin colonies founded in the Mid-Republican period, Rome may have been a deceptive archetype.

The evidence from Roman colonial rural cult sites: a different pattern? It is, however, precisely the Roman situation that is paramount in the argument that the territory of latin colonies was demarcated by frontier sanctuaries. At any rate, literary evidence for the colonies, to put it delicately, is thin. It is just possible to infer the existence of such a constellation from a handful of references. these are taken for the most part from the Corpus Agrimensorum Romanorum. there, reference is made to territorial markers such as altars and wooden stakes or pali sacrificales,37 but references to actual temples are lacking. the main problem with the literary texts, however, is their late date. the relevant texts cannot securely or even probably be regarded as evidence for the era pre-dating the late Republican period.38 for reasons of anachronism outlined above, it is imperative to limit ourselves as much as possible to the evidence for the period under consideration here, the Mid-Republican period, and to focus at least on evidence pre-dating the late Republican and Augustan periods, which were so momentous in terms of the re-creation and invention of a colonial and ‘founding’ discourse (cf. many of the other papers in this volume, esp. Sisani’s). If we do this, the picture changes considerably. In fact, there seems to be no contemporary literary and archaeological evidence for the kind of territorial cult places imagined for Rome and other city-states in the early and Mid-Republican Periods. that is to say, there is no evidence prior to the 1st c. B.C. for any emphasis of colonial territorial boundaries by means of cult places or religious ritual.39 In theory, this lack of evidence may be explained in part by a lack of archaeological visibility. Generally speaking, the body of available evidence for the territorial organization of latin colonies, including extraurban cult places, is imperfect and only seldom collected in a systematic manner. Also, ritual can be difficult to trace in the archaeological record, and it is possible that ephemeral cult places (consisting only of simple altars and the like, without significant votive deposits) have not been identified or documented. However,

34 35 36

37

38

luc. 1.584-638. ziolkowski 2009, 124-26. Scheid 1987, esp. 590; Scheid and Broise 1989; Broise and Scheid 1993. e.g., Lib. colon. I.217-18l for Sutrium (but it is unclear what type of boundary); Lib. colon. II.257l on Pisaurum for a territory marked out by the course of streams, banks and rivers, by termini, and pali. Cf. next note. In the few cases where Mid-Republican colonies are

39

mentioned, the references quite probably refer to late Republican and Imperial refoundation/restructuring of these communities, as is the case for the references in the preceding note, as Gargola 1995, 220 n. 63 admits with respect to Sutrium (Augustan colony?) and Pisaurum (85: possibly refounded in the first century). On a provisional analysis, there also seem to be no clear pattern in later periods, but this is not the object of study here.




fIG. 6. Not your typical Roman frontier sanctuary: the monumental extra-urban sanctuary of Hera (tavole Palatine) at Metapontum (after G. Pugliese Carratelli 1996, 150).

even well-known latin colonies do not suggest a city-state model with frontier shrines marking the borders (see below). essentially, it appears that, overall, no attempts were made at monumental demarcation in terms of architecture, large objects, inscriptions, and so on, and that is significant in itself. Such efforts can be and, famously, have been recognized in other historical contexts.40 In the Greek world, many extra-urban sanctuaries are more monumental than their urban counterparts. Simply said, these Greek extra-urban sanctuaries are moving on an entirely different plane, in terms of monumentality (fig. 6). Confirming our suspicion that the contrast between Greek and Roman colonial situations is not just a result of a lack of archaeological detection in Roman contexts, it should be emphasized that, in general, evidence for cultic activity in areas colonized by Rome is not lacking in this period. the relatively substantial evidence for rural cult places in Roman colonial areas, however, exhibits a different pattern, and therefore it can be safely accepted that the absence of evidence for frontier sanctuaries in Roman colonies of the latin right is somehow significant. even in cases where frontier sanctuaries may already have existed in pre-Roman times, it is difficult to establish whether they continued to serve such a function. In the well-studied case of the Greek colony Poseidonia, which became the latin colony Paestum in 273 B.C., for instance, the archaic and classical cult sites of Albanella, fonte, linora, Getsemani, Acqua che bolle, and probably Agropoli were abandoned by the time the latin colony was founded. Some extra-urban sanctuaries continued to be used or were re-used in the Roman Republican period. this is the case at the Heraion at the foce del Sele, Santa Venera, Capodifiume, and at the so-called sacellum at Camping Apollo.41 However, among these cult places only the Heraion is interpreted as a territorial chora-sanctuary in usual reconstructions of the Greek colony – the other sanctuaries are, instead, located at various distances to the colonial town, from extremely close to moderately far away. Changes in the intensity of cult activity, notwithstanding its dependence on our partial archaeological knowledge, are also interesting: whereas the Heraion clearly loses importance under the Republican colony, the sacellum at Camping Apollo, about 300 m from the city walls, rises from virtually total obscurity to a thriving sanctuary or cluster of sanctuaries in the later 3rd and 2nd c. B.C. In newly colonized areas where the polis model was not already present, a particular pattern may appear

40

ďœšďœś

Besides De Polignac 1994; id. 1995, see, e.g., Sommer 2008 for some examples outside the Greek sphere.

41

torelli 1999, 49-52 with further bibliography.


fIG. 7. the colony of Ariminum with urban center and extra-urban cult sites. the dot-dash line indicates the colonial territory as reconstructed in toynbee 1965. the stars indicate cult sites and the shaded areas indicate settlement sites (R. Kalkers and B. Brouwenstijn).

to emerge. Rural cult places tend to be located within or near clustered rural settlements or other areas with high site density. these complexes of cult sites and settlements are located at varying distances to the colonial center, as the better known examples of Ariminum, Alba fucens, and fregellae show.42 the colony of Ariminum is a particularly clear example. A dense settlement cluster has been identified in the area called Covignano, about 3 km southwest of the ancient center of Ariminum (fig. 7).43 In this

42

Although similar situations can be suggested for other colonial territories, the evidence is generally too thin or chronologically too broad to draw firm conclusions. One example is the ager Praetutianus, where the colonies of Hatria and Castrum Novum were founded in the 290s B.C. and a praefectura was established at Interamna Praetutiorum. Although the territorial definitions of these entities remain unclear to modern interpreters, field research in the area indicates that cult places were preponderantly associated with rural settlements. examples that may have fallen within the territory of the colony of Hatria include a rural cult site administered by a local village community at Cellino Vecchio (CIL I² 1898; the settlement has been recog-

43

nized on the basis of black gloss ceramics: Guidobaldi 1995, 272), a similar situation at Vico-Ornano (ibid., 273), and the excavated site of Barisciano, Contrada S. Rustico. Whereas the first two yield only clearly datable evidence from the late Republican period, at S. Rustico a village and a temple have been uncovered and the earliest material from them may date back to the 3rd c. B.C. (Stek 2009, 149 with n. 167). the only cult place where there is as yet no indication of an associated local rural community is the temple found at Colle S. Giorgio, dated to the late Hellenistic period (Iaculli 1993), but the evidence does not seem to support an interpretation of it as a colonial frontier sanctuary. zuffa 1970; fontemaggi and Piolanti 1995.

ďœšďœˇ


area, a substantial cult place that was in use from at least the 5th c. B.C. appears to have been revitalized, or even monumentalized, after the founding of the colony of Ariminum in 268 B.C. Although the Covignano area has a longer settlement history, beginning in pre-Roman times44 with a floruit in the early Imperial period, there is a well-defined phase that corresponds to the beginnings of the Mid-Republican colony. this phase is documented, amongst other things, by black gloss pottery produced at Ariminum itself.45 there is also ample evidence for cult activity in the area, although precise locations cannot always be determined. for instance, column drums that were found reused in a parish church and Italic-Corinthian capitals attest the existence of at least two different monumental structures.46 the most significant evidence in the context of this paper, however, is the remarkable marble statues that were discovered in 1890 in this area, more precisely near the Villa Ruffi.47 One statue is clearly identifiable as Minerva with aegis and helmet, whereas the other one possibly represents fortuna. As enzo lippolis has shown in a careful study of the sculptural development of Northern Italy, the iconography of the statues and especially their eclectic style, combining Greek and Central Italian elements, should date them to the 2nd half of the 3rd c. B.C.48 this coincides neatly with the arrival of the first colonists at Ariminum. Apparently, a substantial settlement site thrived during the earliest phase of the colony, and the related cult site was adorned with state-of-the-art sculpture that evoked at least one, and perhaps two, of Rome’s hallmark deities at the time (cf. below on the popularity of divine qualities in this period). A similar case in which a cult site is associated with a substantial rural settlement occurs near S. lorenzo in Strada, about 9 km southeast of Ariminum along the via flaminia, which duly curved away so as to pass by the clearly pre-existing site.49 the published architectonic terracottas retrieved here have been dated to the Mid- to late Republican period.50 the sanctuary was located near the settlement later referred to as the vicus Popilius. Alba fucens (modern Abruzzo), founded in the midst of the Apennine mountains in 303 B.C., is a special case. At present, no sanctuaries have been identified archaeologically within the borders of the colonial territory as they have traditionally been reconstructed.51 However, several cult sites related to vicus-type settlements are known in the area from latin inscriptions found on the shores of the fucine lake (fig. 8). the first cult site and settlement pair is attested by inscriptions linking a vicus Aninus to a cult site where the divine virtue or ‘quality’ Valetudo was venerated.52 the inscriptions date to the Republican and early Imperial periods, the earliest at least to the 2nd c. B.C.53 Remains of building material, including limestone slabs, capitals, and column drums found at Castelluccio, lecce nei Marsi, may be associated with the cult and settlement site.54 the second settlement closely associated with a cult site is documented by a bronze sheet bearing a statue dedication found near the fucine lake in 1878.55 the inscription mentions the involvement of two magistrates from the vicus Petinus 56 and can be dated to the late 3rd c. B.C. the third combination of settlement and sanctuary comes from the vicus Supinum, again on the shores of the fucine lake. An inscription dated to the late 3rd or the early 2nd c. B.C. mentions the dedication of a statue of Victoria on behalf of the vecos Supinas.57 Archaeological

44 45 46 47 48 49

50

51



Cf. Cristofani 1995. fontemaggi and Piolanti 1995, esp. 542-45. Marini Calvani 2000. Brizio 1890. lippolis 2000, 251-52. Ghirotti 2007a, 94; id. 2007b; fontemaggi and Piolanti 1995, 538-40. Rebecchi 1998, 191. Newly found terracotta’s are to be published. As for instance visualized by toynbee 1965.

52

53 54 55 56 57

Aninus vecus / Valetudn[e] / donum / dant: CIL Ix 3813 (= CIL I² 391) = Ve. 228 = letta and D’Amato 1975 no. 111; AÉpigr 1978, 286 = AÉpigr 1996, 513; CIL Ix 3812 (= CIL I² 390); letta 2001, 151. CIL Ix 3812 (= CIL I² 390): letta 1997, 332. Grossi 1988, 120 n. 44. letta and D’Amato 1975 no. 188 = CIL I² 2874. Discussion in Stek 2009, 155-56. letta and D’Amato 1975 no. 128 = CIL Ix 3849 (= CIL I² 388).


fIG. 8. the colony of Alba fucens with urban center and extra-urban cult sites. the dot-dash line indicates the colonial territory as reconstructed in toynbee 1965. the stars indicate the discussed cult sites and the dots indicate the epigraphically (and/or archaeologically) attested rural settlements sites (R. Kalkers and B. Brouwenstijn).

remains found in the modern center of trasacco, which include a column drum and a capital, can be related to this complex.58 A fourth combination of a village site and evidence for cult activity has been recognized at Spineto, Colle Mariano, about 2.5 km south of the vicus Supinum. the site, attested by early black gloss pottery, column bases, and the remains of a podium, yielded an inscription with a dedication to Hercules, dated to the late 3rd or the beginning of the 2nd c. B.C.59 the status of these village-cult communities and their relationship to the colony is an issue I have discussed in detail in another context. Although such communities have until recently been interpreted as ‘precociously romanized’ indigenous villages, we may now abandon this view in light of new insights into the institutional role of the vicus in Italy and abroad,60 combined with and supported by analysis of the archaeological evidence and the character of the venerated deities.61 We will come back to this point further below. Here it will suffice to emphasize that, morphologically, the fucine situation fits the image encountered for the other colonial territories for which we have reasonable evidence of their territorial organization and rural cult sites. Moving to Campania, in the territory of the colony of fregellae (founded 328 B.C.), the most sizeable

58 59

letta and D’Amato 1975, 205. Grossi 1988, 113 n. 26.

60 61

tarpin 2002. Stek 2009, Chapter 7.




fIG. 9. the colony of fregellae with urban center and extra-urban cult sites. the dot-dash line indicates the colonial territory as reconstructed in Coarelli and Monti 1998. the stars indicate cult sites and the shaded areas indicate settlement sites (R. Kalkers and B. Brouwenstijn).

sanctuary known is a suburban rather than a rural one, at the site of the future temple of Aesculapius (fig. 9, number 3 on the map). Whereas the monumental temple is dated to the first half of the 2nd c. B.C., cult activity at the site in all likelihood began when the colony was founded at the end of the 4th c. B.C., as is attested by votive material including black gloss ceramics.62 Another probable cult site – though as yet undated – is also located in the immediate vicinity of the colonial town, directly to the north (fig. 9, number 2 on the map).63 the only substantial, truly rural sanctuary in the colonial territory is the cult site indicated, amongst other things, by a Doric capital at località fosso del Medico (fig. 9, number 1 on the map). the site has produced votive material and other finds that allow us to date the active period of the sanctuary from the 3rd to 1st c. B.C. As with the latin colonies discussed above, a relationship with a rural settlement is attested in this case, too, since field surveys have revealed a large contemporary site next to the sanctuary.64 the question now is how we should interpret this particular pattern, in which cult sites and settlement are closely associated at varying distances to the oppidum, and what it might tell us about the organization of colonial communities of latin status in the Mid-Republican period. As we have seen, the available evidence is not compatible with the expected model of frontier sanctuaries based on the territorial city-state model

62



Bouma 1996, 38-39 with bibliography. the material dates almost exclusively to the late 4th to early 2nd c. B.C. for the sanctuary of Aesculapius see Coarelli 1986; it was most probably suburban (e.g. Crawford and Keppie 1984; Coarelli 1998, 62 with n. 432), and

63 64

in any case located within the periphery of the town. Monti 1998, 108 with previous bibliography. Ibid., 103, site 77 with previous bibliography; the contemporaneous settlement site is no. 78.


or the replica of Rome scenario. there is no evidence for the creation of sanctuaries at regular distances from the colonial center to carve out its territory. Instead, the available evidence seems to point to a different distribution of cult sites over the territory at varied distances, where they are notably associated with rural settlements. In terms of methodology, it is unlikely that this pattern is the result of biased or systematically partial archaeological visibility or documentation. On the contrary: as opposed to the more monumental and more readily recognizable, not to mention more highly esteemed, evidence of cult sites, rural settlement sites are usually poorly identified and underrepresented in archaeological studies. this means that we should rather expect a bias towards seemingly isolated cult sites and against rural settlements. It is therefore likely that further archaeological research will confirm rather than contradict the pattern outlined above. With respect to the historical interpretation of this pattern, the general association of cult places with settlement nucleation in the first place helps us better understand the character of such rural settlements. As noted above, archaeological knowledge of clustered sites or villages is generally limited. Already the fact, though, that many of such nucleated settlements could maintain their own sanctuaries, apparently even monumental sanctuaries, is of considerable significance in socio-cultural terms, since it demonstrates that such settlements in colonial areas were not just clusters of barns or farms for town-based people, but rather constituted communities that operated with a certain degree of autonomy in terms of religious and related socio-economic services and their community identity. An important question concerns the status of such communities and their inhabitants. Archaeologically, it is virtually impossible to tell what their ethnic or juridical character was. However, I should like to emphasize that the cults known from these communities point to integration in or strong compatibility with Roman colonial pantheons. this not only accounts for the Covignano area in Ariminum, where the cult statues of Minerva and probably fortuna of the first phase of colony are telling, but also for communities that have often been excluded from the colonial picture, such as the vici at the fucine lake near Alba fucens, which until recently were interpreted as indigenous villages. the cults of Victoria, Valetudo, and Apollo associated with these settlements have usually been interpreted as local indigenous deities or local adoptions,65 but this interpretation seems to be biased by the above-mentioned traditional ‘indigenous’ interpretation of vici in general. It is more reasonable to interpret Victoria and Valetudo as new Roman cults that were introduced in the area. this type of ‘divine qualities’ or personifications is typical of Roman society in the 3rd to 2nd c. B.C. and seems to have had a particular appeal in colonial areas.66 evidence is scarce for ritual practices and potential links between them and settlement organization and different types of territoriality. epigraphic and archaeological evidence of the cult sites discussed above does not permit us to establish a direct hierarchical relationship with the presumed colonial center. Also, there seems to be no evidence for processions or boundary rituals that could be associated with the city-state model, as is documented, for instance, in the Greek world. However, a scenario in which local rural communities played an important role in the colonial settlement organization does not preclude the existence of ritual and religious ties between colonial foundations and rural communities. One of the few glimpses of the ways in which ritual bonds could have been constructed in the first phase of colonization is given by the colony of Ariminum. Many fragments of so-called pocola deorum were found in the city center.67 Many of them mention typically Roman deities, but some also mention pagi (rural districts) and vici. Apparently, both pagi and vici dedicated black gloss cups in the colonial city center. this seems to suggest that these paterae reflect ritual actions that were performed to connect these rural communities to the colonial center.68 Clearly, there is no reason to think that

65 66

67 68

e.g., letta, cited works. Victoria, for instance, received a temple on the Palatine in 294 B.C. by l. Postumius Megellus, whereas Salus received a cult place on the Quirinal in 302 B.C. by C. Junius Bubulcus. for divine qualities, see Clark 2007. franchi De Bellis 1995. there is some discussion of the interpretation of CIL

I² 2897a, reading pagi. Fid[, where Fid[ may be understood as a reference to yet another divine personification, Fides, or as the proper name of the pagus, which would emphasize its own distinctiveness. On the possibly related ritual, which may have had a territorial dimension – for the pagi, not the colony as a whole – cf. Stek 2009, 138-45.




the ritual aimed at geopolitical demarcation, or that territorial claims made by the colonial oppidum over its ager by means of religious ritual in frontier sanctuaries were at stake here. Rather, the inverse seems to be true: rural resident communities from the hinterland come, at least symbolically, to the center.

Conclusion At this point, I would like to briefly return to consider the basis on which frontier sanctuaries in Roman colonies of the latin right were assumed in the first place. the analogy with Rome has obviously played a major role in this, just as it has in reconstructions of the urban appearance of colonies. We should first ask to what extent we need to assume great similarity between the societal organization of Rome and that of the colonies. Recent work has harshly criticized the replica thesis for the urban center as far as the earlier phases of latin colonies are concerned. this raises the fundamental question of whether we should actually accept a priori that the entire socio-political and related settlement organization will have corresponded to the model set by Rome.69 Also, new studies downplay the direct interference of the Roman state in the early phases of colonization in general, which may further strengthen our suspicions that such similarity between Rome and colony cannot be taken for granted. Second, we now can reconsider the imagined, ideal citystate model with clear, sacred territorial demarcation for the Republican period at Rome itself. the evidentiary basis for it is weak, and such a situation may have developed only later in Roman history, if at all. the hypothetical Roman model therefore cannot be considered an appropriate basis for normative expectations about colonial realities. On the basis of the evidence we now have, we must inevitably conclude that the standard view of Roman colonies of the latin right as replicas of Rome or the classical city-state, the territorial integrity of which would be demarcated by border sanctuaries, cannot be maintained. the snippets of contemporary information we have for the Mid-Republican period suggest, on the contrary, an alternative interaction between cult place and community, in which the creation of socio-religious cohesion in local rural communities comes to the fore. these communities may certainly have been connected to the colonial oppidum in religious, ritual, and potentially other ways, but the evidence does not support the image of an urban center exerting dominance over its territory by virtue of explicit sacred boundaries. It is true that the new ‘deviant’ picture sketched here may still be blamed on a methodological failure to detect archaeological evidence (or one may seek to explain my deviant picture away with reference to local diversity and variability). But without a firm model to deviate from in the first place, the picture sketched above suggests, in my view, that we should remain open to quite different scenarios for the socio-political organization of Mid-Republican colonial communities and related models of community definition. We may need to seriously rethink the image of the founding of Roman colonies as the ‘birth of a new city-state’, to come back to the epigraph to this article by frank Brown: what really happened in the newly colonized areas may have been quite different from de Polignac’s vision of the naissance de la cité grecque.

69



Cf. also Brown 1980, 15 on the naturalness of the model being copied in the colonies, on the foundation of Cosa: “It now awaited its colonists, assembled in Rome. What of them? […] Whatever their origins, they

had known no other social or political order than that of the aristocratic city-state. In their new community they would be expected, as a matter of course, to fall into place in the old, inherent order […].”


References Alcock, S. e. 1993. Graecia capta. The landscapes of Roman Greece (Cambridge). Alföldi, A. 1962. “Ager romanus antiquus,” Hermes 90, 187-213. Alföldi, A. 1965. Early Rome and the Latins (Ann Arbor, MI). Ashby, t. 1927. The Roman Campagna in classical times (london). Beard, M., J. North and S. Price 1998. Religions of Rome (Cambridge). Beloch, K. J. 1880. Der italische Bund unter Roms Hegemonie. Staatsrechtliche und statistische Forschungen (leipzig). Beloch, K. J. 1926. Römische Geschichte bis zum Beginn der punischen Kriege (Berlin/leipzig). Bispham, e. 2000. “Mimic? A case study in early Roman colonisation,” in e. Herring and K. lomas (edd.), The emergence of state identities in Italy in the first millennium BC (london) 157-86. Bispham, e. 2006. “Coloniam deducere: how Roman was Roman colonization during the Middle Republic?,” in G. Bradley and J.-P. Wilson (edd.), Greek and Roman colonization: origins, ideologies and interactions (Swansea) 73-160. Boos, M. 2011. “In excelsissimo loco: An approach to poliadic deities in Roman colonies,” in TRAC 2010 (Oxford) 18-31. Bouma, J. W. 1996. Religio votiva: The archaeology of Latial votive religion: the 5th-3rd c. BC votive deposit south west of the main temple at “Satricum” Borgo Le Ferriere (Groningen). Bradley, G. J. 2006. “Colonization and identity in Republican Italy,” in G. J. Bradley and J.-P. Wilson (edd.), Greek and Roman Colonization: Origins, Ideologies and Interactions (Swansea) 161-87. Brizio, e. 1890. “Rimini: statuette di bronzo e sculture marmoree scoperte presso la villa Ruffi,” NSc, 208-209. Broise, H. and J. Scheid 1993. “etude d’un cas. le lucus Deae Diae à Rome,” in O. De Cazanove and J. Scheid (edd.), Les bois sacrés (Naples) 145-57. Brown, f. e. 1980. Cosa: the making of a Roman town (Ann Arbor, MI).

Campbell, J. B. 2000. The writings of the Roman land surveyors: introduction, text, translation and commentary (london). Càssola, f. 1988. “Aspetti sociali e politici della colonizzazione,” DialArch 6 (2), 5-17. Cifani, G. and S. Stoddart (edd.) 2012. Landscape, ethnicity and identity in the archaic Mediterranean area (Oxford). Clark, A. J. 2007. Divine qualities. Cult and community in Republican Rome (Oxford). Coarelli, f. (ed.) 1986. Fregellae, vol. 2: Il santuario di Esculapio (Rome). Coarelli, f. 1998. “la storia e lo scavo,” in f. Coarelli and P. G. Monti (edd.), Fregellae, vol. 1: Le fonti, la storia, il territorio (Rome) 29-69. Coarelli, f. and P. G. Monti (edd.) 1998. Fregellae, vol. 1: Le fonti, la storia, il territorio (Rome). Colonna, G. 1991. “Acqua Acetosa laurentina, l’ager Romanus antiquus e i santuari del I miglio,” ScAnt 5, 209-32. Crawford, M. H. 1995. “la storia della colonizzazione romana secondo i romani,” in A. Storchi Marino (ed.), L’incidenza dell’antico. Studi in memoria di Ettore Lepore (Naples) 187-92. Crawford, M. H. and l. Keppie 1984. “excavations at fregellae, 1978 - 1984. An interim report on the work of the British team,” PBSR 52, 21-35. Cristofani, M. 1995. “Genti e forme di popolamento in età preromana,” in A. Calbi and G. C. Susini (edd.), Pro poplo arimenese (faenza) 145-81. De Polignac, f. 1984. La naissance de la cité grecque: cultes, espace et société VIIIe-VIIe siècles avant J.-C. (Paris). De Polignac, f. 1994. “Mediation, competition, and sovereignty. the evolution of rural sanctuaries in Geometric Greece,” in S. e. Alcock and R. Osborne (edd.), Placing the gods: sanctuaries and sacred space in ancient Greece (Oxford/New york) 3-18. De Polignac, f. 1995. Cults, territory, and the origins of the Greek city-state (Chicago/london). edlund-Berry, I. e. M. 1987. The gods and the place. Location and function of sanctuaries in the countryside of Etruria and Magna Graecia (700-400 B.C.) (Stockholm).




fontemaggi, A. and O. Piolanti 1995. “Il popolamento nel territorio di Ariminum: testimonianze archeologiche,” in A. Calbi and G. C. Susini (edd.), Pro poplo arimenese (faenza) 531-61. franchi De Bellis, A. 1995. “I ‘pocola’ riminesi,” in A. Calbi and G. C. Susini (edd.), Pro poplo arimenese (faenza) 367-91. frazer, J. G. 1929. The Fasti of Ovid, vol. 3: Commentaries on books I-VI (london). Gargola, D. J. 1995. Lands, laws, and gods. Magistrates & ceremony in the regulation of public lands in Republican Rome (Chapel Hill). Ghirotti, l. 2007a. “Sentieri preistorici e sistema viario in età romana nel Riccionese,” in f. Rocchetta (ed.), Luigi Ghirotti. Una vita per l’archeologia. Raccolta degli scritti nel decennale della scomparsa (Riccione) 87-96. Ghirotti, l. 2007b. “la Via flaminia nel territorio di Riccione,” in f. Rocchetta (ed.), Luigi Ghirotti. Una vita per l’archeologia. Raccolta degli scritti nel decennale della scomparsa (Riccione) 97-98. Gjerstad, e. 1973. Early Rome, vol. 5: The written sources (lund). Grossi, G. 1988. “topografia antica del territorio del Parco Nazionale d’Abruzzo (III sec. a.C. - VI sec. d.C.),” in Il territorio del Parco nazionale d’Abruzzo nell’antichità (Civitella Alfedena) 111-35. Guidi, A. 2006. “the archaeology of the early state in Italy,” Social Evolution and History 2, 55-90. Guidobaldi, M. P. 1995. La romanizzazione dell’ager Praetutianus (secoli III-I a.C.) (Naples). Guzzo, P. G. 1987. “Schema per la categoria interpretativa del ‘santuario di frontiera’,” ScAnt 1, 373-79. Iaculli, G. 1993. Il tempio italico di Colle S. Giorgio (Castiglione Messer Raimondo) (Penne). letta, C. 1997. “I culti di Vesuna e di Valetudo tra umbria e Marsica,” in G. Bonamente and f. Coarelli (edd.), Assisi e gli Umbri nell’antichità (Assisi) 317-39. letta, C. 2001. “un lago e il suo popolo,” in A. Campanelli (ed.), Il tesoro del lago. L’archeologia del Fucino e la collezione Torlonia (Pescara) 139-55. letta, C. and S. D’Amato 1975. Epigrafia della regione dei Marsi (Milan).



lippolis, e. 2000. “Cultura figurativa: la scultura ‘colta’ tra età repubblicana e dinastia antonina,” in M. Marini Calvani, R. Curina and e. lippolis (edd.), Aemilia: La cultura romana in Emilia Romagna dal III secolo a.C. all età costantiniana (Venice) 250-78. lugli, G. 1966. “I confini del pomerio suburbano di Roma primitiva,” in Mélanges d’archéologie, d’épigraphie et d’histoire offerts à Jêrome Carcopino (Vendôme) 641–50. Marini Calvani, M. 2000. “uomini e dei: religione e politica sul colle di Covignano,” in A. fontemaggi and O. Piolanti (edd.), Rimini divina. Religioni e devozione nell’evo antico (Rimini) 49-53. Monti, P. G. 1998. “Carta archeologica del territorio,” in f. Coarelli and P. G. Monti (edd.), Fregellae, vol. 1: Le fonti, la storia, il territorio (Rome) 81-111. Nenci, G. 1979. “Spazio civico, spazio religioso, e spazio catastale nella polis,” AnnPisa 9, 459-77. Pelgrom, J. 2008. “Settlement Organization and land Distribution in latin Colonies before the Second Punic War,” in l. de ligt and S. J. Northwood (edd.), People, Land and Politics. Demographic Developments and the Transformation of Roman Italy 300 BC-AD 14 (leiden) 333-72. Pugliese Carratelli G. (ed.) 1996. The Western Greeks (london). Quilici Gigli, S. 1978. “Considerazioni sui confini del territorio di Roma primitiva,” MÉFRA 90, 567-75. Quinn, J. C. and A. Wilson 2013. “Capitolia,” JRS 103, 117-73. Rebecchi, f. 1998. “Scultura di tradizione ‘colta’ nella cisalpina repubblicana. evoluzione e cronologie aperte,” in G. Sena Chiesa and e. A. Arslan (edd.), Optima via (Cremona) 189206. Riva, C. and S. Stoddart 1996. “Ritual landscapes in archaic etruria, in Approaches to the study of ritual,” in J. B. Wilkins (ed.), Approaches to the study of ritual. Italy and the ancient Mediterranean (london) 91-109. Scheid, J. 1987. “les sanctuaires de confins dans la Rome antique: réalité et permanence d’une représentation idéale de l’espace romain,” in


L’Urbs: espace urbain et histoire, Ier siècle av. J.C. - IIIe siècle ap. J.-C. (Rome) 583-95. Scheid, J. 1990. Romulus et ses frères. Le collège des Frères Arvales, modèle du culte public dans la Rome des empereurs (Rome/Paris). Scheid, J. and H. Broise 1989. “Recherches au bois sacré de Dea Dia (la Magliana, Rome),” RA, 199-202. Settis, S. (ed.) 1984. Misurare la terra: centuriazione e coloni nel mondo romano (Modena). Sommer, M. 2008. “Bauen an der Grenze. Überlegungen zur Monumentalisierung kultureller Identitäten,” in f. Pirson and u. Wulf-Rheidt (edd.), Austausch und Inspiration. Kulturkontakte als Impuls architektonischer Innovation (Mainz) 202-13. Stek, t. D. 2008. Sanctuary and Society in CentralSouthern Italy (3rd-1st centuries BC). A study into cult places and cultural change after the Roman conquest of Italy (Amsterdam). Stek, t. D. 2009. Cult places and cultural change in Republican Italy. A contextual approach to religious aspects of rural society after the

Roman conquest (Amsterdam). tarpin, M. 2002. Vici et pagi dans l’Occident romain (Rome). taylor, l. R. 1960. The voting districts of the Roman republic (Rome). torelli, M. 1999. Tota Italia. Essays in the cultural formation of Roman Italy (Oxford). toynbee, A. J. 1965. Hannibal’s legacy. The Hannibalic War’s effects on Roman life (london, New york). Vallet, G. 1968. “la cité et son territoire dans les colonies grecques d’Occident,” in La città e il suo territorio (Naples) 67-142. Wissowa, G. 1912. Religion und Kultus der Römer (Munich). zifferero, A. 1995. “economia, divinità e frontiera: sul ruolo di alcuni santuari di confine in etruria meridionale,” Ostraka 4, 333-50. ziolkowski, A. 2009. “frontier sanctuaries of the ager romanus antiquus: did they exist?,” Palamedes 4. zuffa, M. 1970. “Abitati e santuari suburbani di Rimini dalla protostoria alla romanità,” in Studi sulla città antica (Bologna) 299-315.




fIG. 1. Plan of Minturnae. the original colonial settlement is marked as castrum to the left (from lackner 2008, 358).


Livy 27.38 and the vacatio militiae of the maritime colonies luuk de ligt

In his book on the demographic, economic and social impact of Roman warfare during the Middle Republic Nathan Rosenstein discusses the amount of land which was needed to meet the property qualification for military service during the final decade of the Second Punic War. As he points out, the ownership of 7 iugera (1.75 ha) must have been enough to qualify a citizen for service in the legions. However, he argues there is one text which suggests that even those citizens who owned a mere two iugera (0.5 ha) could be called up for the field army.1 the textual evidence in question consists of a well-known passage concerning the levy of 207 B.C. As livy reports, the Roman supply of iuniores had been severely thinned out at this point. therefore the consuls decided to call up even the iuniores of the maritime colonies, despite the fact that the colonists on the coast were said to have an inviolable right to exemption from military service (sacrosancta vacatio militiae). When Ostia, Alsium, Antium, Anxur, Minturnae (fig. 1), Sinuessa and Sena refused to comply, the consuls ordered them to defend their claim before the Senate, and on the appointed day, the representatives of these 7 colonies duly appeared. they read out the documents which granted them exemption, but the Senate decided that these documents could not be regarded as having any validity “whenever an enemy remained in Italy” (cum in Italia hostis esset). Only in the case of Ostia and Antium was the exemption from legionary service applied as before (observata), but the colonists of these two towns had to swear an oath that they would not spend more than 30 nights outside the town walls for as long as the enemy remained in Italy.2 from livy’s account, it would seem to follow that the adult male citizen population of the maritime colonies, or at least a large proportion of that population, was eligible for service in the legions. Indeed, the fact that the colonists had been awarded a formal vacatio militiae (exemption from military service) suggests that this had been the case when the first maritime colonies were established, that is to say, as early as the second half of the 4th c. B.C.3 However, we also happen to know that citizens who participated in some maritime colonies founded before the Second Punic War received allotments of no more than two iugera. the basic information is found in livy’s brief reference to the foundation of the colony of Anxur/tarracina in 329 B.C. where each colonist received two iugera.4 It is by combining

1 2 3

4

Rosenstein 2004, 58. liv. 27.38. According to t. Cornell, the first fortified settlement at Ostia was established between 380 B.C. and 350 B.C., (Cornell 1995, 321) but more recent publications suggest a foundation date around 300 B.C., based on the ceramics found in the fossae of the castrum walls. See zevi 1996. the colony at Antium was founded in 338 B.C. liv. 8.21.11. Note that we have no information on the size of allotments given to colonists sent out to maritime colonies of the 3rd c. B.C. Since the 7-iugera allotments given to viritane settlers in Sabinum shortly after 290 B.C. were much larger than the plots re-

ceived by viritane settlers of the 5th and 4th c. B.C. (with the possible exception of those settled in the territory of Veii, but cf. Diod. Sic. 14.102.4), it would be unwise to rule out the possibility that the later policy of providing settlers sent out to coloniae civium Romanorum with allotments measuring between 5 and 10 iugera started sometime in the 3rd c. B.C. (Val. Max. 4.3.5; Columella, Rust. praef. 14; liv. 39.44; liv. 39.55; liv. 40.29). Cf. Bispham 2006, 122-23, for the observation that, contrary to what most publications on colonization in Roman Republican Italy suggest, we do not actually know that all maritime colonies of the 4th and 3rd c. B.C. received 300 colonists.




this piece of information with livy’s account of the levy of 207 B.C. that Nathan Rosenstein arrives at the conclusion that during the 4th and 3rd c. B.C., those citizens who owned only half a hectare of land must have been assidui rather than proletarians. In the absence of this assumption, he argues, the episode of 207 B.C. cannot be explained. the theory that the ownership of a mere two iugera of arable land would have qualified a Roman citizen for service in the legions is by no means new. earlier proponents include Philipp Huschke, theodor Mommsen, George Botsford and, more recently, Michael Crawford. Although their arguments differ in detail, most of them have tried to buttress their case by pointing to the Varronian tradition that two iugera formed the standard heredium of Roman peasants in the Regal period.5 Since peasants formed the backbone of the Roman Republican army, this tradition seems to presuppose that the ownership of a two-iugera plot would have qualified a Roman citizen for legionary service. Precisely because stories about the tiny heredium of the Regal period do not appear before the second half of the 1st c. B.C., there is every reason to doubt the reliability of this tradition. As emilio Gabba has pointed out, the tradition that Romulus granted two-iugera allotments to his followers is not found in the annalistic tradition represented by livy and Dionysius. Similarly, while both livy and Dionysius report that in 504 B.C. the Sabine leader Attus Clausus (Appius Claudius) migrated to Roman territory with 5,000 of his clients, neither of them provide information concerning the size of the plots which were given to these immigrants. the story that each of them received a two-iugera allotment is found only in Plutarch.6 the marginal place of the tradition concerning the heredium in the literary tradition leads Gabba to infer that the concept of the two-iugera heredium is unlikely to have been very old. In his view, we might well be dealing with an antiquarian construct of the 2nd c. B.C. the principal aim of which was to account for the fact that a centuria of land contained 100 two-iugera plots.7 More recently Stephen Oakley has suggested a different origin for the tradition represented by Varro and later writers. He argues that all references to two-iugera allotments in stories concerning the Regal or early Republican period can be explained as back projections of the more recent practice of granting twoiugera plots to Roman citizens sent out to the maritime colonies of the 4th c. B.C.8 It seems reasonable to conclude there are some good grounds to dismiss all references to the bina iugera of Regal and early 5th c. B.C. Rome either as purely fictitious constructs or as anachronistic retrojections. It must, however, be emphasized that this conclusion does not apply to various references to

5

6

7



Huschke 1838, 643-44; Mommsen 1887, III.1, 248-49; Botsford 1909, 86; Crawford 1985, 23-24. Cf. Varro, Rust. 1.10.2; Plin., HN 18.7. Attus Clausus: liv. 2.16; Dion. Hal., Ant. Rom. 5.40. two-iugera plots given to his followers (Plut., Vit. Pobl. 21). Gabba 1978; cf. Capogrossi Colognesi 2012, 62, n. 3. those inclined to accept the two-iugera heredium of Regal times as a historical fact must either posit the existence of an ‘infield/outfield’ system with easy access to arable land not held in private ownership (De Neeve 1984, 65, n. 13), or the existence of a dependent peasantry paying rents in kind for the right to use some of the land belonging to rich members of the elite in addition to their tiny freeholds (e.g. Crawford 1985, 24). yet another solution has been proposed by luigi Capogrossi Colognesi (Capogrossi Colognesi 2012, 62-65). He ingeniously suggests that the colonial

8

schemes of the Regal and early Republican periods involved the assignation of two iugera for each adult male member of the families participating in them (cf. liv. 5.30 for the assignations carried out in the ager Veientanus). However, in his brief reference to the establishment of the maritime colony of Anxur/tarracina, livy explicitly states that two iugera were given to each of the 300 colonists rather than to each adult male member of their families (liv. 8.21.11). It would be unwise to rule out the possibility that the source lying behind livy misinterpreted the basic rules governing this colonial venture, but it seems methodologically preferable to look for a solution that does not require us to make this assumption. Oakley 1997, 676-77; cf. Smith 2006, 194, n. 30.


Roman citizens receiving tiny plots of land in the late 5th or early 4th c. B.C. We are told by livy, for instance, that each of the 1,500 colonists sent out to labici in 418 B.C. received two iugera of land. In another passage he states that each of the 2,000 colonists sent out to Satricum in 395 B.C. was given an allotment measuring two-and-a-half iugera.9 It is not immediately apparent why livy, or any of his predecessors, should have gone to the trouble of inventing a tradition crediting Roman colonists of the late 5th and early 4th c. B.C. with such tiny plots of land. this observation applies a fortiori to livy’s casual reference to the two-iugera allotments granted to the settlers of tarracina, which even the most determined critics accept as reliable. It has often been observed that Roman colonial policies were based on the assumption that those sent out to newly conquered territories would be capable of defending their holdings and indeed of contributing men for new military exploits. for this reason alone it seems unlikely that any Roman magistrate or politician should have advocated a policy of providing any type of colonist with allotments whose size fell short of the property requirement for military service. It would seem to follow from this that the ownership of two or two-and-a-half iugera of arable land was enough to meet this threshold. As we have seen, this inference is supported by the indisputable fact that the colonists sent out to Antium in 338 B.C., whose privately owned allotments are unlikely to have been larger than those granted to the colonists of tarracina, were given a formal vacatio militiae in order to make sure that their number would not be depleted by future levies. In other words, the conclusion that throughout the 4th and 3rd c. B.C. even very poor Roman citizens could be called up for normal legionary service seems inescapable. It is only when we take a closer look at the Roman system of property classes in its late 3rd c. B.C. form that some weaknesses in Rosenstein’s theory come to light. Building on various publications by elio lo Cascio and Dominic Rathbone, Rosenstein operates with a threshold of 1,100 libral asses (corresponding to 5,500 sextantal asses of the period 212-141 B.C.) for membership of the fifth class during the final decades of the 3rd c. B.C.10 Since the figure of 1,100 not only appears in the manuscript of Cicero’s De Re Publica but also seems to lie behind the property requirement of 11,000 (sextantal) asses which we find in livy’s description of the centuriate organisation allegedly introduced by Servius tullius, this is a reasonable assumption.11 However, if the hypothetical threshold of 1,100 libral asses corresponds to the notional value of two iugera of arable land, we end up with a land price of 550 libral asses per iugerum, corresponding to 1,100 sesterces of the 2nd c. B.C. this land price is higher than the price of 1,000 sesterces per iugerum which Columella gives for the late 1st c. A.D.12 In this context it must be remembered that Columella is referring to land purchased for the setting up of a slave-run villa producing for the urban market. Such land would have been in a good location and

9 10

11

12

liv. 4.47 and 6.15; cf. Pelgrom 2012, 85. Rosenstein 2004, 57; building on lo Cascio 1988 and Rathbone 1993. Cic., Rep. 2.40. the original manuscript reading in this passage was undoubtedly mille centum aeris (‘1,100 asses’), but modern scholarship remains divided over the question as to whether the late antique corrector altered this to mille quingentos aeris (1,500 asses). See e.g. lo Cascio 1988, 286-88; Rathbone 1993, 139-42. Whatever view we take of this paleographic problem, the sum of 11,000 (sextantal) asses mentioned by livy as the original threshold for membership of the fifth class speaks strongly in favor of mille centum aeris as the correct reading. Columella, Rust. 3.3.8, to be consulted with De Neeve 1985. As De Neeve points out, the enormous

variation in land prices which must have existed in all periods of Roman history undermines the concept of a notional standard price. Cf. Northwood 2008 for the suggestion that for the purposes of the census land plots were valued at current market prices. On the other hand, the indisputable fact that land prices were determined or influenced by a rich variety of economic and non-economic factors need not entirely undermine the idea that contemporary observers did in fact operate with notional standard prices for particular categories of land. Note that the epigraphic evidence concerning sales of land by auction in Classical Athens has been interpreted as pointing to a standard price of 50 drachms per plethron (one-eleventh of a hectare) of ordinary arable land. Cf. next footnote.




would have been worth more than most of the land worked by ordinary peasants. Partly for this reason, Richard Duncan-Jones has argued that even in early Imperial Italy average land prices must have been in the order of 500 sesterces per iugerum. Regardless of the merits of this inference, it seems incredible that land prices in the Mid-Republican era exceeded those prevailing in the second half of the 1st c. A.D.13 It also seems worth pointing out that in livy’s description of the ‘Servian’ system of classes, which is widely seen as a back projection of the system in its 3rd c. B.C. form, the property qualification for membership of the first class is about nine times higher than that for membership of the fifth class. Hence, if it is assumed that in the decades before the Second Punic War the property qualification for membership of the fifth class stood at 1,100 libral asses and that this sum corresponded to the notional value of two iugera of arable land, we must also accept that any Roman citizen owning between 18 and 20 iugera (4.5 hectares) of arable land would have qualified for membership of the first class. this outcome is manifestly absurd. A solution to this conundrum may be found by looking at the history of the Roman system of property classes. In the centuriate organization described by Cicero, livy and Dionysius we find five property classes, with property thresholds of 100,000, 75,000, 50,000, 25,000 and 11,000 (or 12,500) asses.14 the literary sources are unanimous in attributing the creation of this system to king Servius tullius, but the thresholds mentioned by livy and Dionysius seem much too high for the Regal period and there are good grounds to think that the centuriate organization was originally based on fewer than five classes. An important clue concerning the military and political structures of Regal and early Republican Rome is a passage from Gellius who reports that originally there was just one classis, with a threshold of 125,000 asses. All those whose assets fell short of this threshold were considered to be infra classem.15 Building on this clue, modern scholarship has developed the theory that in late Regal and early Republican times the term classis must have referred to those economically capable of equipping themselves as hoplites. Only these men would have served in the legions.16 An inevitable corollary of this reconstruction is that those citizens who were infra classem must have served as light infantry. In other words, if Gellius can be relied upon, the light infantry of Regal and early Republican Rome were originally drawn from all poorer independent property-owners.

13



Duncan-Jones 1982, 48-52. Rathbone, suggests that around 140 B.C. the notional price of one iugerum of grain land may have been ca. 400 sesterces (Rathbone 1993, 146, n. 20). fragmentary data on Athenian and Roman horse prices suggest that Rathbone’s estimate is of the right order of magnitude for the 3rd and 2nd c. B.C. In 4th c. B.C. Athens 500 drachmas was the median price of a war horse, although both lower and much higher prices are attested (Kroll 1977, 89; Spence 1993, 275-76; Oliver 2006, 118-19). In this period, the sum of 500 drachmas would have bought about 0.9 ha of arable land (Andreyev 1974; Hanson 1995, 471; lambert 1997, 229-33, 257-65). According to Varro, 1,000 libral asses (corresponding to 5,000 sextantal asses) were enough to buy a public horse, presumably in the mid-3rd c. B.C. (Varro, Ling. 8.71). If we assume that the ratio between land prices and horse prices prevailing in Mid-Republican Rome was approximately the same as that which can be reconstructed for early Hellenistic Athens, one iugerum of grain land would have cost about 275 libral asses, cor-

14

15

16

responding to 550 sesterces of the 2nd c. B.C. If part of the two-iugera plots received by those sent out to the maritime colonies were used for market gardening or were planted with fruit trees (as seems likely), these plots would have been worth more than grain land, but the hypothetical price of 275 libral asses per iugerum of grain land is almost certainly too high for the 330s and 320s B.C. Cic., Rep. 2.39-40; liv. 1.43.1-8; Dion. Hal., Ant. Rom. 4.16-21. Gell. 6.13. like the thresholds referred to by livy and Dionysius, the minimum property qualification of 125,000 asses is far too high for the 6th or 5th c. B.C., but Gellius’ claim that originally there was just one classis is generally seen as plausible (see e.g. Cornell 1995, 184; Rathbone 1993, 127). e.g. Cornell 1995, 183-84, building on the findings and theories of various earlier scholars, especially those of the eminent Italian ancient historian Plinio fraccaro.


In principle, only those citizens owning two or more iugera of arable land may have been called up for this type of service, but none of our admittedly extremely fragmentary sources refers to the existence of such a threshold. According to tim Cornell, this state of affairs lasted until 406 B.C., when the Roman state first began to pay wages (stipendium) to its soldiers.17 the bronze needed for these payments was collected by imposing a direct ‘war tax’ (tributum) on Roman citizens.18 Since it was decided to distribute the burden of taxation in accordance with people’s wealth, the introduction of stipendium and tributum required the creation of a more complicated system of property classes. Cornell argues in favor of a scenario in which the single classis of heavily-armed legionaries was split up into three sub-classes while those property-owners who had previously been infra classem were assigned to two further classes. those owning insufficient property to qualify for membership of the second of these classes (the fifth class of the reformed comitia centuriata) must have been released from the obligation to pay tributum.19 In short, if we accept Cornell’s theory, the year 406 B.C. saw the simultaneous introduction of the five property classes still existing in Middle and late Republican times, the integration of the light infantry into the centuriate organization, and the creation of a clear demarcation line between light infantry on the one hand and proletarians not liable to normal military service on the other. In this system those liable to impositions of tributum were the same people who could be called up for normal military service and were therefore entitled to receive military pay.20 the elegance of Cornell’s hypothesis cannot be denied, but that does not mean it must be correct. In fact, the dearth of reliable evidence relating to the development of the armies and political assemblies of early Republican Rome makes it relatively easy to put together various alternative scenarios leading from a single classis to the five-classes system of the mid-3rd c. B.C. the outlines of one such scenario have been sketched by Rathbone. Building on Gellius’ statement about the original meaning of the term classis, he argues that the single classis of Regal and early Republican times might well have survived intact until the late 4th c. B.C. He buttresses this hypothesis by pointing out that the latter period witnessed a whole cluster of important changes. these included a considerable increase in the number of assidui through settlement schemes, the abolition of nexum in 326 or 313 B.C., the regular levy of an enlarged force of four legions (from 311 B.C.), and the adoption of the manipular formation and lighter body-armor. Rathbone suggests that these developments may well have prompted the introduction of a new four-classis system one of whose effects was to bring down the property threshold for legionary service. As a next step, the fifth classis was created, presumably at some date in the first half of the 3rd c. B.C. Since light infantry now began to be recruited exclusively from this class, its creation must have had the effect of debarring property-owners

17

18

19

I cannot accept Cornell’s inference that before 406 B.C. military service must have been the preserve of a wealthy group who could afford their own armor and weapons (ibid., 188). Recent studies of Greek warfare in Archaic and Classical times have called into question the existence of armies consisting exclusively of heavily-armed hoplites (e.g. Van Wees 2004, 61-65; trundle 2010, esp. 141). Cornell 1995, 187. Cf. liv. 4.59.11-60.8, but see Northwood 2008, 265, n. 30, for the observation that livy does not explicitly state that tributum was introduced in 406 B.C. Cornell 1995, 186-89. Note that this ingenious theory stands or falls with the assumption that regular payments of stipendium continued after the capture of Veii. taking a skeptical view of the reliability of the

20

annalistic tradition, W. V. Harris suggests that regular payments of stipendium are unlikely to have started before 280 B.C., when Roman began to mint coins on a significant scale (Harris 1990, 507). However, as Kristian Mohr Mersing observes, earlier payments might have been made in aes rude (Mersing 2007, 231-32). On this point, Cornell’s position is identical to that of Claude Nicolet who has defined tributum as “un impôt payé par les mobilisables au profit des mobilisés” (Nicolet 1976, 29). However, note that proletarians serving in Roman fleets of the 3rd and early 2nd c. B.C. received stipendia without being liable to tributum. the somewhat mysterious accensi velati who seem to have accompanied the legions of the 4th and 3rd c. B.C. might have included proletarians (Mommsen 1887, vol. 3, 283-84; Brunt 1987, 402; Daly 2002, 72-73).




below its census from normal military service and its rewards.21 In other words, instead of being created at one specific moment in Roman Republican history, the five-classis system of Mid-Republican times is more likely to have developed gradually, and the old practice of drawing infantry from all poorer independent property-owners might well have persisted until the early or mid-3rd c. B.C. Of course the nature of the surviving evidence makes it more or less impossible to decide which of these competing reconstructions of the history of the classis system is more likely to be correct. to make things worse, there may be room for further theories which are equally plausible and equally untestable.22 Nonetheless, it seems worth pointing out that Rathbone’s theory makes it possible to account for the existence of a threshold of 1,100 libral asses for membership of the fifth class during the second half of the 3rd c. B.C. and for the vacatio militiae awarded to the maritime colonies of the 4th c. B.C., which seems to presuppose a much lower threshold for military service – or even the complete absence of such a threshold. I now move on to another difficulty raised by the fragmentary evidence concerning the maritime colonies of the 4th c. B.C.: what are the economic implications of the two-iugera allotments received by the settlers sent out to tarracina in 329 B.C. and presumably by other groups of colonists sent out to other maritime colonies of this period? Among specialists in the field of Roman agrarian history there has been some debate about whether it was possible for a small peasant family to survive on a plot of land of 7 iugera, the equivalent of 1.75 hectares. the majority view seems to be that plots of this size could have sustained a family of four if the land was worked intensively and if almost none of it was fallowed.23 However, the very existence of a debate about the viability of farms of 7 iugera shows that a plot of two iugera cannot possibly have produced all or even most of the food needed to sustain a Roman family. faced with this inescapable conclusion, some scholars have argued that colonists who were given half a hectare of privately owned land must have had access to communal pasture land. A prominent proponent of this view is Gabba, who has argued that large numbers of peasants throughout Central and Southern Italy must have been structurally dependent on access to public pasture land because their holdings were too small to sustain them. He buttressed this theory by pointing to the evidence from Medieval and early Modern Italy where various kinds of communal grazing land also played a crucial part in the rural economy. In similar vein, Crawford writes that “in the late Republic there was a category of common land and it is reasonable to suppose that a peasant might supplement an income from his freehold by grazing on such common land”.24 While these suggestions are clearly an important step in the right direction, I feel that the Medieval analogy is potentially misleading. the main reason for this is that, at least in those parts of Republican Italy which belonged to the ager Romanus, public land was not used exclusively as pastureland but included extensive tracts of arable land which could be occupied by farmers, the so-called ager occupatorius. In an important article, which appeared in 2003, Rathbone pointed out that large tracts of newly conquered land were used for colonization or viritane assignations. We also happen to know that at least some state-owned

21 22



Rathbone 1993, 146-47. the surviving evidence might also be accounted for by distinguishing between five stages of development: 1. A single class of heavy infantry dominates the comitia centuriata; 2. At some point during the ‘Struggle of the Orders’ light infantry obtain votes in this assembly; 3. Tributum is imposed on all those liable to be called up as heavy or light infantry, with the possible exclusion of those owning fewer than two iugera of arable land; 4. further sub-classes are created; 5. After the extensive settlement schemes of the late 4th and early 3rd c. B.C. the light infantry are upgraded, 1,100

23 24

libral asses becoming the new threshold for military service (and for liability to tributum?). upgrading of the light infantry is suggested by the contrast between the light-armed citizen soldiers of Archaic and Classical Greece, many of whom did not wear helmets and fought with slings and stones, and the well-equipped Roman velites described by Polybius (Polyb. 6.22.1.-2; Van Wees 2004, 62; cf. the probably fictitious description of the equipment of the fifth class in liv. 1.43.7). Roselaar 2010, 204-6. Gabba 1977; Crawford 1985, 24. Cf. also Dilke 1971, 178-79, n. 4.


land was sold off. from these indications Rathbone infers that many Roman peasants did not have access to ager occupatorius.25 this might well be correct for large parts of Italy in the 2nd c. B.C. In the specific case of the maritime colonies, however, it must be kept in mind that these colonies were founded in areas where the Roman state had confiscated substantial tracts of land. If we also take into account the small number of settlers and the small size of the area which they received in private ownership, it is more or less impossible to avoid the conclusion that, at least in the early stage of the existence of these new communities, the colonists had easy access to additional ager publicus which could be used to grow crops.26 We must therefore reckon with the possibility that the tracts of arable land cultivated by many colonists of the first generation (and by their descendants) were quite extensive. Since, as far as we know, holdings consisting of public land did not count towards the property ratings determining membership of the five classes, the mere fact that the amount of arable land used by the settlers and their descendants far exceeded the bina iugera which they owned ex iure Quiritium would, in and by itself, not have qualified them for service in the legions.27 However, if I am right in supposing that these people controlled large plots of arable ager publicus in addition to their private holdings, it must surely be inferred that many of them owned a solidly built farmhouse, a pair of oxen, one or two slaves and a dozen sheep or goats.28 On this basis it

25 26

27

28

Rathbone 2003. Mason 1992, 86-87 sees use of arable land not owned by the colonists as part of the solution, alongside access to communal pasture land and consumption of wild plants collected from marginal areas or fallow land. While plebeians may have encountered difficulties when trying to occupy ager publicus in the vicinity of Rome, exclusion of plebeians by patricians – or of poor plebeians by wealthy nobiles – seems less likely after 367 B.C., especially in a colonial context (e.g. Cornell 1995, 269, referring to Cassius Hemina fr. 17 P). In estimating the value of a citizen’s property, the censors took account of the bona censui censendo. Mommsen held that originally only res mancipi (mainly Italian land, slaves and large farm animals) owned ex iure Quiritium fell into this category (Mommsen 1887, vol. 2, 389). During the Mid-Republic the value of many kinds of moveable property, such as clothing and jewellry had to be declared, but there is nothing to suggest that the range of goods taken into account by the censors was ever widened to include holdings of public land (see e.g. Nicolet 1980, 69; Northwood 2008, 260). J. W. Rich suggests that holdings of public land counted towards the maximum of 500 iugera which Roman citizens were permitted to possess under the lex Licinia of 367 B.C., but he does not question the existence of a clear distinction between private and public land in this period (Rich 2008). liv. 22.14 refers to farm buildings in the territory of Sinuessa. On land prices see above, at notes 12-13. According to Gellius, the Lex Aternia Tarpeia of 454 B.C., laid down that the value of an ox equaled that of 10 sheep or

of 100 pounds of bronze, corresponding to 100 (or 120?) libral asses of the early 3rd c. B.C., but the reliability of this tradition is doubtful (Crawford 1985, 19-20; Gell. 11.1.2; Magdelain 1995, 115-16). Note that Cicero, who seems to attribute these valuations, or similar ones, to the lex Iulia Papiria of 430 B.C., regarded them as advantageous (levis) for those who had been fined, suggesting that the valuations in bronze were set at an artificially low level, thereby encouraging people to pay their fines in metal (Cic., Rep. 2.60). If we ignore the slave prices referred to in Plautus and terence, many of which give the impression of being inflated for comic effect, there are almost no data on slave prices in Republican Italy, Plutarch’s assertion that Cato the elder never paid more than 1,500 drachmas (= 1,500 denarii?) for a slave being one of the rare exceptions (Plut., Vit. Cat. Mai. 4.5). According to livy and Plutarch, Roman war captives who had ended up as slaves in Greece were ransomed in 195 B.C. at a rate of 500 denarii each (liv. 34.50.6; Plut., Vit. Flam. 13.5). this looks like a ‘normal’ slave price, but both lower and higher ransom rates are reported in other sources and the relationship between ransom rates and slave prices is likely to have been flexible (e.g. liv. 22.5: 100 quadrigati = 150 denarii for a slave, but 300 quadrigati = 450 denarii for a foot soldier and 500 quadrigati = 750 denarii for a cavalryman, immediately after Cannae). After a more or less comprehensive review of the evidence Walter Scheidel concludes that Republican slave prices are likely to have been similar to those prevailing in Classical Athens where a slave typically cost between 200 and 500 drachmas (Scheidel 2005, 16).




might be conjectured that even at the start of the Second Punic War, when the property threshold for membership of the fifth class seems to have stood at 1,100 libral asses (corresponding to 5,500 sextantal asses of the period 212-141 B.C.), at least a certain proportion of the citizen population of the maritime colonies owned sufficient property to qualify them for normal legionary service.29 In this context it must be remembered that around 212 B.C. the threshold for legionary service is thought to have been lowered to 4,000 sextantal asses.30 A side effect of this reduction would have been to make a somewhat larger proportion of the citizens of the maritime colonies liable for military service – unless of course the descendants of the colonists continued to enjoy complete immunity from all forms of military service as a result of the vacatio militiae which the first settlers had received more than a century earlier.31 up to a point, this hypothetical reconstruction is supported by the archaeological evidence from the territory of tarracina. As is well known, traces of a centuriation grid comprising 7 complete and 6 incomplete centuriae have been discovered to the north-west of the town (fig. 2). unfortunately, the interpretation of this grid is hedged in with difficulties. the most frustrating of these is that it cannot be securely dated. It is true that most specialists are inclined to connect it with the foundation of the colony, but the absence of any hard evidence has allowed focke Hinrichs to argue in favour of a date in the triumviral period.32 Another difficult question concerns the spatial relationship between this centuriation grid and the twoiugera plots referred to by livy. It seems reasonable to suppose that the small allotments of which the colonists were granted the ownership were intended as places where they could build their farms, lay out their gardens and plant a couple of fruit-bearing trees. Why, though, should all these farms and gardens have been clustered in a relatively small centuriated area? Interestingly, there are no traces of any further early land division scheme outside the area covered by the supposedly 4th c. B.C. grid. It is true that traces of other grids have been discovered to the west of tarracina, along the Via Appia, but none of these grids seems to have been laid out before the third quarter of the 2nd c. B.C. According to a passage in the Liber coloniarum, the land of tarracina “was left unsurveyed” (in absoluto est dimissus).33 As a general description of the territory of tarracina this notice is certainly invalid for the final years of the Republic or for the early empire, when a substantial part of the territory of

29



It has often been suspected that most of the citizens sent out to the coloniae civium Romanorum of the 4th and 3rd c. B.C. were proletarians (see e.g. Càssola 1988, followed by erdkamp 2011). Saskia t. Roselaar argues that since the value of iugera of land must have fallen short of the property qualification for legionary service, the colonists must have remained proletarians after moving to their new homes (Roselaar 2009). from this, she infers that the aim of the vacatio militiae was to make sure that the colonists and their descendants would remain exempt from military service if they acquired additional property at a later stage. In my view, it is better to assume that most of the colonists or their descendants became proletarians as a result of the creation of the fifth class or of the introduction of a higher property threshold for membership of that class at some date in the 3rd c. B.C. Note that this hypothesis is in perfect accordance with the fact that citizens of the maritime colonies were recruited for service in the fleet in 191 B.C. (liv. 36.3.46). Normally only proletarians were called up for this

30

31

32

33

type of service. Contrary to what is suggested by Johannes H. thiel, there is nothing to suggest that citizens of the coloniae maritimae had been called up for naval service before 191 B.C. (thiel 1946, 276-77). Gabba 1976, 5-6; Brunt 1987, 402-3; Rathbone 1993, 141 and 144. Mutatis mutandis the same would have been true of the descendants of settlers sent out to maritime colonies of the 3rd c. B.C., although the latter may have received allotments larger than two iugera. Cf. above. In favor of an early date: Cancellieri 1990; triumviral date: Hinrichs 1974, 55-56. Chouquer et al. assign the centuriation grid to the triumviral period but claim to have found a much earlier system of parallel lines producing square-shaped two-iugera plots on the north side of the Via Appia (Chouquer et al. 1987, 105-9). the existence of these early division lines is disputed by Margherita Cancellieri (Cancellieri 1990). for a recent survey of the discussion see Pelgrom 2012, 99-103. Lib. colon. 186-87 (ed. Campbell 2000).


fIG. 2. traces of an early Roman land division scheme recognized in the territory of terracina (after Cancellieri 1990, 69, fig. 6).

the town had been centuriated. this had led Campbell to suggest that the passage in the Liber coloniarum might be corrupt.34 However, it seems worth pointing out that an irregular pattern of land division is exactly what one would expect to find if most of the land to the west of tarracina had originally been ager occupatorius, of which the colonists and their descendants had taken possession in a haphazard fashion. Perhaps then the Liber coloniarum should be understood either to refer to the period before the triumviral assignations or to conditions in a certain part of the territory of tarracina.35 the third and final problem which will be examined in this paper concerns livy’s use of the expression sacrosancta vacatio to refer to the exemption claimed by the colonists (cf. Sisani in this volume). In some older publications it is argued that the colonists must have based their claim on a tribunician law which could be called ‘sacrosanct’ because the tribunes themselves were regarded as ‘sacrosanct’.36 However, while livy often uses the term sacrosanctus to refer to the ‘inviolable’ status of the tribunes, there is nothing to suggest that all of the laws proposed by the tribunes of early and Mid-Republican times were ‘sacrosanct’ by definition. In order to obviate this difficulty, Mommsen and many later scholars have interpreted livy’s use of the expression sacrosancta as indicating that the colonists were protesting against the infringement of a privilege which had been awarded by a lex sacrata, that is to say, by a law which had been confirmed by an oath

34 35

Campbell 2000, 389-90. A third possibility is that the Liber coloniarum refers to the disintegration of the triumviral grid in early Imperial times.

36

e.g. Ruperti 1843, 775; Dumont 1844, 23.




sworn by the concilium plebis.37 While this theory is superficially attractive, it runs up against the difficulty that all known leges sacratae belong to the early or mid-5th c. B.C.38 the only possible exception is the preGracchan lex de modo agrorum, which is probably to be identified with the lex Licinia of 367 B.C. If Appian is to be believed, this agrarian law was confirmed by an oath.39 In a valuable book on the meanings of the latin words sacer, sacrosanctus and various related terms Claudia Santi suggests that the exemption from military service claimed by the maritime colonies could be called ‘sacrosanct’ because it had been awarded in a treaty (foedus) stating that any transgressor would be sacer (dedicated to the gods). In practice this would have amounted to declaring all transgressors liable to the death penalty.40 the principal weakness of this theory is that there is absolutely nothing to suggest that the relationships between Rome and the maritime colonies were regulated by treaties.41 Moreover, Santi’s interpretation is guided by the assumption that the basic meaning of the term sacrosanctus must be something like ‘guaranteed by declaring any transgressor to be a homo sacer’, that is to say, by granting general permission to kill such a person with impunity.42 At first sight this interpretation might seem to be supported by the well-known tradition concerning the oath sworn by the plebs of early Republican Rome to kill anyone who would dare to violate the sacrosancta potestas of any of the plebeian tribunes.43 However, as Mommsen noted long ago, a close reading of the sources concerning the early history of the tribunate leaves no doubt that the term sacrosancta was originally used to refer to the protection afforded by the collective oath, so that the expression sacrosancta potestas must mean something like “power guaranteed by an oath” rather than “power which makes anyone who lays hands on its bearer sacer”.44

37



festus, Gloss. Lat. 422. lindsay, defines leges sacratae as those laws which made the transgressor sacer, but commencing with Mommsen many scholars have argued that this definition is either too restrictive or based on a misunderstanding (Mommsen 1887, II. 1, 303 n. 2). In their view, any law which had been reinforced with an oath could be called a lex sacrata (see e.g. Maschke 1906, 27-28; Von fritz 1976, 385-86; Cornell 1995, 259, and the exhaustive discussion by Marottoli 1979). Note that festus refers to an alternative tradition according to which the leges sacratae were those laws which the plebs had approved and reinforced with an oath (iurata) during its meetings on the Mons Sacer (festus loc. cit.). Mommsen and Von fritz speculate that an oath was added to give the law awarding the privilege of vacatio greater permanence than ordinary laws (Mommsen 1887, I, 242-43; Von fritz 1976, 376). Combining the two traditions reported by festus, Max Kaser infers that all sworn leges of early Republican times must have prescribed the penalty of sacratio (in effect, the death penalty) for transgressors (Kaser 1949, 52-53). However, he also interprets the prohibition on killing a Roman citizen without a trial (interfici indemnatum) which Cicero and later sources attribute to the twelve tables as indicating that as early as the mid-5th c. B.C. divine retribution was the only penalty entailed by sacratio, at

38

39

40 41 42 43

44

least in cases which had not been judged by the comitia centuriata (Crawford 1996, vol. 2, 696-700). See the brief list in Botsford 1909, 265 n. 1, and the survey by Von fritz 1976. e.g. Maschke 1906, 67; Roselaar 2010, 102. the latter keeps open the possibility that Appian ascribed an oath to the lex Licinia which had not actually been sworn. Hölkeskamp interprets the fact that the plebiscitum ‘ne quis postea populum sevocaret’ of 357 B.C. prescribed the death penalty for transgressors as an indication that this could have been another lex sacrata, but the passage from livy which is our only source of information for this law does not mention an oath (Hölkeskamp 1987, 95-96, liv. 7.16). Santi 2004, 187-89. thus correctly De Sanctis 1972, 350. e.g. Santi 2004, 195. Cf. also Scheid 2003, 26. liv. 3.55; Dion. Hal., Ant. Rom. 6.89; App., B Civ. 2.108; festus, Gloss. Lat. 424 lindsay. Mommsen 1887, vol. 2, 286-87. the best recent discussions of the term sanctus are thomas 1988 and id. 1993. His main finding is that the sources use this term to refer to objects (such as city walls) or people whose inviolability had been guaranteed by a law or by an oath. In my view (cf. main text below), there are good reasons to think that already in Mid-Republican times the term sacrosanctus acquired a less specific meaning after the


In my view, the most economical solution to the problems posed by livy’s use of the term ‘sacrosanct’ in connection with the vacatio militiae of the maritime colonists is that the latter’s claim to exemption from military service rested on a provision contained in a charter regulating the rights and duties of the first settlers and their descendants. If livy’s account can be relied upon, the representatives of the 7 recalcitrant colonies could present documents supporting their claim to vacatio militiae. these documents can plausibly be identified as copies of the leges datae regulating the rights and obligations of the citizens of the maritime colonies. It seems worth pointing out that the Caesarian lex Ursonensis contains a reference to the militiae munerisque publici vacatio sacrosancta (‘the sacrosanct exemption from military service and public duties’) enjoyed by the pontifices and augures of urso.45 the law explicitly says that the pontifices of Rome enjoyed the same privilege. there is some literary evidence to suggest that the exemption enjoyed by the pontifices of Republican Rome had been awarded by a Regal or early Republican law,46 but as the lex Ursonensis shows, the immunity enjoyed by the priests of a colonial community might rest on a provision included in the lex data imposed by Rome at the time of the colony’s foundation. On this basis it does not seem far-fetched to suppose that the legal charters of the maritime colonies of the 4th and 3rd c. B.C. contained a provision exempting their citizens from military service. It also seems possible that the right of exemption awarded to the colonists was declared to be ‘sacrosanct’ in order to expel any doubts concerning its ‘inviolable’ character.47 Piling Ossa on Pelion, it might be speculated that this particular use of the term ‘sacrosanct’ belongs to a period in which the early Republican practice of reinforcing certain laws by swearing collective oaths had fallen, or was beginning to fall, into abeyance.48 under these circumstances it would have begun to make sense to affirm the unchangeable character of various rights and privileges by declaring them to be just as ‘sacrosanct’ as if they had been awarded in a sworn law. In other words, rather than belonging to the period of the leges sacratae, the expression vacatio sacrosancta might well have been created at a time when the meaning of the term sacrosanctus had already begun to evolve from its original meaning ‘guaranteed by an oath’ to its later meaning ‘inviolable’. the other side of the coin is that there is absolutely no evidence to suggest that any ‘sacrosanct’ rights or privileges awarded by laws or municipal charters which had not been confirmed by oaths were systematically ignored. this suggests to me that even after the early Republican practice of reinforcing certain laws by collective

45

conceptual connection with oath-taking became progressively looser. In his translations of the lex Gabinia Calpurnia of 58 B.C. and of the Caesarian lex coloniae Genetivae Michael H. Crawford translates sacrosanctus rather vaguely as ‘prescribed by what is sacred’ (Crawford 1996, 349, 423). the fact that various late Republican laws have the closing formula si quid sacro sanctum est, quod non iure sit rogatum, eius hac lege nihil rogatur supports this less specific interpretation (ibid., 2324). Stretching the meaning of the term sacrosanctus further, Cicero can refer to possessiones sacrosanctae and Pliny the younger to the memoria sacrosancta of a deceased friend (Cic., Cat. 2.18; Plin., Ep. 7.11.3). Lex Coloniae Genetivae (FIRA I, 21), Chapter lxVI. As noted by Jörg Rüpke, the fact that Cicero refers to the vacatio militiae of the pontifices and the augures as a vacatio deorum suggests that the ‘sacrosanct’ status of their immunity was seen as having a religious back-

46

47

48

ground, but this might well be a late Republican reinterpretation (Cic., Acad. Pr. 2.121; Rüpke 1990, 63-64). See App., B Civ. 2.150; Dion. Hal., Ant. Rom. 2.21.3, 4.62.5, 4.74.4 and 5.1.4. Mommsen 1887, vol. 3.1, 243, suggested that a formal obligation of permanent residence (Domicilzwang) must have been placed on all citizens sent out to a newly established maritime colony. Going one step farther, edward togo Salmon interprets livy’s reference to the oath imposed on the citizens of Ostia and Antium as a reaffirmation of an existing regulation (Salmon 1969, 77). While there is nothing to contradict these inferences, it is equally possible that the oath of 207 B.C. was a novelty and that it was only at this stage that the obligation of quasi-permanent residence was made explicit. Of course, the late Republican iusiurandum in legem, sworn by magistrates or by magistrates and senators, is something completely different (see Roselaar 2010, 101-2).




oaths had fallen into disuse, any rights and privileges which had been declared ‘sacrosanct’ in later laws or charters were still being regarded as ‘inviolable’ in practice. So what happened in 207 B.C.? If my hypothetical reconstruction is correct, the colonists had a strong case. Although at least some of them are likely to have owned sufficient property to qualify for membership of the fifth class, they could prove that the leges datae of their communities gave them a ‘sacrosanct’ exemption from legionary service. In addition to this, they could point to the fact that before 207 B.C. the citizens of the maritime colonies had never been called up for this type of service.49 these were powerful arguments in a society where formal rights and precedent were important in political and legal debates. What justification could the Senate possibly have had for setting aside the claim of the colonists? In my view, the most likely answer is that Hasdrubal’s advance into Italy had prompted the Roman government to declare a state of emergency (tumultus) and that the Senate took the line that the privilege of vacatio enjoyed by the colonists was invalidated by these exceptional circumstances. During the Second Punic War a regional tumultus seems to have been declared after Cannae. At this time even slaves and men imprisoned for debt were enrolled in the Roman legions.50 Some historians have expressed their surprise about the fact that livy does not refer to any freeborn proletarians being recruited by means of tumultuary levies between 216 B.C. and 207 B.C.51 Whatever the explanation for livy’s reticence might be, there can be little doubt that the dilectus of 207 B.C. was a tumultuary levy.52 One of the clues pointing in this direction is livy’s emphasis on the seriousness of the military threat created by Hasdrubal’s advance. As he writes, ‘the dread occasioned by the war had been doubled by the advance of a new enemy into Italy’.53 We are also told that the Senate justified its decision to reject the appeal of the maritime colonies by arguing that the privilege of vacatio was invalid ‘for as long as an enemy remained in Italy’. this justification recalls a famous passage in Appian from which it appears that the declaration of a tumultus Gallicus resulted in the immediate suspension of all exemptions from military service, including that enjoyed by the priests of Rome.54 Various other sources, including the lex Coloniae Genetivae, refer to a distinction between a tumultus Italicus, declared because of an emergency in peninsular Italy, and a tumultus Gallicus, announced in the event of an impending Gallic invasion. the emergency created by Hasdrubal’s advance into Italy obviously belongs to the former category.55 If the colonists’ claim was set aside on the grounds that their privilege of vacatio militiae was invalid in the event of a tumultus, how do we explain the fact that it was decided not to call up any legionaries from the colonies of Ostia and Antium? Were these perhaps the only colonies whose foundation charters contained an explicit reference to an unconditional vacatio militiae or the only ones whose right of exemption had been granted through sworn laws?56 In my view, neither of these explanations is likely to be correct. Why, for instance, should the rights and obligations of the colonists of Antium have differed substantially from those sent out to tarracina, which was founded only 9 years later?57 As Wolfgang Kunkel and Roland Wittman noted more than 15 years ago, the correct explanation might simply be that the Senate was guided by the purely pragmatic consideration that at least two stretches of coastline in close proximity to Rome had to be

49

50 51

52 53 54



the line of argumentation used by the representatives of the seven colonies presupposes that citizens of maritime colonies had not been called up for military service during the campaigns against Pyrrhus or during the first Punic War. liv. 22.57.11 and 23.14.3-4. Rich 1983, 291. He suggests that at this time almost all ingenui living outside the city of Rome were assidui. thus correctly Bellen 1985, 24-25. liv. 27.38. App., B Civ. 2.150.

55

56 57

e.g. Cic., Phil. 5 53 and Lex Col. Genetivae lxII, to be consulted with Crawford 1996, 433 and 435. It does not seem far-fetched to suggest that it was precisely the events of 207 B.C. which prompted the Senate to develop the concept of a tumultus Italicus, although this could be presented as a natural extension of the earlier concept of tumultus Gallicus. the latter view was defended by Maschke 1906, 47. livy dates the colony of Antium to 338 B.C. (8.14) and that of tarracina to 329 B.C. (liv. 8.21).


guarded by full garrisons of maritime colonists.58 If strategic considerations did play a role, the decision to maintain the garrison at Ostia at full strength is entirely unproblematic. the decision to call up legionaries from tarracina but not from Antium would have been more or less arbitrary.59 yet it is not difficult to understand the Senate’s decision to keep the defending force of only one of these colonies at full strength. My conclusion falls into three parts. firstly, I think the tiny size of the allotments received by those sent out to maritime colonies in the 4th and 3rd c. B.C. can be explained only if it is assumed that the settlers were expected to occupy substantial amounts of arable ager publicus. Secondly, I have tried to demonstrate that the decision to provide colonists who had been allotted only two iugera of privately-owned land with a formal vacatio militiae presupposes a property threshold for normal military service falling dramatically short of the requirement of 1,100 libral asses referred to by Cicero and implied by livy’s figure of 11,000 sextantal asses. Although the surviving evidence is hopelessly deficient, the most economical solution for this difficulty is surely Rathbone’s theory that the threshold of 1,100 libral asses was introduced sometime during the 3rd c. B.C., that is to say, some considerable time after the foundation of Ostia, Antium and tarracina.60 thirdly, I have suggested that after Hasdrubal’s march from Spain to Italy the Senate set aside the colonists’ appeal to the privilege of exemption on the grounds that all exemptions from military service had been suspended by the proclamation of a tumultus Italicus. the second of these conclusions can only be supported with fragmentary evidence concerning land prices in early Imperial Italy and with general considerations concerning the historical evolution of the five classes of the comitia centuriata. While this is regrettable, it is also inevitable given the state of the surviving evidence.

References Andreyev, V. 1974. “Some aspects of agrarian conditions in Attica in the fifth to the third centuries B.C.,” Eirene 12, 5-46. Bellen, H. 1985. Metus Gallicus, metus Punicus: zum Furchtmotiv in der römischen Republik (Stuttgart). Bispham, e. 2006. “Coloniam deducere: how Roman was Roman colonization during the Middle Republic?,” in G. Bradley and J.-P. Wilson (edd.), Greek and Roman colonization: origins, ideologies and interactions (Swansea) 73-160. Botsford, G. 1909. The Roman assemblies from their origin to the end of the Republic (New york). Brunt, P. A. 1987. Italian Manpower, 225 B.C.-A.D. 14 (2nd edn., Oxford). Campbell, J. B. 2000. The writings of the Roman land surveyors: introduction, text, translation

58 59

Kunkel and Wittman 1995, 432, n. 1. toynbee stresses the strategic importance of Antium, but the same could be said of all or almost all harbor

and commentary (london). Cancellieri, M. 1990. “Il territorio Pontino e la Via Appia,” ArchLaz 10, 61-72. Capogrossi Colognesi, l. 2012. Padroni e contadini nell’Italia repubblicana (Rome). Càssola, f. 1988. “Aspetti sociali e politici della colonizzazione,” DialArch 6 (2), 5-17. Chouquer, G. et al. 1987. Structures agraires en Italie centro-méridionale: cadastres et paysages ruraux (Rome). Cornell, t. J. 1995. The beginnings of Rome. Italy and Rome from the Bronze Age to the Punic Wars (c. 1000-264 B.C.) (london). Crawford, M. H. 1985. Coinage and money under the Roman republic: Italy and the Mediterranean economy (Berkeley/los Angeles). Crawford, M. H. (ed.) 1996. Roman statutes, vol. 1

60

towns and landing places (toynbee 1965, 183). for a variant of Rathbone’s theory which also obviates the difficulties posed by liv. 27.38 see note 60.




(london). Daly, G. 2002. Cannae: the experience of battle in the Second Punic War (london). De Neeve, P. W. 1984. Colonus: private farmtenancy in Roman Italy during the republic and the early principate (Amsterdam). De Neeve, P. W. 1985. “the price of agricultural land in Roman Italy and the problem of economic rationalism,” Opus 4, 77-109. De Sanctis, G. 1972. “Per rinunziare l’acrisia (rev. edn.),” in G. De Sanctis (ed.), Scritti minori, novamente editi da Aldo Ferrabino e Silvio Accame, vol. 3 (Rome) 282-360. Dilke, O. A. W. 1971. The Roman land surveyors: an Introduction to the Roman Agrimensores (Newton Abbot). Dumont, C. 1844. Essai sur les colonies romaines (Brussels). Duncan-Jones, R. 1982. The economy of the Roman Empire: quantitative studies (2nd edn., Cambridge). erdkamp, P. 2011. “Soldiers, Roman citizens, and latin colonists in Mid-Republican Rome,” AncSoc 41, 109-46. Gabba, e. 1976. “the origins of the professional army at Rome (english transl.),” in e. Gabba (ed.), Republican Rome, the army, and the allies (Oxford) 70-131. Gabba, e. 1977. “Considerazioni sulla decadenza della piccolo proprietà contadina nell’Italia centro-meridionale del II sec. a.C.,” Ktèma 2, 26984. Gabba, e. 1978. “Per la tradizione dell’heredium romuleo,” RendIstLomb 112, 250-58. Hanson, V. 1995. The other Greeks: the family farm and the agrarian roots of Western civilization (New york). Harris, W. V. 1990. “Roman warfare in the economic and social context of the fourth century B.C.,” in W. eder and C. Ampolo (edd.), Staat und Staatlichkeit in der frühen römischen Republik (Stuttgart) 494-510. Hinrichs, f. t. 1974. Die Geschichte der gromatischen Institutionen: Untersuchungen zu Landverteilung, Landvermessung, Bodenverwaltung und Bodenrecht im römischen Reich (Wiesbaden). Hölkeskamp, K.-J. 1987. Die Entstehung der Nobili-



tät: Studien zur sozialen und politischen Geschichte der römischen Republik im 4. Jhdt v. Chr. (Stuttgart). Huschke, P. 1838. Die Verfassung des Königs Servius Tullius als Grundlage zu einer römischen Verfassungsgeschichte (Heidelberg). Kaser, M. 1949. Das altrömische Ius: Studien zur Rechtsvorstellung und Rechtsgeschichte der Römer (Göttingen). Kroll, J. H. 1977. “An Archive of the Athenian Cavalry,” Hesperia 46, 83-140. Kunkel, W. and R. Wittman 1995. Staatsordnung und Staatspraxis der römischen Republik, zweiter Abschnitt: die Magistratur (Munich). lackner, e. M. 2008. Republikanische Fora (Munich). lambert, S. D. 1997. Rationes centesimarum: sales of public land in Lykourgan Athens (Amsterdam). lo Cascio, e. 1988. “Ancora sui censi minimi delle cinque classe ‘serviane’,” Athenaeum 76, 273302. Magdelain, A. 1995. De la royauté et du droit de Romulus à Sabinus (Rome). Marottoli, l. 1979. Leges sacratae (Rome). Maschke, R. 1906. Zur Theorie und Geschichte der römischen Agrargesetze (tübingen). Mason, G. G. 1992. “the agrarian role of coloniae maritimae: 338-241 B.C.,” Historia 41, 75-87. Mohr Mersing, K. 2007. “the war-tax (tributum) of the Roman republic: a reconsideration,” ClMed 58, 215-35. Mommsen, t. 1887. Römisches Staatsrecht (leipzig). Nicolet, C. 1976. Tributum: recherches sur la fiscalité directe sous la république romaine (Bonn). Nicolet, C. 1980. The world of the citizen in Republican Rome (Berkeley/los Angeles). Northwood, S. J. 2008. “Census and tributum,” in l. de ligt and S. J. Northwood (edd.), People, land, and politics: demographic developments and the transformation of Roman Italy 300 B.C. - A.D. 14. (leiden) 257-70. Oakley, S. P. 1997. A commentary on Livy: books VIX, vol. 1: Introduction and book VI (Oxford). Oliver, G. J. 2006. “Polis economies and the cost of the cavalry in early Hellenistic Athens,” in P. G. Van Alphen (ed.), Agoranomia: studies in


money and exchange presented to John H. Kroll (New york) 109-24. Pelgrom, J. 2012. Colonial landscapes: demography, settlement organization and impact of colonies founded by Rome (Ph.D thesis, leiden university). Rathbone, D. 1993. “the census qualifications of the assidui and the prima classis,” in H. SancisiWeerdenburg (ed.), De Agricultura: In Memoriam Pieter Willem de Neeve (1945-1990) (Amsterdam) 121-52. Rathbone, D. 2003. “the control and exploitation of ager publicus in Italy under the Roman Republic,” in J.-J. Aubert (ed.), Tâches publiques et entreprise privée dans le monde romain (Neuchâtel) 135-78. Rich, J. 1983. “the supposed Roman manpower shortage of the later second century B.C.,” Historia 32, 287-331. Rich, J. 2008. “lex licinia, lex Sempronia: B.G. Niebuhr and the limitation of landholding in the Roman Republic,” in l. de ligt and S. J. Northwood (edd.), People, land, and politics: demographic developments and the transformation of Roman Italy 300 B.C. - A.D. 14. (leiden) 519-72. Roselaar, S. t. 2009. “Assidui or proletarii? Property in Roman Citizen Colonies and the vacatio militiae,” Mnemosyne 62, 609-23. Roselaar, S. t. 2010. Public land in the Roman Republic: a social and economic history of Ager Publicus in Italy, 396-89 BC (Oxford). Rosenstein, N. 2004. Rome at war: farms, families, and death in the Middle Republic (Chapel Hill/london). Ruperti, G. 1843. Regierung und Verwaltung des römischen Staats (Hannover). Rüpke, J. 1990. Domi militiae: die religiöse Konstruktion des Krieges in Rom (Stuttgart). Salmon, e. t. 1969. Roman colonization under the

Republic (london). Santi, C. 2004. Alle radici del sacro: lessico e formule di Roma antica (Rome). Scheid, J. 2003. An introduction to Roman religion (edinburgh). Scheidel, W. 2005. “Real slave prices and the relative cost of slave labor in the Greco-Roman world,” AncSoc 35, 1-17. Smith, C. J. 2006. The Roman clan: the gens from ancient ideology to modern anthropology (Cambridge). Spence, I. 1993. The cavalry of Classical Greece: a social and military history (Oxford). thiel, J. H. 1946. Studies on the history of Roman sea-power in Republican times (Amsterdam). thomas, y. 1988. “Sanctio: les defenses de la loi,” L’écrit du temps 19, 66-84. thomas, y. 1993. “De la ‘sanction’ et de la ‘sainteté’ des lois à Rome: remarques sur l’institution juridique de l’inviolabilité,” Droits: revue francaise de theorie juridique 18, 135-51. toynbee, A. J. 1965. Hannibal’s legacy: the Hannibalic War’s effects on Roman life, vol. 1: Rome and her neighbours before Hannibals’ entry (Oxford). trundle, M. 2010. “light troops in Classical Athens,” in D. M. Pritchard (ed.), War, democracy and culture in Classical Athens (Cambridge) 139-60. Van Wees, H. 2004. Greek warfare: myths and realities (london). Von fritz, K. 1976. “leges sacratae und plebei scita,” in K. von fritz (ed.), Schriften zur griechischen und römischen Verfassungsgeschichte und Verfassungstheorie (Berlin/New york) 374-87. zevi, f. 1996. “Sulle fasi più antiche di Ostia,” in A. Claridge and A. Gallina zevi (edd.), ‘Roman Ostia revisited’: archaeological and historical papers in memory of Russell Meiggs (london) 69-89.




PART II COLONIAL LANDSCAPES. COLONISTS AND NATIVES SHAPING THE URBAN, NATURAL AND SOCIAL ENVIRONMENT


Gellius, Philip II and a proposed end to the ‘model-replica’ debate Jamie Sewell

Introduction The title of this paper refers to Gellius’ hackneyed passage describing Rome’s colonies as “small copies and likenesses” (quasi effigies parvae et simulacra) of their mother city.1 These words have exerted tremendous terminological influence on, or perhaps have even given rise to a modern debate: the degree to which Rome’s Republican colonies physically emulated their mother city. Proponents of Gellius’ statement, such as Edward Togo Salmon, have described the colonies’ “uniform tendency to imitate Rome” (cf. Pelgrom and Stek in this volume for the development of Salmon’s position on this matter),2 and Frank Brown, the excavator of Cosa, saw the prototype of that Latin colony as being “Rome itself”.3 Subsequently, Gellius has been invoked on innumerable occasions in discussions on the urban topography of the colonies.4 As a consequence, popular terms derived from Gellius, such as “little Romes”, have been used to describe the colonies.5 More recently, the basis for what Henrik Mouritsen describes as “the simple model-replica perception of the relationship between founder-city and colony”6 has, justifiably, received increasing critical attention,7 to the point that the discussion is now on the opposite tack. Perceived ‘Gellian’ and ‘normative’ models of Mid-Republican colonization have been criticized from various historiographical and archaeological perspectives.8 Although many modern works which cite Gellius imply or state directly that he compared the colonies to the urbs (in a description of a speech by Hadrian), in fact, as Paul Zanker points out,9 he actually likens the colonies to the greatness (amplitudo) and majesty (maiestas) of the Roman people. Thus there was never a basis for employing Gellius’ description in modern comparisons of the built environments of Rome and its colonies (cf. Sisani and Torelli in this volume). But a tradition to do so has developed nonetheless which has served as a distraction. To argue whether the colonies were or were not replicas,10 likenesses or images of Rome does not help us understand colonial towns, or their relationship to Rome, or indeed Roman colonization. There is an

1

2 3 4

5

Gell., NA 16.13.9. “Small copies and likenesses” is how Zanker 2000, (41) translates it; Bispham 2006, 79: “small representations and images of a sort”. Salmon 1969, 18: “miniatures and […] reproductions”. I am grateful to Michael Crawford and Mario Torelli for their comments regarding the meaning and translation of this passage. Salmon 1969, 18. Brown 1980, 12. For discussions on how Gellius has been interpreted: Bispham 2000, 157-58; id. 2006, especially 78-85. Examples of how Gellius’ statement has been employed in discussions: Richardson jr. 1957, 49; Torelli 1988; id. 2007, 164 and 179; Zaccaria Ruggiu 1995, 33-38; Coarelli 1998, 57; id. 2005, 24-28; Mouritsen 2004. Cf. Bispham 2006, 75; Diana Kleiner’s Yale lecture on

6 7 8

9 10

Roman architecture of the towns in the Western Provinces, delivered on 14th April 2009, was entitled “Making Mini Romes on the Western Frontier”: http://www.youtube.com/watch?v=_OTCjTceIAo (viewed 09.05.11). Mouritsen 2004, 64. Torelli 1988; Fentress 2000; Zanker 2000, 40-41. Compare the diverse approaches in: Zanker 2000; Crawford 1995; Bispham 2000; id. 2006; Mouritsen 2004; Bradley 2006; Pelgrom 2008; Sewell 2010, 67-85. Cf. Coarelli 2005. Zanker 2000, 4. Although I choose to use the term ‘model-replica debate’ for the sake of brevity, the word ‘replica’ is not an accurate translation of any part of Gellius’ description (see note 1 above).




innate logic in believing Rome desired to reproduce itself in its colonies, or in a perceived desire for the colonists to surround themselves with the familiar by creating facsimiles of the metropolis. Yet this is an overly simplistic view; it potentially masks important historical processes, some of which I hope to elucidate in this paper. By doing so, and in acknowledgement that today most would agree the model-replica analogy is outdated, an opportunity arises to demonstrate how far thinking on this aspect of colonization has come in recent years. The built environment of the Mid-Republican colonies’ primary settlements is the main focus here, looking at the various Roman and foreign influences to have shaped their physical forms, some of which can be traced to areas outside Italy. If the Romans were willing to adapt foreign ideas in their development of suitable types of colonial settlement, perhaps they might have looked to others when considering the best way to proceed in other aspects of colonization. To end with, then, we will examine the similarities and the contemporaneity of the strategies for expansion attributed to Rome and to Philip II of Macedonia and explore a possible link between them.

Adapting Roman models To reveal how the replication concept distorts our understanding, one needs to emphasize the diverse functions and capabilities of the metropolis and its colonial foundations.11 There were certain powers Rome possessed that she surely avoided passing on to its colonies. If the colonies had had the ability to declare war, for example, or to augment their territories unilaterally, as Rome was able to do, the system would likely have collapsed. For Roman colonization to function the colonies had to have an extremely limited ability to form their own foreign policies. Necessarily, colonial political systems had to recognize their permanent subordination to Rome in particular ways, and the condition of being permanently subordinate was obviously not replicated from the Res Publica of Rome (cf. Sisani in this volume). With this caveat, I will now review some of the physical aspects of the Mid-Republican Latin colonies that were arguably derived from the capital but, given the specific colonial context, they should be understood as adaptations rather than replications. By accepting that certain Roman models were suitable for the colonies and others not, it permits consideration of the reasons for this selectivity (cf. Stek in this volume). The political system of Rome was adapted for the Latin colonies to some unknown degree, attested by the use of Roman magisterial titles,12 and arguably represented by the comitium-curia complexes on colonial forums (fig. 1).13 Controversy surrounds the issue of whether the cavea of the Comitium on the forum Romanum had a similar circular form to that of the Latin colonial assembly-places, as reconstructed by Fillipo Coarelli.14 Insufficient of the original has survived to perhaps ever allow a reconstruction that everyone can live with, and arguing that the original had the form of its derivatives is circular reasoning.15 In terms of our understanding of colonization, this debate has somewhat overshadowed a more important issue since, interestingly, no one seems to disagree with the labeling of the colonial structures as comitium-curia complexes. Implicitly, then, it is accepted that they represent adaptations of the metropolitan original. This is due, almost certainly, to the argument first presented by Lawrence Richardson Jr.,16 who identified the structure at Cosa as a comitium-curia complex through its similarities with some essential aspects of the form and location of the metropolitan original known from literary sources.17 Like the colonial examples,

11 12 13

14



Sewell 2010, 83. Salmon 1969, 85-87. Lackner 2008, 258-65 for an up-to-date summary of the archaeological evidence for Latin colonial comitia and references to the debate on the character of the structure at Rome. Coarelli 1983, 139-60; id. 1985, 11-21; id. in LTUR 1 (1993) 309-14. s.v. “Comitium”. Contra Carafa 1998, 150-51; Amici 2004-5, 359.

15 16 17

Mouritsen 2004, 40. Contra Coarelli 2005, 25-26. Richardson jr. 1957, 49. See note 13. Other discussions and interpretations of these sources: Sjöqvist 1951; Krause 1976; LTUR 1 (1993) 331-32, s.v. “Curia Hostilia” (F. Coarelli). Not all the sources are relevant to one particular building phase.


the Comitium was a stepped construction, situated on the northern side of the Forum and orientated towards the south. Overlooking it, directly to the north, was the Curia Hostilia which, judging from the later Curia Iulia, might have been a rectangular structure. Furthermore, the steps of the colonial caveæ do not seem generally to be wide enough to sit on,18 indicating that citizens stood at assemblies which, according to Cicero, was the custom at Rome.19 It is easy to imagine how the archaeological discovery of these colonial structures nourished the model-replica analogy. Thus the argument that the architecture of Latin colonial comitium-curia complexes was, to some degree, derived from the metropolitan original, remains strong. The processes that led to the construction of Latin comitium-curia complexes to begin with are particularly poorly understood. In some cases they appear to have been constructed some time after the colonies were founded (although the dating problems are considerable). Did the colonies have to or did they choose to build these structures? As yet, all the excavated northern sides of Mid-Republican Latin colonies have revealed a comitium-curia complex. Does this reflect a program, or peer polity interaction, or some other obscure process? We are not even close to understanding this, but since this architecture was apparently derived from Rome, the processes involved seem to have been centripetal in character. Latin colonies were set up as self-governing states and, for this reason, Salmon argued, they “must have been free to choose any constitution they liked, but they preferred to order themselves according to the forms of Roman political organization, and because of this readiness, if not positive eagerness, to imitate Rome they present a fairly uniform picture.”20 Here we witness Gellian terminology in action, channeling trains of thought. Conceivably, Latin colonies might have been as “separated politically” from Rome as Salmon maintained,21 but more recent epigraphic discoveries reveal that Rome had a greater say than previously thought (cf. Pelgrom in this volume). We now possess direct testimony to the actions of a triumuir, one of the board of commissioners sent by Rome to found colonies. In this case, his activities seem to relate to the re-founding of Aquileia in 169 B.C., and our commissioner, T. Annius, was responsible for composing the colony’s laws and, on three separate occasions, co-opting its senate.22 In this case, Rome’s representative was proactive in the development of Aquileia’s political system. Why? The colony came into existence through a decision taken and supported in Rome for the purpose of serving Rome’s best interests; its role was initially determined by the metropolis. Based on this premise, no wonder the laws of a colony were given to it. Perhaps, but the problem with this argument is its demotion of Annius to a pawn sent by the metropolis, whose personal agency counted for little. After all, the inscription records that he composed (composiuit) the laws, reflecting an element of self-determination. Moreover, the commissioner’s choices might have been influenced by the character and concerns of the new colonial community (cf. Bradley in this volume). That the local situation had a profound influence on the process of founding a new colony is reflected in the urban centres themselves. The commissioners were also responsible for taking the first steps to establish the physical settlement.23 In this task a high degree of pragmatism was required due to the unique nature and circumstances of each colonial foundation. Variety characterizes the urban centres, ranging in size from as little as 7.5 ha to as much as ca. 200 ha,24 with some placed on inland hilltops, others on coastal flood plains; some were founded ex novo, others within pre-existing settlements. Individuality has been highlighted as a quality associated with the colonies’ religious landscapes,25 and with the ethnic mix of their populations.26 Thus one should expect to see the built environment of each colony develop in a unique manner. Even

18 19 20 21 22

Krause 1976, 59. Cic., Flac. 15-17. Salmon 1969, 85-86. Cf. Mouritsen 2004, 64; Coarelli 2005, 24; Sewell 2010, 81-85. On the discovery of the inscription: Zaccaria 1996, cols. 182-84. Discussions on the inscription: Galsterer 2006, 53; Bispham 2007, 37, 152-56; Sewell 2010, 84-85.

23 24 25 26

Gargola 1995, 71-87. Lackner 2008, 240-44. Bispham 2006. Bradley 2006.




FIG. 1. Long insulae bordering Latin colonial fora, reconstructed as elements of their original town planning. From left to right: Fregellae (after Coarelli 1998, tav. III, 4), Alba Fucens (after Mertens 1969, plan I), Paestum (after Neutsch 1956, fig. 115), Cosa (after Brown, Richardson and Richardson 1993, 6, fig.1).

though the comitium-curia complex seems to have been necessary architecture for a Latin colony, no two are the same. Not only do they vary in size, but the architectural solutions to the curia buildings vary considerably, and they also changed over time in unique ways. This is particularly interesting, for Henrik Mouritsen questioned the idea that the curious post-pits found on all Latin forums represent a cultural practice derived from Rome because of the heterogeneity of the form, sizes and layouts of the pits.27 Yet the curia buildings on Latin forums are also each unique in size, layout and architectural development, but here there seems to be no dissent to the idea that they were derived from the Curia Hostilia. Uniqueness was also expressed by the Latin colonies in their ability to mint their own coins; a reflection of their individuality as a politico-religious community (cf. Bradley in this volume). Since the concept of the post-338 B.C. Latin colonies was devised by Rome, and since part of this concept included allowing them to produce their own coinage, there is strong argument for it having been Rome’s intention for each colony to have a unique identity. And yet, since the urban centres were clearly planned, it is possible to pick out features that Latin colonies of the 4th and 3rd c. B.C. shared with one another in their town planning. It is highly unlikely these similarities are coincidental, meaning that there appears to have been a set of conventions for planning Latin colonies. Fregellae (328-13 B.C.) had an elongated forum, and all known subsequent Mid-Republican forums had proportions of between 1:2 and 1:3 (fig. 1).28 Similarly long Greek agorai are not unheard of, but since there is no known Latin colony without a long forum, it suggests that there was a planning convention to design them this way. Another distinctly Roman planning convention was to create long narrow insulae which flanked the sides of Latin colonial forums (fig. 1).29 This can be reconstructed at Fregellae, Alba Fucens, Paestum and might have been an element of Cosa’s original 3rd c. B.C. plan. These blocks appear to have been destined exclusively for the forum’s public architecture, and often the majority of their surface areas came to be taken up with rows of tabernae. The success of this model – already developed by the late 4th c. B.C. – is that it seems to have influenced the planning of the larger Roman colonies of the 2nd c. B.C., for the same scheme has been plausibly reconstructed by Eva-Maria Lackner at Luna.30 Fregellae was probably the first ex novo Latin colony founded,31 and thus the first ever town the Romans

27 28 29 30



Mouritsen 2004, 64. Conventi 2004, 197-200. Sewell 2010, 58-62. Lackner 2008, 117 and 356-57.

31

All other Latin colonial sites, including those founded before 338 B.C., have revealed that they existed as settlements before the dates when our sources indicate the colony was set up (Lackner 2008, 240-43). Ostia is discussed below.


set up from scratch. It is all the more striking to see a town-planning convention at Fregellae, established during the infancy of the Roman town-building tradition, survive the test of time and continue to be employed by planners through to the 2nd c. B.C. As we will see, the Romans also adapted a great deal from Greek town-planning practices, but there appears to be no Greek precedent for creating long narrow insulae along the edges of agorai. This therefore lends Roman planners a foresight not seen in the Greek context, since the sense behind inscribing these blocks into the town plan seems to be related to an anticipation of how forum architecture was going to develop in the future. Whether these planning ‘conventions’ might be more accurately described as ‘customs’ or indeed ‘policy’ is difficult to determine, but they are likely to have been derived from the layout of the forum Romanum, which also possessed elongated proportions and was lined with tabernae.32 This is suggestive of colonial forums having been designed for political, ceremonial and commercial activities adapted from the metropolis. So a well-defined model for how Latin colonial forums should be planned had emerged by the later 4th c. B.C. that was apparently not substantially deviated from in subsequent foundations during the 3rd c. B.C. and beyond. The original town planning is a rare contemporary source of data that can inform us about important aspects in the development of Roman colonization. It was a template for the subsequent physical realization of the settlement. The planning of the town was almost certainly undertaken during the threeyear tenure of the triumviri. In most if not all cases the earliest dated architectural feature in each Latin colony is the fortifications. Both Livy and Polybius refer to the existence or construction of 3rd c. B.C. colonial town walls during the tenure of the triumviri.33 Determining the line of the curtain wall and planning the town itself were interrelated processes, testified by Greek literary and epigraphic sources relevant to the 4th and 3rd c. B.C.,34 and the Romans certainly learnt much from the Greeks. So as well as necessarily having to work pragmatically due to the specific circumstances of each foundation, it appears as though the triumviri (or their subordinates) also arrived at each new site with apparently fixed concepts of how certain aspects of the new urban center were to be planned. The physical development of settlements planned in this manner was consequentially influenced by certain parameters set at the start.

Adapting foreign models. By suggesting that aspects of the Latin colonies’ urban design were derived from Rome itself, this would seem to support for the model-replica argument. In fact, this leads us directly to another reason for the redundancy of model-replica analogies, for although the urban centres were clearly influenced by metropolitan precedents, they also comprised planning conventions typical of contemporary Greek townplanning.35 The model for the urban centres comprised the selective incorporation both of foreign elements and others derived from the metropolis.36 Although a great deal of confidence is reflected in early Roman town planning, there can be no doubt they sought foreign expertise at some point. Although an obvious point, the orthogonality of colonial town plans was not derived from Rome, and despite orthogonal planning being already a centuries-old practice by the later 4th c. B.C., the signs of contemporary Greek influence are found in the length of the Latin colonies’ insulae. With a few exceptions, the proportional length of Greek insulae shortened over the centuries, and the length of those from the Latin colonies are generally compatible with those of Greek towns founded from the second half of the 4th c. B.C. onwards. Other signs are to be found in the fortifications: at Alba Fucens, Paestum and Cosa we see the application of 4th c. B.C. Greek innovations, such as the arched gate, the gate court, the portcullis and, at Cosa, towers which

32

33

Coarelli 1985, 140-52; LTUR 5 (1999) 14-15, s.v. “Tabernae Novae” (E. Papi). Liv. 21.25, 3-4; Polyb. 3.40.9-10.

34 35 36

Merritt 1935, 369 no. 1; Lib., Orat. 11.88-92. Cf. Terrenato 2008, 253-54. Sewell 2010, 21-86.




contained artillery. Early forums, such as that at Alba Fucens, are even skirted by streets on all four sides, mimicking a Greek agora (cf. Liberatore in this volume). Yet selectivity was at work, for other Greek models found no place on Latin colonial forums. Since the Romans clearly had knowledge of the Greek architectural vocabulary, the absence of the quintessentially Greek stoa on the forums of Roman towns during the 4th and 3rd c. B.C. stands out particularly,37 as this period marks the zenith of the stoa’s popularity in other Hellenistic contexts. Instead, Latin colonial forums came to be lined with tabernae. Arguably, in this selective integration of both foreign and metropolitan precedents, a Roman attempt to create a Latin colonial urban model may be perceived. Turning to the much smaller colonies of Roman citizens, it appears now as if they were adapted from a specific foreign model. Ostia might have been the first planned and fortified settlement built de novo by the Romans. The earliest material evidence from the site is dated between 380 and 340 B.C.38 It has a distinct rectilinear form with a gate in the center of each wall, with the main streets forming a crossroad at the center of the settlement. From what is known of the castrum at Minturnae, founded in 296 B.C., its plan was probably very similar. Since the 1960s French archaeologists have been comparing the coloniae maritimae to Olbia, a permanently manned Greek fort found near Hyères-Les-Palmiers in Provence, established by Massilia in the second half of the 4th c. B.C.39 It was located directly on the coast, was square in plan and is exactly the same size as Minturnae (with each side being 160 by 65 m long). Also like Minturnae, towers were situated on its corners. Its planning is identical to that of Ostia in as much as its two widest streets were set perpendicularly to form a crossroad at the settlement’s center and, just like Ostia, the one running east-west is slightly wider than the one running north-south (fig. 2).40 Ostia had no forum, and Olbia had no agora or any architecture representative of self-government. Olbia was situated ca. 66 km to the east of Massilia and is considered to have been a small permanently manned outpost of Massiliot citizens situated so as to guard the coast. Other rectilinear Greek forts are known – the shape was simply a response to building on level terrain – but the exceptional degree of similarity between Olbia, Ostia and Minturnae in terms of date, location, size, form, function and their permanent populations dependent on a metropolis, makes a direct link between Roman and Massiliot strategies for coastal defense almost beyond question. According to Justin, a foedus aequum existed between Rome and Massilia by the time these settlements were founded.41 By the 4th c. B.C. the Phocaeans of the western Mediterranean were old hands at building compact coastal colonies, since they had been at it since the early 6th c. B.C., starting with Massilia. Subsequently they founded other settlements along the shores of the Iberian Peninsula and of Liguria.42 This makes it more than likely that the model for Rome’s coastguard colonies was provided by Massilia. Additional support for this hypothesis is provided by the distinctive intersection of the two main streets at the center of Olbia, since rather than being expedient to the needs of a coastal fort, it seems to have been a distinctive feature of Massiliot townplanning. Strabo describes Nicaea,43 another late 4th c. B.C. foundation of Massilia, as being square as well, but with a wall circuit 16 stadia long, equaling a surface area of ca. 55 ha, making it more than 20 times larger than Olbia. Yet Strabo states that Nicaea had streets set at right angles and 4 gates, all 4 of which could be seen from a stone set up in the gymnasion, presumably situated at the center of the town. To return to the model-replica debate, in terms of the physical settlement, the similarity between Olbia

37 38

39 40



Cf. Sewell 2010, 64-67. http://www.ostia-antica.org/intro.htm (viewed 15.05.11). A late 5th c. B.C. date is favoured by Coarelli 1992, 25 and Torelli 2007, 162. Coupry 1964; id. 1986; Verdin 1997. At Olbia the widths are 5.2 m and 4.2 m: ibid., 429. At Ostia, 8.7 m and 7.35 m: Brandt 1985, 3334, 41.

41

42 43

Just., Epit. 43.5.8-10. Other reports of strong ties between Rome and Massilia: Just., Epit. 43.3.4, 43.5.3; Liv. 34.9.10; Strab. 4.1.5. Garbini 1996, 127-30. Strab. 12.4.7.


FIG. 2. A comparison of Brandt’s reconstruction of Minturnae (left) with the actual state of remains at Olbia (right) (Minturnae: after Brandt 1985, fig. 24; Olbia: after Verdin 1997, fig. 1).

and Minturnae is as close as I have seen to actual replication, but here again it would serve us better to understand it as an adaptation. Yet the success of the Roman coastguard colony seems to be related to it having been based on the cumulative experience of others. This case also highlights the importance of town planning as a contemporary testimony to important historical processes. But with the model for Rome’s first citizen colonies being situated in an area that is not even contiguous to the Italian peninsula, it serves to reinforce the point that our comprehension of Roman colonization is likely to be enhanced when placed in a Mediterranean context, and it also highlights the degree of interconnectivity in the region. Evidence for the degree of Roman confidence in planning towns from the later 4th c. B.C. onwards is mounting, for it seems that a number of existing settlements annexed by Rome received an orthogonal street-grid as a consequence of their incorporation into the Roman state. Above all, the purpose of replanning a town was to prepare it for future development, signifying that those responsible were in possession of an ideological concept of how the town should develop in the future. Paolo Sommella lists 11 towns he believes were affected in this way,44 and a recent study proposes that several Campanian towns, including Capua and Cumae, were either partly or completely re-planned and given new defenses between the end of the 4th and beginning of the 3rd c. B.C.45 These changes were possibly linked with a re-organization of the settlement pattern in their territories, a reminder that if the Romans had developed a model for how to physically organize towns, there is every likelihood they had likewise developed an organizational model for territoria as well; a colonia originally was, by definition, a settlement of cultivators.

44

Sommella mentions Alatri, Aletrium, Aquinum, Arpinum, Artena, Ferentinum, Fundi, Privernum, Saepinum, Teanum Sidicinum and Venafrum (Sommella 1988, 25-26).

45

Rescigno and Senatore 2009.




A Macedonian connection? By focusing on the trajectory of Roman urbanism in the later 4th c. B.C., it seems to indicate that a considerable degree of thought had gone into the development of colonial urban models by this time. In a consideration of what might have triggered this flurry of inventiveness and the Romans’ acceptance of, or even hunger for related foreign expertise, my attention has inevitably been drawn to the constitutional changes of 338 B.C.46 Given the demonstrable willingness of the Romans to adapt the ideas of others in relation to urbanism, I conclude by turning away from urban topography to ask whether Rome might have been influenced by other foreign ideas in the early development of its colonization model. The following analysis is based on a premise that in 338 B.C., or thereabouts, the Romans thought carefully about questions linked to the formation of a ‘colonial policy’: I) How should territorial expansion (whether won through military victory or by other means) be consolidated to ensure its permanency? II) How should this augmented territory be defended, while maintaining political stability and propagating the capability for further conquest? If the Romans considered these issues at this time, and given their willingness to adapt the ideas of others, what successful contemporary models could they have drawn upon? The reason to suppose that these matters were reflected upon in 338 B.C. can be found in the constitutional changes of that year, following the Latin revolt. This year is thus traditionally marked as a turning point in the history of Roman expansionism, but it was also a year which witnessed the “end of Greek liberty”,47 at least for many poleis of mainland Greece, as a result of the battle of Chaeronea. In just 20 years Philip II had, in the words of his latest biographer, “transformed Macedonia from a disunited, weak backwater into an imperial power”.48 It has been argued that, during his reign, Philip doubled the area of his kingdom and trebled the size of his army.49 Given the uproar among the poleis caused by Philip’s growing power, surely the Romans had got wind of this heated political situation. Hence we find Philip’s strategies for expansionism reaching the pinnacle of success at the very moment the Romans are engaged in a constitutional reform now considered as a momentous step for the subsequent conquest of peninsular Italy. This is reason enough, I believe, to evaluate whether the Romans might have adapted some of Philip’s strategies. Before doing so, some issues should not be overlooked: by choosing to examine Mid-Republican colonization from the perspective of questions I) and II), it implies that expansionism was its driving force, but the sources indicate there were other reasons for the founding of colonies (cf. Terrenato in this volume).50 Other potential influences on Rome’s colonization model will not be discussed below, such as the experience with the Latin League.51 It should also be emphasized that certain commonalities in the Roman and Macedonian situations are not attributable to potential cultural contact in the later 4th c. B.C. On the geopolitical level, the area of Philip’s enlarged Macedonian kingdom and the neighboring regions which came under his influence comprised both inland and coastal regions. Nomadic, tribal pastoralists in mountainous areas were under the king’s sway, as were urbanized agriculturalists inhabiting fertile flood plains. The variety of ecologies, economies, landscapes and settlement forms (including Greek poleis) within Philip’s Macedonian kingdom and its allied territories is remarkably similar to that encountered by the Romans during their territorial expansion of the 4th and 3rd c. B.C.52 Another comparability is to be found in a primary cause of both Roman and Macedonian expansionism, being a strong expectation in the mentalities of both cultures that when an enemy was beaten in battle, generally, his land would be forfeited permanently to the victor;53 a prerequisite

46 47 48 49



Cf. Cornell 1995, 347-52; Toynbee 1965, 129-41. Worthington 2008, 136. Ibid., 1. Hammond 1989, 152 and 177.

50 51 52 53

Cf. Bradley 2006; Patterson 2006. Cf. Cornell 2000. Cf. Coarelli 1992, 23-24; Dench 2005, 166 and 216-17. Anson 2008, 17. Cf. Dench 2005, 163-64.


for the consideration of questions I) and II). These arguably coincidental similarities are striking, so even if the reader chooses to discount the possibility of a link between Macedonian and Roman expansionist strategies, the following comparison might still prove thought provoking. Naturally, Philip’s success was linked to his extraordinary personal abilities in diplomacy, military leadership and administration,54 and to his much discussed military reforms.55 Thus, in an evaluation of whether a Rome’s model for expansionism was influenced by Philip’s, it is most appropriate to examine scholarly opinion on how Philip dealt with questions I) and II) above.56 Where we find a strong resonance is in how land was put to use. A key element of Philip’s success was the practice of granting land in return for either infantry or cavalry service.57 Through both viritane assignments and by grants to those relocated in existing or newly founded cities, land was improved by first-time landowners who owed their new prosperity, and thus also their loyalty, to the king. Many former landless could now afford to arm themselves in order to attain a place in the Macedonian army and thus have an opportunity to defend their newly acquired property.58 So in the Macedonian context, scholars identify a direct relationship between Philip’s reconfiguration of land and the growing strength of his army. The (re)development of land was an essential aspect of the Roman conquest of Italy and there are many signs that this was being undertaken by the late 4th c. B.C.59 One of the main results of Latin colonization was the creation of new landowners who augmented Rome’s army with contingents of financially independent colonial troops.60 In the late 4th and early 3rd c. B.C., many Campanian towns belonging to communities which received Roman citizenship without the vote are believed to have been re-planned and their territories reorganized.61 In this case, receiving the citizenship was accompanied not only by obligatory military service in the legions,62 but also apparently by a reconfiguration of the new citizens’ land. We also hear of viritane grants to Roman settlers in the early 3rd c. B.C. from land appropriated from the vanquished.63 Although little is known about Macedonian royal land, some parallel may be drawn with ager populus populi Romani, for the king could rent it or grant it revocably to individuals.64 Founding and redeveloping towns, a phenomenon closely associated with the conquest of Italy, was a strategy undertaken by Philip on a scale greater than any of his predecessors.65 To this end “he transplanted people and cities hither and thither”.66 Here we are getting close to the familiar mechanisms of Roman colonization. Yet the sources on Philip resettling populations state or imply that they were relocated

54 55 56

57

58

59

60 61

Worthington 2008, 1-5. Ibid., 26-32. The character of the Macedonian and Roman armies and their tactics were notably diverse, and when they met, the Romans did rather better (Harl 2008). Billows 1995, 132-37; Hatzopoulos 1996, 268-71; Anson 2008, 23-25. Miltiades B. Hatzopoulos emphasizes Philip’s innovation of dramatically increasing the numbers of the land-owning class (Hatzopoulos 1996, 270-71). Since the newcomers owed loyalty to the king for their new possessions, this move simultaneously reduced the power of the old aristocracy who had caused his predecessors so many problems. Purcell 1990, 14-20; Dench 2005, 163-64. Cf. Harris 1979, 60-62. Gabba 1988. Rescigno and Senatore 2009.

62 63 64

65

66

Cf. Dench 2005, 122-23. Coarelli 1992, 28-29; Cornell 1995, 362. Hatzopoulos 1996, 205. Richard A. Billows argues that, during the Hellenistic period, the king also nominally had rights over land granted to Macedonian settlers within the territories of towns (Billows 1995, 134-37). Very few sources relevant to royal land under Philip exist, so comparisons with ager publicus of the MidRepublic are difficult since much more is known about the latter; e.g. I am not aware of usufruct having existed in the Macedonian context (cf. Roselaar 2010, esp. 88-119). For parallels and potential influences in the contractual arrangements for granting land in the Roman context and under Hellenistic monarchy, see: Bengston 1972; cf. Billows 1995, 111-45. Hammond 1979, 654-73; Hatzopoulos 1996, 482; Anson 2008, 20. Just., Epit. 8.5.7-6.2. Transl. J.R. Ellis, 1969.




compulsorily.67 Forced resettlement was undertaken by the Romans already by the 3rd c. B.C.,68 but voluntary registration is more closely associated with Roman colonization.69 Philip’s approach towards the poleis on the coast has been described as ‘eclectic’,70 although Miltiades Hatzopoulos, probably more accurately, imagines it to have been governed by political expediency.71 Certainly the king was more lenient towards the poleis than is often presumed. Very few Greek cities were destroyed, others were re-founded, and some were augmented with Macedonian settlers.72 It is thought that poleis situated on the periphery of the ‘Macedonian commonwealth’ became allies and kept their autonomy.73 Roman handling of the Greek cities of Southern Italy and Sicily – even more lenient – was also arguably driven by political expediency, leading to a similar variety of outcomes,74 but the Western Greek context at this time was notably diverse to that of the northern Aegean.75 Urbanization and its accompanying human resettlement were related to unlocking the potential of land, but they also had other functions. It has been argued that Philip founded towns along his western border with the Illyrians and the Dardanians, and strengthened the line further with fortifications,76 and is reported to have done similar on his Thracian border.77 It is stated clearly by Diodorus that this was undertaken to create a strategic network,78 a policy subsequently adopted by the Seleucids.79 Similarly, Rome’s colonies of the MidRepublic are believed to have been deliberately situated so as to function as a strategic network,80 albeit one that grew over a longer period during which strategic concerns changed (cf. Pelgrom and Stek, and Bradley, in this volume). Of course, the term ‘network’ implies communication between its nodes, and there might well have been a Macedonian precedent to the Roman tradition of building long distance roads.81 Hammond believed that Philip augmented the system of royal roads begun by his predecessors,82 and that military roads were created for rapid troop deployment.83 If so, we might imagine this Macedonian tradition to have been derived from the Achaemenid kings,84 since there is little evidence of a similar habit among the ancient Greeks.85 Philip’s policy of founding towns was not carried out throughout the kingdom, and one can sense his pragmatism in maintaining the territorial and political organization of mountainous Upper Macedonia with its scattered villages.86 Tentatively, Hatzopoulos suggests that Philip may have equated rural “territorial units with poleis, and their councils of peliganes with civic boulai”.87 But there is insufficient textual evidence from Philip’s reign which can be used as a basis for comparing how he and the Romans organized the settlement structure in the countryside during the later 4th and 3rd c. B.C.88

67

68 69 70 71 72 73 74

75



Listed and discussed in Ellis 1969. See also Hammond 1979, 661-62; Hatzopoulos 1996, 70. Dench 2005, 165. Cf. Càssola 1988. Hammond 1979, 662. Hatzopoulos 1996, 190. Ibid., 181-89 and 482; Anson 2008, 20. Hatzopoulos 1996, 482. Some examples of varied Roman treatment of Greek cities include the foedus aequum with Neapolis 326 B.C.: Toynbee 1965, 102; Lomas 1993, 44-48. Locri, Rhegium and Croton placing themselves in the fides of Rome between 285 and 282 B.C.: Clemente 1990, 35; Cornell 1995, 363. The Roman re-founding of Poseidonia as a Latin colony in 273 B.C.: Torelli 1999; Crawford 2006. By which I mean many of the poleis of Magna Graecia had fallen to Lucanians and Bruttians by 338 B.C. (Musti 2005, 261-84), so that their socio-political and

76 77 78 79 80

81 82 83 84 85 86 87 88

ethnic situations were very different to those of the cities encountered by Philip. Hammond 1979, 654; id. 1981. Ellis 1969, 17. Diod. Sic. 16.71.1-2. Winter 1971, 38-39. Toynbee 1965, 158-60; Salmon 1969, 55-81; Coarelli 1992, 27; id. 1998, 29-33. Cf. Dench 2005, 165. Cf. Laurence 1999. Hammond 1979, 658. Ibid., 673. Cf. Just., Epit. 7.4.1; Dench 2005, 164-65. Purcell 1990, 12; Rackham 1990, 105-6. Hatzopoulos 1996, 77-104. Ibid., 482. A period for which we are also poorly informed regarding the equivalent Roman situation (cf. Pelgrom 2008; Stek 2009, esp. 123-70).


Still, in terms of the strategies and policies attributed to Philip during his reign, there is arguably a resonance with subsequent Roman approaches to the conquest of Italy. For example, Anson considers that a ‘key’ to the king’s success was his destruction of relationships which subordinate political entities within Macedonia enjoyed with each other, and substituting these with bonds of allegiance to himself.89 A similar policy was adopted by Rome in 338 B.C. in a way, since many of the Latins were made citizens of Rome, disrupting their former loyalties, and the members of the Latin League not to be made Roman citizens were, to some degree, denied political relations with one another in the future.90 The conquest of Italy, through the granting of the citizenship and the legal obligations arising from treaties Rome forced on the communities it subjugated, and on those she founded, created a centripetalism which dominates our perception of the Roman system, reflected so forcefully in the network of viae publicae. Yet if one looks more closely at methods employed to administer the commonwealth, the similarity between Roman and Macedonian approaches starts to evaporate. Importantly, there appears to be no Macedonian equivalent to the Roman concept of citizenship. Nicholas Hammond suggested that the Macedonians usually possessed multiple citizenships,91 but this idea has been convincingly rejected.92 From the epigraphic evidence, Hatzopoulos concludes that “all the free inhabitants of the national territory, regardless of their ultimate ethnic origins were citizens of one Macedonian civic community (city, ethnos, or sympolity) and together formed the Macedonian ethnos.”93 Thus it was possible to be a citizen of an individual community within Macedonia and enjoy the local rights which went with it, but in relation to Macedonia as a whole, one would only be a member of the Macedonian ethnos.94 Macedonian ‘citizenship’ was not only not conferred in the way Roman citizenship was, it did not exist in the way Roman citizenship existed. Macedonia was a geographical area, albeit an elastic one, but Rome was a city; thus, no comparison can be made. There was a Macedonian assembly of some kind that was involved in appointing new kings, in decisions in capital cases, and in ratifying decisions made by the king’s council, for example, but its character is poorly understood.95 Settlers who augmented the population of existing urban centres became citizens of that community without being legally differentiated from its other inhabitants.96 The king possibly imposed a high degree of constitutional uniformity on the cities, but they largely ran their own affairs, paid no direct tax to the central authorities, and several cities at least seem to have been able to vote in their own laws.97 Superficially, this might not seem so different to the situation of Latin colonial towns, but in the Macedonian context, each city probably had a single, and incomparable, chief magistrate: an epistates.98 The role of this controversial figure is disputed, but his responsibilities are enormously important, to my mind, for they seem to determine the degree of centralization one can attribute to the Macedonian state. It appears that the epistates was some form of intermediary, for the “kings and their representatives always correspond not with the Council and the People of the cities but with the epistates, who appears thus personally responsible for the execution of the measures taken by the central authorities”.99 Hence the epistates has traditionally been seen as the king’s overseer or functionary, but there appears to be insufficient evidence to argue whether he was appointed by the king or elected locally.100 Either way, he had no equivalent in Rome’s colonial foundations.

89 90 91 92

93 94

Anson 2008, 17 and 21-25. See also note 52 above. Cornell 1995, 347-52. Hammond 1979, 647. Cf. Errington 1980; Hatzopoulos 1996, 206-9; Anson 2008, 20. Hatzopoulos 1996, 208-9. Membership of the Macedonian ethnos was not conferred by the state and did not have a status associated with it within Macedonia. The state ethnikon of ‘Makedones’ is a term occurring only in documents found

95 96 97 98 99 100

outside Macedonia and applied for the purpose of ethnic identification (Errington 1980, 79; Hatzopoulos 1996, 168). Cf. Hatzopoulos 1996, 261-322. Ibid., 182; Anson 2008, 20. Hammond 1979, 651; Hatzopoulos 1996, 158 and 483. Cf. Hatzopoulos 1996, 381-429. Ibid., 394. Ibid., 427; Hammond 1999, 375.




So where we see the greatest similarity in the approaches of Philip and the Romans is in the development of land through the resettling of populations and urbanization, leading to the generation of wealth and the augmentation of their armies. Both placed towns and other fortified centres strategically to defend newly acquired territory and were interested in linking them together with roads (cf. Bradley in this volume). Great effort was made by both to develop bonds of allegiance to the central authority at the cost of former alliances. If the Romans did adapt elements of Philip’s strategies, how did they find out about them? Ancient textual sources are entirely silent in this regard,101 and so I have no direct basis to argue some form of cultural contact took place. We actually have someone arriving in Italy in 334 B.C. who had intimate knowledge of both Philip the man and very likely his expansionist policies: Alexander I of Epirus, the Molossian. Philip had been assassinated two years earlier at the wedding of the Molossian, who had been previously placed on the throne of Epirus by Philip. Both Livy and Justin report some kind of friendly contact between the Romans and Alexander during his campaign on the peninsula since they shared common enemies in the form of the Samnites.102 There has been much speculation on the nature of this contact.103 Naturally one cannot discount this meeting as a possible channel (or as the basis for developing a channel) through which the Romans learnt something of Macedonian strategies, but it is perhaps too convenient. The timing is not quite right and, as I have stated before,104 I am generally not in favor of scholars burdening historical figures with extra duties – especially transferring knowledge – just because they happen to be in the right place at the right time. In the context of the colonies’ urban centres, archaeology reveals the existence of processes by which architectural and townplanning concepts were transmitted great distances across the Mediterranean. These processes represent agency not reported in textual sources, reminding us that there was much more going on during this period than described in the fragmentary and often distorted surviving textual sources relevant to it. Based on the degree of interconnectivity in the region and the outstanding character of Philip’s accomplishments, it would seem likely that the Romans were aware of the king and his impact. Whether their awareness was of sufficient depth to allow them to adapt some of the king’s strategies is another matter. By suggesting that the Romans approach to colonization was informed by Philip’s Macedonia, one also runs the risk of implying that Roman inventiveness was limited in this respect. If Philip was able to innovate, then surely the Romans could too? One could hardly respond in the negative, yet the possibility remains that some influences on Rome’s developing colonization model may have had origins beyond the Italian peninsula. If so, there might well have been more than those discussed here.

Acknowledgements My gratitude is owed to all the contributors to this work for their stimulating presentations at the workshop from which this volume is derived, but especially to the organizers, Tesse Stek and Jeremia Pelgrom. I am also grateful to Edward Anson and the anonymous referee whose useful comments enabled me to improve this paper.

101

102



The literary evidence for Rome’s interaction with Macedonia is reviewed in Walbank 1981. Liv. 8.17; Just., Epit. 12.2.

103 104

Cf. Zevi 2004. Sewell 2010, 51.


References Amici, C. M. 2004-05. “Evoluzione architettonica del comizio a Roma,” RendLinc 77, 351-79. Anson, E. M. 2008. “Philip II and the transformation of Macedonia: A reappraisal,” in T. Howe and J. Reames (edd.), Macedonian legacies. Studies in ancient Macedonian history and culture in honor of Eugene N. Borza (Claremont) 18-30. Bengston, H. 1972. “A Hellenistic land-conveyance: the estate of Mnesimachus in the plain of Sardis,” Historia 21, 45-74. Billows, R. A. 1995. Kings and colonists: aspects of Macedonian imperialism (Leiden). Bispham, E. 2000. “Mimic? A case study in early Roman colonisation,” in E. Herring and K. Lomas (edd.), The emergence of state identities in Italy in the first millennium BC (London) 157-86. Bispham, E. 2006. “Coloniam Deducere: How Roman was Roman Colonization during the Middle Republic?,” in G. J. Bradley and J.-P. Wilson (edd.), Greek and Roman Colonisation: Origins, Ideologies and Interactions (Swansea) 73-160. Bispham, E. 2007. From Asculum to Actium: The Municipalization of Italy from the Social War to Augustus (Oxford). Bradley, G. J. 2006. “Colonization and identity in Republican Italy,” in G. J. Bradley and J.-P. Wilson (edd.), Greek and Roman Colonization: Origins, Ideologies and Interactions (Swansea) 161-87. Brandt, J. R. 1985. “Ostia, Minturno, Pyrgi. The Planning of three Roman Colonies,” ActaAArtHist 5, 25-87. Brown, F. 1980. Cosa: the making of a Roman town (Ann Arbor, MI). Brown, F. E., E. Hill Richardson. and L. Richardson 1993. Cosa. III, the buildings of the forum: colony, municipium, and village, MAAR 37 (University Park, PA). Carafa, P. 1998. Il comizio di Roma dalle origini all’età di Augusto (Rome). Càssola, F. 1988. “Aspetti sociali e politici della colonizzazione,” DialArch 6-2, 5-17. Clemente, G. 1990. “Dal territorio della città all’egemonia in Italia,” in Storia di Roma, vol. 2: L’impero mediterraneo. 1, La repubblica imperiale (Turin) 19-39.

Coarelli, F. 1983. Il foro romano, vol. 1: Periodo arcaico (Rome). Coarelli, F. 1985. Il foro romano, vol. 2: Periodo repubblicano e augusteo (Rome). Coarelli, F. 1992. “Colonizzazione e municipalizzazione. Tempi e modi,” DialArch 10, 21-30. Coarelli, F. 1998. “La storia e lo scavo,” in F. Coarelli and P. G. Monti (edd.), Fregellae 1. Le fonti, la storia, il territorio (Rome) 29-69. Coarelli, F. 2005. “Pits and fora: a reply to Henrik Mouritsen,” PBSR 73, 23-30. Conventi, M. 2004. Città romane di fondazione (Rome). Cornell, T. J. 1995. The beginnings of Rome. Italy and Rome from the Bronze Age to the Punic Wars (c. 1000-264 BC) (London). Cornell, T. J. 2000. “The city-states in Latium,” in M. H. Hansen (ed.), A comparative study of thirty city-state cultures (Copenhagen) 209-28. Coupry, J. 1964. “Les fouilles d’Olbia, a Hyères,” CRAI 108, 313-21. Coupry, J. 1986. “Les fortifications d’Olbia de Ligurie. Propositions, questions,” in P. Leriche and H. Tréziny (edd.), La fortification dans l’histoire du monde grec. Actes du colloque, Valbonne 1982 (Paris) 389-99. Crawford, M. H. 1995. “La storia della colonizzazione romana secondo i romani,” in A. Storchi Marino (ed.), L’incidenza dell’antico. Studi in memoria di Ettore Lepore (Naples) 187-92. Crawford, M. H. 2006. “From Poseidonia to Paestum via the Lucanians,” in G. J. Bradley and J.-P. Wilson (edd.), Greek and Roman Colonization: Origins, Ideologies and Interactions (Swansea) 59-72. Dench, E. 2005. Romulus’ Asylum. Roman identities from the age of Alexander to the age of Hadrian (Oxford). Ellis, J. R. 1969. “Population-transplants under Philip II,” Makedonia 9, 9-17. Errington, R. M. 1980. “Review of: N. G. L. Hammond and G. T. Griffith, A history of Macedonia, volume II,” CR 30-1, 78-80. Fentress, E. 2000. “Frank Brown, Cosa, and the idea of a Roman city,” in E. Fentress (ed.),




Romanization and the city: creation, transformations and failures (Portsmouth, RI) 11-24. Gabba, E. 1988. “Aspetti militari e agrari,” DialArch 6, 19-22. Galsterer, H. 2006. “Die römischen Stadtgesetze,” in L. Capogrossi Colognesi and E. Gabba (edd.), Gli statuti municipali (Naples) 31-56. Garbini, G. 1996. “The Phoenicians in the Western Mediterranean (through to the fifth century B.C.),” in G. Pugliese Carratelli (ed.), The western Greeks (London) 121-32. Gargola, D. J. 1995. Lands, laws, and gods. Magistrates & ceremony in the regulation of public lands in Republican Rome (Chapel Hill). Hammond, N. G. L. 1979. “The internal organisation of Macedonia and of Macedonian conquests in the Balkans,” in N. G. L. Hammond and G. T. Griffith (edd.), A History of Macedonia, vol. 2: 550-336 BC (Oxford). Hammond, N. G. L. 1981. “The western frontier in the reign of Philip II,” in H. J. Dell (ed.), Ancient Macedonian Studies in Honour of C. F. Edson (Thessaloniki) 199-217. Hammond, N. G. L. 1989. The Macedonian state: origins, institutions, and history (Oxford). Hammond, N. G. L. 1999. “The roles of the epistates in Macedonian contexts,” BSA 94, 369-75. Harl, K. W. 2008. “Legion over Phalanx: the battle of Magnesia, 190 B.C.,” in T. Howe and J. Reames (edd.), Macedonian legacies. Studies in ancient Macedonian history and culture in honor of Eugene N. Borza (Claremont) 257-82. Harris, W. V. 1979. War and imperialism in Republican Rome, 327-70 BC (Oxford). Hatzopoulos, M. B. 1996. Macedonian institutions under the kings, vol. 1: A historical and epigraphic study (Athens). Krause, C. 1976. “Zur baulichen Gestalt des republikanischen Comitiums,” RömMitt 83, 31-69. Lackner, E. M. 2008. Republikanische Fora (Munich). Laurence, R. 1999. The roads of Roman Italy. Mobility and cultural change (London). Lomas, K. 1993. Rome and the Western Greeks, 350 BC-AD 200. Conquest and acculturation in Southern Italy (London). Merritt, B. D. 1935. “Inscriptions of Colophon,” AJP 56, 358-71.



Mertens, J. (ed.) 1961. Alba Fucens I-II. Rapports et études (Brussels). Mouritsen, H. 2004. “Pits and politics. Interpreting colonial fora in Republican Italy,” PBSR 72, 3767. Musti, D. 2005. Magna Grecia. Il quadro storico (Rome/Bari). Neutsch, B. 1956. Archaologische Grabungen und Funde im Bereich der unteritalischen Soprintendenzen von Tarent, Reggio di Calabria und Salerno (1949-1955), Arch. Anz. 71, 193- 450. Patterson, J. R. 2006. “Colonization and historiography: the Roman Republic,” in G. J. Bradley and J.-P. Wilson (edd.), Greek and Roman Colonization: Origins, Ideologies and Interactions (Swansea) 189-219. Pelgrom, J. 2008. “Settlement Organization and Land Distribution in Latin Colonies before the Second Punic War,” in L. de Ligt and S. J. Northwood (edd.), People, Land and Politics. Demographic Developments and the Transformation of Roman Italy 300 BC-AD 14 (Leiden) 333-72. Purcell, N. 1990. “The creation of provincial landscape. The Roman impact on Cisalpine Gaul,” in T. Blagg (ed.), The early Roman empire in the West (Oxford) 7-29. Rackham, O. 1990. “Ancient landscapes,” in O. Murray and S. Price (edd.), The Greek city from Homer to Alexander (Oxford) 85-111. Rescigno, C. and F. Senatore 2009. “Le città della piana campana tra IV e III sec. a.C. Dati storici e topografici,” in M. Osanna (ed.), Verso la città. Forme insediative in Lucania e nel mondo italico fra IV e III sec. a.C. Atti delle Giornate di studio, Venosa, 13-14 maggio 2006 (Venosa) 415-62. Richardson jr., L. 1957. “Cosa and Rome, Comitium and Curia,” Archaeology 10, 49-55. Roselaar, S. T. 2010. Public land in the Roman Republic: a social and economic history of Ager Publicus in Italy, 396-89 BC (Oxford). Salmon, E. T. 1969. Roman colonization under the Republic (London). Sewell, J. 2010. The formation of Roman urbanism, 338 - 200 B.C.: between contemporary foreign influence and Roman tradition (Portsmouth, RI). Sjöqvist, E. 1951. “Pnyx and Comitium,” in G. Mylonas (ed.), Studies presented to David M. Robinson, vol. 1 (Saint Louis, MO) 400-11.


Sommella, P. 1988. Italia antica. L’urbanistica romana (Rome). Stek, T. D. 2009. Cult places and cultural change in Republican Italy. A contextual approach to religious aspects of rural society after the Roman conquest (Amsterdam). Terrenato, N. 2008. “The cultural implications of the Roman conquest,” in E. Bispham (ed.), Roman Europe (Oxford) 234-64. Torelli, M. 1988. “Aspetti ideologici della colonizzazione romana più antica,” DialArch 6, 65-72. Torelli, M. 1999. Paestum romana (Paestum). Torelli, M. 2007. “Parte prima. L’età regia e repubblicana,” in P. Gros and M. Torelli (edd.), Storia dell’urbanistica. Il mondo romano (2nd edn., Rome) 3-198. Toynbee, A. J. 1965. Hannibal’s legacy. The Hannibalic War’s effects on Roman life (London / New York). Verdin, F. 1997. “Un exemple de voirie grecque en territoire indigène: Olbia de Provence, Hyères-lesPalmiers, Var (v. 340 - v. 50 av. J.C.),” REA 99, 427-42.

Walbank, F. W. 1981. “Livy, Macedonia and Alexander,” in H. J. Dell (ed.), Ancient Macedonian Studies in Honour of C. F. Edson (Thessaloniki) 335-56. Winter, F. E. 1971. Greek fortifications (Toronto). Worthington, I. 2008. Philip II of Macedonia (New Haven/London). Zaccaria, C. 1996. “La base di T. Annius Luscus,” AquilNost 67, 179-84. Zaccaria Ruggiu, A. 1995. Spazio privato e spazio pubblico nella città romana (Rome). Zanker, P. 2000. “The city as symbol: Rome and the creation of an urban image,” in E. W. B. Fentress (ed.), Romanization and the city: creation, transformations and failures (Portsmouth, RI) 25-41. Zevi, F. 2004. “Alessandro il Molosso e Roma,” in Alessandro il Molosso e i condottieri in Magna Grecia. Atti del quarantatreesimo convegno di studi sulla Magna Grecia, Taranto – Cosenza 26 – 30 Settembre 2003 (Taranto) 793-832.




Republican colonization and early urbanization in Central Adriatic Italy: the valley of the River Flosis Frank Vermeulen

Introduction: the general setting In the central Adriatic regions of Picenum and the ager Gallicus, urbanism was generally a comparatively late phenomenon, seldom taking place before the 3rd c. B.C. Compared to regions as Campania, the Roman cities in Picenum are small and do not always contain the whole set of typical urban public buildings we can find in the important cities in Italy.1 On the other hand, the density of Roman towns in this region was surprisingly high, even if these population centers were not as large as those on the central-Tyrrhenian side of the peninsula.2 The origin of this particular urban landscape is difficult to reconstruct as the archaeological record for the pre-Roman and early Roman periods is limited and preponderantly funerary in nature. This dearth, which is certainly one of evidence but also of imagination, makes it problematic to cross the border between Iron Age prehistory and Roman history, both conceptually and chronologically.3 However, the time has come to bridge the gap, and some on-going research is indicating useful directions to take. The first official contacts of the peoples on the Adriatic side of the peninsula with Rome were made during the early decades of the 4th c. B.C. and arose in an anti-Samnite (to the south) and anti-Celtic (to the north) context. Towards the end of the 4th c. B.C., Roman contacts with the easternmost Umbrians led to the first treaty in the area east of the central Apennines.4 Some years later, after the battle of Sentinum (295 B.C.), the decisive battle in the Third Samnite War in which the Romans were able to overcome a coalition of Samnites, Etruscans, Umbrians and their Gallic allies, the whole ager Gallicus (283 B.C.) and most of the ager Picenus (268 B.C.) was confiscated by Rome, with the exception of the federate towns of Ancona and Asculum. This annexation was followed by the foundation of a series of colonies of Latin and Roman citizens in the area between Rimini and Ascoli. The part of the Picene population that was not deported to the region of Salerno after the conquest in 269-68 B.C. first became members of a civitas sine suffragio, and later full Roman citizens.5 The Latin and Roman colonies founded on the Adriatic side between 289 (Hadria, Sena Gallica) and 184 B.C. (Pisaurum, Potentia) were ‘maritime’, in the sense that they were all located on the coast.6 This first phase of colonization in the area was finalized with the foundation of the colonia civium Romanorum of Auximum, probably in 157 B.C.,7 a site which had perhaps already begun as a forum of Roman commercial activity a couple of generations earlier.8 Although Auximum does not lie directly on the coast, it is situated very close to it, in the hinterland of friendly Ancona and its port. Also the inner parts of the newly conquered land soon attracted full Roman attention (cf. Hermon in this volume). In the Umbrian districts of Camerinum and Matilica, the Roman confiscation was followed by land divisions to viritane colonists. This most likely already happened around 300 B.C.9 One generation

1 2 3 4

De Ligt 2012. Vermeulen and Mlekuz 2012. Williams 2001. With Camerinum in 310 B.C. Cf. Liv. 9.36; Frontin., Str. 1.2.2.

5 6 7 8 9

Paci 2008, 423. Delplace 1993. Vell. Pat. 1.15.3. Sisani 2007, 569. Ibid., 129, 184.




after the Roman incorporation of the entire area between the mountains and the Adriatic Sea, the Lex Flaminia de Gallico et Piceno agro viritim dividundo of 232 B.C. radically restructured this part of the Italian peninsula. This law focused on the creation of new administrative centers, not necessarily urban in character. The initiative by the tribune Gaius Flaminius had an enormous impact on the settlement system since, with the exception of the federate territories of Asculum and Ancona, it led to great numbers of viritane colonists being settled in the ager Gallicus et Picenus. On the institutional level, in this period the formation of typical Roman-style entities such as pagi and vici can be seen, which are well attested in places such as Cupramontana, Cingoli and Montoro.10 This radical reorganization, known mostly from texts and inscriptions, clearly marked the dominance of Rome over the land and the people of the region. At the same time it facilitated the incorporation of the indigenous population in the Roman state and census.11 Regrettably, the precise number of Roman immigrants settled in this area by Flaminius is unknown. However, some indirect proof of their large number can be found in the difficulties encountered by the Carthaginian army during its passage through Picenum, which can be interpreted as resulting from the resistance of many loyal Roman citizens that already lived there.12 The massive arrival of Roman and Latin colonists in Marche led to further large scale annexation and radical reorganization of former Picenian, Senonian and Umbrian territory. At least part of this process was linked to the evolution of the road network, most prominently the Via Flaminia, which crossed the Apennines in eastern Umbria to connect Rome with the Po plain via the ager Gallicus. There is a strong relationship between the developments of this road, colonial foundations, areas distributed ad viritim, as well as with the federate centers (for this last point, cf. Bradley in this volume).13 Apparently, the majority of the praefecturae which were established in the region were positioned either on this road or on one of its diverticula. Recent studies have shown that the transformation from praefecturae to municipia in Umbria happened immediately after the Social War. This can be deduced also from the fact that these cities were governed by a college of four magistrates (quattorviri). In the territory of the Senoni and in Picenum this process of municipalization occurred somewhat later, during the second half of the 1st c. B.C.14 In this latter case, the new municipia were headed by duoviri.15 Despite the rather late advance of municipia in this region, it is clear from epigraphy and archaeology that many central places did not wait for the formal municipalization to develop certain urban structures.16 As will be shown in more detail below, various centers already had clear urban characteristics in the early 1st c. B.C.17 Nonetheless, it is clear that the urbanization process intensified in the second half of the 1st c. B.C. A crucial factor for this development must have been the settlement of veterans in the interior of Picenum, during the period of the second Triumvirate (43-33/32 B.C.).18 The Liber coloniarum mentions seven veteran settlements in the middle and lower valleys of Picenum: Potentia, Ricina, Trea and Septempeda (River Potenza), Pausulae and Tolentinum (River Chienti), and Urbs Salvia (River Fiastra).19 Later, new land assignations were made along the coast as well as in the interior by Augustus, among them at Camerinum and Matilica. New military colonial (re-)foundations dated to the period between the Triumvirate and Augustus are known for Ancona, Pisaurum, Firmum Picenum and probably also for Fanum Fortunae, Falerio and Asculum Picenum.20 It seems that none of the territories with good arable land in the region escaped the

10 11 12 13 14 15 16



Capogrossi Colognesi 2000; Stek 2009. Tarpin 2002, 245. Polyb. 3.77.4-7; Sisani 2007, 58. Coarelli 1988, 35. Sisani 2007. Dall’Aglio and Castagnoli 2002. Marengo and Paci 2004.

17

18 19 20

E.g. regular street grids are known from the early 1st c. B.C. in Cupra Maritima (2 actus) and Falerio (2:3 actus). Cf. Perna 2012. Paci 1998. Blume et al. 1848, 253. Branchesi 2007.


effect of these new assignations. The fact that even upland areas (for instance near Amandola where a cippus from that period was found) were included shows that these new land allotments even extended to the more marginal landscapes; an indication that large numbers of people must have (been) moved into the region.21

Recent research into urbanism in central Adriatic Italy Until recently, the study of urbanism and especially of Republican colonial town planning in central Adriatic Italy has tended to concentrate heavily on the coastal towns with a history of continuity, such as Rimini, Fano and Pesaro. The urban plans proposed for these towns are rather hypothetical as modern construction prevents a reliable spatial examination of these sites. More problematic even is to understand the urban development of these towns over time, especially their transition from the Iron Age to the Roman period. There is some evidence to suggest that the central-Adriatic coastal areas witnessed population movement from hill sites to the central area of the coastal plain already in the late 6th c. B.C. Hence, the zones around the later colonies of Fanum Fortunae, Ariminum, Pisaurum and Sena Gallica are likely to have been occupied very early so as to make use of the river estuaries as transit ports.22 Detailed analysis of the archaeological records of these later colonial towns indicates that they were already occupied in the 4th c. B.C. However, as the overview below demonstrates, much of this evidence is still very scattered and it is therefore difficult to reconstruct the pre-Roman phase of these settlements in any detail: I) The plan of Sena Gallica, probably created between 290 and 284 B.C., is still very hypothetical, because it is based only on the study of the modern street network and on some data from an excavated domus.23 Reconstructions therefore diverge: Paolo Sommella proposes a plan based on the module of 60 x 60 m blocks,24 whereas Pier Luigi Dall’Aglio postulates a street pattern with more or less east-west oriented longitudinal blocks of 70 by 35 m (2 x 1 actus).25 The possibility that two slightly differing orientations existed in the urban grid of Sena Gallica cannot be excluded: one linked to the early colonial foundation and one with a triumviral re-organization of the city. II) More information is available for Pesaro. The Roman colony of Pisaurum (184 B.C.) was certainly founded on a preceding settlement. This is testified by the nearby Picene necropolis and by the finds of Greek pottery in the urban area. The rectangular city plan (slightly crooked as it was adapted to accommodate the river bend) has maximum dimensions of 484 x 400 m The town was divided into four not fully equal parts by two intersecting roads which met near the forum and it housed 14 or 15 insulae.26 III) Archaeological evidence is also available for the oldest phases of Rimini. Under the later Roman town, remains of a earlier settlement of some 20 ha were found, which are dated to the 4th c. B.C. and first quarter of the 3rd c. B.C. Datable elements are Greek coins, 4th c. B.C. Black Glazed pottery, a warrior stela and Gnathia ware. There is convincing evidence to suggest that the site was already under Roman control in the early 3rd c. B.C., that is before the founding of the colony of Ariminum in 268 B.C.27 Even more problematic is the reconstruction of urban developments in the inland towns. As many of these sites are situated on high ground (mostly hilltops), they have remained settled in Medieval times and beyond. This applies to places like Camerinum, Matilica, Pausulae and Tolentinum. These sites were certainly

21 22 23 24

Paci 1998, 212. Dall’Aglio, De Maria and Mariotti 1991. Luni 2004, 183. Sommella 1988.

25 26 27

Dall’Aglio, De Maria and Mariotti 1991. Sommella 1988. Maioli, Malnati and Miari 2012.




important pre-Roman centers which gradually transformed into Roman cities, especially after the municipalization movement of the 1st c. B.C. However, since not much archaeological research is conducted, information is scarce and comes mostly from their partly preserved walls.28 More reliable information exists for those inland sites which have been deserted in Medieval or Early Modern times. For these sites, which have been the subject of more intensive research during the past two decades, it is possible to reconstruct their entire layout and early urban development. Recent research by teams from the universities of Macerata, Bologna and Ghent especially has contributed enormously to our knowledge of those sites. Sites such as Suasa, Ostra, Urbs Salvia, Potentia, Trea, Ricina and Septempeda are now being uncovered by means of intensive surveys combined with excavations in carefully selected areas. These recent projects provide us with important insights in the early phases of town formation in this region. For example, recent excavations of the University of Bologna in Suasa, in the Gallic Senoni area, have found proof of the establishment of a praefectura on the site, which was followed later by the creation of a municipium. The case of Suasa shows that Republican phases can be found if regular excavations are conducted in the right areas. Important finds include 3rd c. B.C. finds (including a Latin inscription!) and 2nd c. B.C. architecture, such as a sanctuary which is located underneath the later forum. Also the first building phase of an important domus of the aristocratic Coedii located close by the forum has been identified.29 The forum was probably only built in late Augustan-Tiberian times, when the town was also enriched by a series of buildings for the staging of spectacles (theater, amphitheater) and other public monuments.30 At Urbs Salvia, the excavations of the University of Macerata have not found any evidence for pre-Roman occupation of the site. The oldest archaeological and epigraphic evidence points to a foundation date in the later part of the 2nd c. B.C., when a possible spring sanctuary and mansio were constructed along the main road, the Salaria Gallica, which later became the main cardo of the city. Some building structures with walls of river cobbles and earth floors have been partly revealed here. Early 1st c. B.C. votive pits and parts of a Late Republican public building seem to have been connected to this early urban phase, which some researchers link to the foundation of Pollentia in the late 2nd c. B.C.31 However, it seems most likely that the town was only granted colonial status under Augustus, when an impressive phase of monumentalization took place, including the building of the forum and a new spring sanctuary.

Case study: colonization and early urbanism in the valley of the Potenza (Flosis) By focusing now on a particular case study area in this central Adriatic region of Marche, where since 2000 a systematic survey project has been conducted by a team of Ghent University, an attempt will be made to reveal certain long-term trends in urbanization processes which are difficult to detect when focusing on a local spatial scale, or on a narrow chronological period.32 This valley, a circa 80 km long corridor along the River Potenza just north of Macerata, was a much frequented passageway connecting Tyrrhenian and Adriatic Italy from the later Prehistoric period onwards. In Roman times, a diverticulum of the Via Flaminia followed the Potenza Valley and thus linked Rome with the harbor city Ancona. The importance of this route grew significantly during the Late Republic which must also have been the prime motive behind the internal colonization of the Potenza Valley.33 Today the rural landscape of the valley is little affected by modern occupation, allowing for intensive artifact surveys and geophysical prospecting, as well as remote sensing applications. Using such a multidisciplinary approach, the Ghent team under my direction has detected more than 120 proto-historic and Roman settlement sites (figs. 1 and 2).34 In a second phase of the project (2005-10) a selection of these sites

28 29 30 31



Cf. Auximum. Dall’Aglio and De Maria 2010. Giorgi and Lepore 2010. Perna 2006; Fabrini 2012.

32 33 34

Vermeulen et al. 2009; Vermeulen 2012c. Dall’Aglio and De Maria 2010, 40. Vermeulen et al. 2003; Percossi, Pignocchi and Vermeulen 2006; Vermeulen 2012a.


FIG. 1. Location of the Roman urban centers in Northern Picenum and the border areas with Umbria and the ager Gallicus. Also shown are the known Roman rural settlements in the Potenza Valley (with an indication of the systematically surveyed transect zones of the PVS survey).

has been analyzed in more detail, using an intensive intra-site survey: the colony of Potentia on the coast and the municipia of Ricina, Trea and Septempeda in the interior.35 In the course of this period, a re-analysis of some older excavation data was also carried out, together with a reassessment and a general mapping of all known sites and find spots throughout the whole valley (in total some 350 sites) dated between the later Bronze Age and the end of Roman dominance of the area.36 Small-scale stratigraphic excavations complemented some of the survey data, especially in the colonial town of Potentia.37 Pre-Roman central places The Potenza Valley Survey (PVS) project was able to distinguish three main types of settlement dating to the pre-Roman, Late Picene period (5th-3rd c. B.C.): small farmsteads, hamlets/villages and hillfort sites.38 Numerous small farmsteads have been identified scattered over the landscape, which occasionally cluster together to form larger sites such as small hamlets or composite villages. For this latter category, there is good excavated evidence in the Umbrian district of Matelica, where several settlements of different sizes have been found near the modern town, with an average distance of approximately 1.5 km from each other

35 36

Vermeulen and Mlekuz 2012. Percossi, Pignocchi and Vermeulen 2006.

37 38

Vermeulen et al. 2011. Boullart 2003; Vermeulen and Mlekuz 2012.

ďœąďœ´ďœľ


FIG. 2. Location of pre-Roman hilltop centres and surrounding clusters of Iron Age rural settlement in the Potenza Valley.

and each with its own burial ground. The distribution of these clustered, undefended areas of habitation in the valleys seems to suggest complete control over the territory and possibly even over some of the nearby Apennine passes.39 There is some evidence to suggest that a similar situation existed near modern Treia, where a number of proto-historic settlement foci in the immediate vicinity of the later Roman town site of Trea have been found.40 Both the composite village landscape near Matelica and the cluster near Treia are located on an important communication route to Ancona. The third and most interesting category in this context of settlements can be defined as larger hill-fort sites, which are sometimes also important sanctuary sites.41 The combination of new survey evidence and old excavation data have made it possible to identify at least five such sites in the Potenza basin dating to the 5th c. B.C.,42 which are rather evenly spread over the valley and located on the most prominent positions in the landscape (fig. 2). These sites are: Monte Primo near the Umbrian Apenine gorge of the town of Pioraco,43 Monte Pitino and Monte Franco near San Severino Marche,44 the hilltop covered by the modern town area of Recanati45 and the lower plateau of Montarice overlooking the mouth of the River Potenza.46

39

40 41 42

ďœąďœ´ďœś

De Marinis and Silvestrini 2005, 139. For such nucleated settlement systems in the Samnite area see Stek and Pelgrom 2005. Vermeulen et al. 2009, 102-4. Letta 1992; Naso 2000; Vanzetti 2004. The hilltop centers of Matelica and Camerino, in the Apennine Umbrian district, which strictly speaking are

43

44

45 46

not part of the River Potenza basin, are excluded. Lollini 1979; Bonomi Ponzi 1992; Vermeulen et al. 2009. Lollini 1958; Boullart and Vermeulen 2005; Vermeulen et al. 2009. Percossi Serenelli 2003. Vermeulen et al. 2003.


The position of these five sites, controlling the coast and river, suggests that they played an important role in controlling the exchange of goods between the coastal area (including the emporium site of Numana on Monte Conero) and hinterland communities. The hill-forts are all characterized by defense systems that enclosed relatively small surface areas of 4 to 7 ha. Most are surrounded by cemeteries which have yielded some very wealthy graves.47 Some of these strategically located sites developed valley-based ‘antennae’ shortly before and during the phase of first contacts with the Romans in the 3rd c. B.C. It is clear that these sites functioned as central places with important political, commercial and religious functions. This aggregative quality makes the analysis of such sites valuable to the study of social change in the communities of pre-Roman Picenum. The surveys have yielded important indications of the way these centers were either transformed or gradually declined during the early romanization process, and of their eventual obliteration in the Late Republic. The hilltop site of Monte Pitino, which was the principal Picene center of aristocratic control in the central part of the valley, witnessed strong depopulation in the Late Republican period and most probably had been fully abandoned by the time of Augustus when two municipia, Trea and Septempeda, developed in the immediate neighborhood.48 The position of these new municipia, closely linked to the wider inland road network of central Italy, gave them the opportunity to develop a strong Roman profile in the course of the 1st c. A.D. (see below). Comparable to the case of Ricina, new research demonstrates more clearly than before that these inland towns developed along an important inland road with pre-Roman roots and gradually acquired urban features to become real towns, that is, Roman municipia with a city wall and forum, in the 1st c. B.C. Fieldwork confirmed that their success was at least partially attributable to the local topography, which offered easy fording places on the river and other geo-morphological advantages.49 Early Roman urbanization in the interior Picenum and the adjoining western border area of Umbria were characterized by a dense urban network. The average distance between towns in this region was 13.5 km, which is slightly higher than the corresponding figure in Latium and Campania.50 This urban density is particularly well visible in the upper Potenza Valley where a compact network of small towns with municipal status was located. These municipalities, as argued above, had their roots in pre-Roman settlements (Trea, Septempeda, Matilica, Camerinum), which suggests that, to some extent at least, the distinct dense pattern of Roman towns here resulted from a pre-existing pattern of settlement. At the same time, one can also recognize clear Roman influences in the development and spatial positioning of these towns. The Roman administrative system which was linked to the new road network certainly influenced the spatial distribution of central places. The subsequent urbanization of these places in the beginning of the 1st c. B.C., is most likely connected to the enfranchisement of the Italic elites after the Social War. As new Roman citizens they began to pay more attention to the city of Rome and to political competition there.51 Accordingly, they began to embellish the newly created municipia and turned their backs completely on their old power bases on hilltops. It is not a coincidence that in precisely this period the municipia of Septempeda, Trea and Ricina in the Potenza Valley fully developed. These sites were all located along the earlier mentioned diverticulum, the Via Flaminia per Picenum Anconam. The new Ghent University surveys of the town sites of Septempeda, Trea and Ricina are beginning to reveal the size and urban pattern of these organically grown Roman centers (fig. 3).52 With their moderate size of 12 to 22 ha intra muros, they remained small centers throughout their entire existence, and it is

47

48 49

Conati Barbaro, Manfredini and Silvestrini 2003; Vermeulen et al. 2009, 106-8. Vermeulen et al. 2009. Vermeulen and Mlekuz 2012.

50 51 52

De Ligt 2012. Patterson 1987. Vermeulen 2012c.




FIG. 3. General plan of the 4 Roman towns in the Potenza Valley according to recent survey evidence.

likely that their oldest core, tentatively dated to between the 3rd and 2nd c. B.C., was even smaller. Nevertheless, as a result of their promotion to the status of municipia shortly after the mid-1st c. B.C. and the subsequent settlement of colonists there under the second Triumvirate, they gradually acquired an infrastructure worthy of full-fledged cities. In the area of Trea (fig. 3) surveys have identified a series of small Picene settlements, unevenly scattered over an elongated plateau covering at least 30 ha, roughly 4 km northeast of the Monte Pitino hill-fort. The precise dating of these is difficult, but at least some sites seem to belong to the later phases of the Iron Age (5th-3rd c. B.C.). The eastern part of this plateau developed during the Republican period into a more aggregative settlement, clustered around the east-west road running from the Potenza Valley in the direction of Ancona (i.e. the above-mentioned diverticulum of the Via Flaminia).53 This roadside settlement – perhaps a conciliabulum or forum for commercial exchange or a praefectura – probably acquired a regular layout only around the time that the Via Flaminia was constructed in 220 B.C. At the latest around the mid-1st c. B.C., an area of approximately 12 ha was enclosed by an imposing city wall, in opus quasi-reticulatum.54 After a first phase of urbanization which consisted of organizing the housing blocks alongside the main road axis came the monumentalization of Trea which was given a large forum, several temples, a basilica and colonnaded street. In Septempeda (fig. 3), near San Severino Marche, the large-scale development of the intra- and extramural zones is now also discernible thanks to the new survey data. This town first developed during the

53



Vermeulen et al. 2009.

54

Vermeulen, Slapšak and Mlekuz 2012.


3rd-2nd c. B.C. in the vicinity of the road in the valley bottom. The oldest finds that attest to a settlement in this spot consist of a series of graves of that period containing quite rich grave goods including Black Glazed pottery, unguentaria, lagynoi and iron strigiles.55 The latest geophysical surveys and aerial photography data suggest that the core of this settlement was formed by a Hellenistic-style sanctuary, but this discovery needs further confirmation.56 Comparable to the settlement history of Trea, archaeological finds indicate important changes in this region in the 3rd c. B.C. The new settlement that developed on the valley floor site along the road as a kind of conciliabulum in this period began to take over the role as central place from the Picene hilltop site of Monte Pitino, which lies only 5 km northwest of Septempeda. The town, still called oppidum in the Liber coloniarum, only later became a municipium.57 This rise in status probably coincided with, or was even preceded by, the erection of a monumental sandstone town wall in opus quadratum in the late 2nd or the course of the 1st c. B.C. In this period, the incoherent habitation nuclei were developed into an enclosed built-up area of approximately 15 ha and, although there never was a fully regular street pattern, a series of streets developed parallel to the main central artery.58 In order to convey the idea of a Roman town, at least four extremely monumental entrance gates and a forum surrounded by public buildings (including baths) and monuments were erected.59 The available archaeological evidence does not allow a precise dating of these major urban developments, but inscriptions and typical architectural terracotta’s belonging to an important cult place and a major public building constructed in the last decades of the 1st c. B.C.60 suggest an Augustan date for this spectacular bloom.61 Finally, similar developments can also be observed in Ricina (fig. 3). Again the town grew gradually – from at least the 2nd c. B.C. – along the valley bottom road which eventually became the decumanus maximus of the city plan.62 The site is located some 7 km southwest of the presumed central Picene hilltop site, now located under modern Recanati. At Ricina, however, no finds preceding the Late Republic have been documented and the evidence for the erection and chronology of a regular city wall, enclosing a town area of some 22 ha, is still scanty.63 Major buildings around its center, including a large theater and a temple complex and an amphitheater have been discovered in recent years by the Ghent team.64 The two buildings for the staging of spectacles are especially remarkable. The presence of an amphitheater for gladiatorial games in such a small place was surely an important attraction for the veterans65 and served a wider region.66 It is very likely that both buildings were simultaneously designed and financed in Augustan times, which seems to testify to important investments by elite families. The presence of such rich investors is also suggested by the series of funerary monuments erected west of the urban center from the second half of the 1st c. B.C.67 Potentia: colonization of the coastal area It is especially in the coastal area, in and around Potentia near Porto Recanati, that fieldwork by the Ghent team since 2002 has offered the best possibilities to make highly detailed reconstructions of urban infrastructure and development.68 By integrating large-scale survey approaches with legacy data (e.g. epigraphic and historical evidence) and some good stratigraphical evidence, a reliable plan of the town could

55 56 57 58

59 60

Landolfi 2003, 51. Vermeulen 2012c. Blume et al. 1848, 253. Interestingly, the city wall also incorporates a small hilltop or arx, which could be interpreted as the core of pre-Roman settlement near the river, which functioned as an ‘antenna’ for the Monte Pitino central site. Vermeulen et al. 2009. CIL XI 5711.

61 62 63 64 65 66 67 68

Landolfi 2003, 58. Mercando 1971; Percossi Serenelli 1989. Ead. 1999. Vermeulen 2005; id. 2012c. Golvin 1988. Patterson 2006, 139. Mercando, Bachielli and Paci 1984. Vermeulen 2008; id. 2012b; id. 2012c; Vermeulen and Carboni 2010.




FIG. 4. General plan of Potentia, reflecting the third phase of town development.

be reconstructed (fig. 4). The plan indicates the exact location of the colonial city, with its town wall, including three (possibly four) city gates, the complete street network, the central forum and several other monumental complexes, many elements of city housing, three extra-mural funerary areas and a large segment of the suburban and rural settlement system and roads pertaining to the territory of the ancient city. Although some geomorphological constraints mask the visibility of some of the ancient structures, especially in the western and southern parts of the city area, it is still possible to distinguish a strictly rectangular town plan measuring roughly 525 by 343 m (almost 18 ha measured intra muros), lying parallel to the coast. The urban area was subdivided by a regular network of streets oriented parallel with the walls, forming city blocks (minimum 58) of different dimensions. The cardo and decumanus maximus intersect at the center of the town and delimit the rectangular forum square which is located to the southwest of the crossroad. Until recently, it was believed that Potentia was created ex novo.69 Recently, however, several archaeological indications have come to light, which seem to undermine the theory of a settlement of Roman colonists in a totally empty area. A short enumeration of the most convincing evidence to contradict the traditional view can suffice here:

69



Cf. Vermeulen and Verhoeven 2004. Exceptions were Percossi Serenelli 2001 who suggested the existence a possible Roman praefectura predating the installation of the colony, and some scholars who pointed at fragile evidence for contacts between Greek merchants from nearby Numana and natives of the Montarice promontory – where some Greek pottery was found out of con-

text – that potentially undermined this theory (Baldelli 2004, id. 2007; Luni 2004). No proof to the contrary was found during the 20-year long excavations of a temple area in the city center. The nearby proto-historic site of Montarice (see above), located 600 m to the north, was never seriously studied and was indeed largely shrugged off as an important Bronze Age settlement.


I) The Ghent surveys found datable Greek and Late Picene finds (5th-4th c. B.C.) on Montarice hill, especially on its southern-facing slopes. More importantly, some probably Greek imports have also been discovered in the center and on the coastal side of the later Roman town area. II) The, as yet unpublished, data from the Percossi excavations also hint at the presence of a precolonial phase of the settlement dated to the 5th to 4th c. B.C., which is buried under a one-metre-thick clay deck on which the later colony was built.70 This evidence seems to suggest that after a period of many inundations of the Potenza plain, responsible for the deposition of a thick alluvial layer and the dramatic shift of the river course to its later Roman bed, the zone of the later colony was reoccupied. Some excavated structures include: a small oven, Late Picene type pottery of late 3rd to early 2nd c. B.C. date, a deep (ritual?) pit with descending steps and remains of wooden and sandstone structures. Whether these finds indicate the presence of people of Picene descent or of groups of cives romani occupying a kind of conciliabulum cannot yet be determined. III) The discovery of the existence of a clear connection between the urban lay out of the city and the suburban road system (cf. above) – discovered in our aerial photographic survey since 2002 – further supports the possibility that a military praefectura, statio or vicus preceded the foundation of the colony in 184 B.C.71 Such scenarios have been attested in the northern Adriatic areas, such as the Veneto region. Colonial settlements in that region were not necessarily immediately followed by the construction of town walls. In Trieste (Roman Tergeste), for example, the wall was probably only built some 20 years after the installation of the colony.72 This demonstrates that, at least occasionally, the construction of urban settlements was not completed at once, but was accomplished in different building phases. IV) Recent geomorphological study of the area of the later colony of Potentia also reveals that only this precise place was suitable for permanent settlement in the immediate area of the river mouth. This makes pre-Roman settlement of the spot even more plausible, even though the still unpublished palynological data on the river mouth area show that this landscape remained dominated by wetlands until the second half of the 2nd c. B.C.73 Whatever the precise character of the early settlement might have been, it is clear that on the narrow beach ridge immediately north of the river mouth the colonia civium romanorum of Potentia was founded in 184 B.C.74 Reviewing all the new and the old archaeological data, and taking into account the scarce historical information, the hypothesis can now be put forward that there were three major phases in the development of the urban plan of Republican and Early Imperial Potentia (fig. 4).75 The first phase coincides with the first years after the official foundation of the colony and represents the initial settling of the arriving colonists, who might have been veterans of wars in Spain, Sicily or Sardinia.76 The chosen town area, which was probably surrounded by a ditch (fossa) and maybe a palisade or small agger, may have been a regular rectangular area of only some 525 by 300 m (16 ha). Its NNW-SSE orientation and putatively elongated shape are attributable to the presence of the longitudinal beach ridge which runs parallel to the nearby coastline. Not long afterwards, according to Livy for 174 B.C., the site of the colony was fully urbanized. Thanks to the financial intervention of the censor Q. Fulvius Flaccus, brother of one of the three co-founders of the colony, the ‘military-looking’ settlement developed into a real town. In that same year a whole series of direct interventions by the Roman censors in Roman ‘maritime’ colonies is

70 71 72 73 74

Percossi Serenelli 2012. Liv. 39.44.1. Di Filippo Balestrazzi 1999. Goethals, De Dapper and Vermeulen 2009. Liv. 39.44.1.

75 76

Vermeulen and Verhoeven 2004. A connection with the navy has been suggested (cf. Salmon 1969; Laffi 2001, 521). If so, they were settled in response to problems with the piracy of the Istri in the Adriatic (cf. Sisani 2007, 60).




recorded.77 This suggests that the role of these originally military settlements changed, and that the economic potential of these towns near the coast, including the production of, and the trade in, wine and textiles was recognized.78 At the same time, these interventions by Roman magistrates also demonstrate the incapacity of such small centers, whose autonomy from Rome was very limited, to create an urban infrastructure on their own.79 The earliest urban structures, as listed by Livy for Potentia as well as for its twin city Pisaurum, consisted of a temple of Jupiter, a circuit wall with three arched gates, a regular street network with sewers, an aqueduct, and a portico with shops to close off the forum square. Current archaeological data comply almost completely with this detailed historical information relating to 174 B.C.: I) During our 2007-10 excavations, the circuit wall – a fine ashlar construction of sandstone flanked by an interior agger – and at least one gate were discovered and preliminarily dated to the second quarter of the 2nd c. B.C.80 The two other gates are presumed on the basis of survey evidence; they are located centrally in the two short sides (north and south) and the western long side of the town; II) The street network has been clearly confirmed by surveys and its oldest phase has now also been revealed during the excavations of the decumanus maximus near the west gate; sewer structures have been found in this same area and near the excavated temple; III) Electromagnetic surveys revealed a long, narrow rectangular forum (length-width circa 3:1) with flanking porticos and tabernae in the very center of the original colonial settlement. Although its features are clearly Late Republican, its chronological evolution and the character of some of the (public?) buildings on its short sides still have to be established; IV) The temple of Jupiter, mentioned by Livy, can still not be identified with absolute certainty with the excavated Republican temple directly east of the forum. The excavated temple was built later in the 2nd c. B.C.81 However, using recent geophysical prospection my team has discovered a major templelike structure on the northern side of the forum. As this building seems to dominate the forum axially on its short side, it has all characteristics of a Capitolium. However, excavations are needed to confirm its identification and date, before entering the discussion about whether this early 2nd c. B.C. maritime colony of Roman citizens already received a Capitolium shortly after its foundation (on Capitolia, and their differential significance and backgrounds, cf. the contributions of Boos and Torelli in this volume). The very regular rectangular town plan of Potentia during this second phase thus displays a series of distinct characteristics of a maritime colony: the insulae oriented to the two perpendicular main street axes, crossing near the center, the forum gravitating to the southeast in the direction of the presumed harbor (as prescribed by Vitruvius). Its early history is certainly comparable to the general evolution of coastal colonial settlements such as the well-studied colony of Cosa.82 Finally, the still somewhat hypothetical third phase in the development of the town concerns an enlargement of the total inhabitable intra-mural urban space, possibly dated after the mid-1st c. B.C. Evidence for this theory is found in: I) Survey data indicating that the intra-mural town area was enlarged some 50 m to the east in a later stage of the settlement’s history.

77 78 79 80 81



Cf. Gros and Torelli 2007, 184. Salmon 1969, 16. Guidobaldi 1995. Vermeulen et al. 2011. Percossi Serenelli 2001; ead. 2012.

82

Gros and Torelli 2007, 174-5. In a way Potentia still follows the typical Mid-Republican system of city division, with insulae often covering 2:3 actus, but not organized per scamna, with a longitudinal axis parallel with the main cardo, but per strigatio.


II) Aerial photographs showing a larger and more pronounced tracée in the eastern part of the city, which may indicate that the original eastern side of the city wall was later replaced by a street. III) The fact that excavations in the north-eastern corner of the city have not revealed the presence of early structures.83 IV) A destruction layer in the temple sector dating to approximately the mid-1st c. B.C.84 This seems to confirm Cicero’s statement that a major earthquake destroyed Potentia in that period.85 This archaeological evidence could indicate that the town suffered heavily from this catastrophic event, and that a major reorganization and rebuilding was required in the decades after the earthquake. V) Traces of major rebuilding under Augustan rule; these have also recently been observed in the west gate area, where the entrance into the town was fully rebuilt.86 VI) The fact that the postulated enlargement and rebuilding of the town fits nicely with the abovementioned new assignations of territory and the arrival of new colonists during the second Triumvirate. This influx of people would have necessitated a larger town center. In that phase, the now better equipped urban center probably also incorporated a growing number of people that had formerly lived dispersed throughout the colonial territory, but were by then increasingly attracted by the amenities offered by the coastal town. Apart from these major urban developments in the general plan of the city, some further remarks can be made about the urban topography and functional zoning within and around the city limits. Inside the walled city, other public buildings located so far are: the excavated temple and macellum directly east of the forum, both with major Augustan building phases, a small theater near the eastern city wall and a possible bath complex in the southern part of the city (identified by survey but still to be dated precisely). The geophysical prospecting has helped to distinguish the remains and partial plans of many housing structures, some of the domus type, others clearly tabernae and other simple habitations and shops, interspersed between the many insulae in the urban center. Outside the walls, the three roads (detected from the air) leaving the city gates were bordered by cemeteries with funerary monuments facing the road. Recent excavations in the northern necropolis have uncovered a series of funerary monuments dating to the end of the Republic and to Augustan times.87 At least for a certain period, some of these immediate suburban areas were also partially used for manufacturing activities. Between circa 120 B.C. and A.D. 100, several flourishing amphora workshops, connected with the wine- and olive-oil-producing villas in the immediate hinterland, operated directly north and south of the city.88 Indications from the aerial survey of possible harbor facilities at the mouth of the River Potenza, just south of town, add to this general picture. Recent research has also highlighted aspects of the well-organized, ‘centuriated’ agrarian landscape beyond the city limits.89

Discussion and conclusion To sum up, recent work in northern Picenum and bordering eastern Umbria indicate that although historical data mention some disruption in the demography of the region, there was also an important aspect of continuity between the pre-Roman and early colonial communities. Evidence for urban developments, religious foci and rural transformation reveal that the adaptation to Roman culture between circa 300 B.C. and the era of Augustus was a slow process. In particular, the gradual or sporadic monumentalization of the early towns has been highlighted. Important stimuli behind these developments have been recognized in the aspirations of new colonists and veterans who settled in the region and in

83 84 85 86

Mercando 1979. Ibid. Cic., Har. resp. 28. 62. Vermeulen et al. 2011.

87 88 89

Percossi Serenelli 2007. Vermeulen et al. 2009; Monsieur 2011. Corsi 2008; Corsi and Vermeulen 2010.




the ‘self-Romanization’ of local elites. Some fragile evidence suggests that the latter phenomenon might have started already in the period immediately preceding the Roman conquest (5th to early 3rd c. B.C.). Most importantly perhaps, the new data shows that the urbanization process in this region did not follow the Roman conquest as abruptly as has been assumed, but seems to have been a more gradual process which might have commenced before any Roman involvement in the area, or was perhaps linked to less intrusive Roman activities in the area before the official colonization. Nevertheless, the fundamental role played by early Roman interventions, mostly the establishment of new towns for colonists, almost consistently located on the coast, and the massive settling of viritane colonists in many parts of the region after 232 B.C., should not be underestimated. The main phases that can be distinguished in the Potenza Valley on the basis of our archaeological investigations mirror general patterns of gradual urbanization and colonization in the wider Picenum area and in the easterly districts of Umbria and the ager Gallicus, which have been reconstructed on the basis of epigraphic evidence.90 The earliest colonization of the region east of the Apennines in the vicinity of the upper Potenza Valley seems to have been linked to the increased Roman intrusion in the Umbrian districts of Camerinum and Matilica, initially undertaken as a finalization of the conquest of Umbria around 300 B.C. and later following the construction of the Via Flaminia (completed in 220 B.C.). During the third and the first quarter of the 2nd c. B.C., the exposure of the region to Etrusco-Latial influences is evident and is most likely connected to early viritane settlement programs. Evidence of such tight links have been recognized in votive offerings and in an inscription (T. Apanei) on a Black Glazed vessel from the urban area of Matelica.91 In these still federate areas, the settlement dynamics during and immediately after the phase of the Roman conquest were guided mostly by internal dynamics. Centers like Matilica and Camerinum which are located in the vicinity of the upper Potenza Valley display typical protourban developments along indigenous lines at the time of their conquest. The central site is located on hilltops and the irregular plan of such ‘archaic towns’ seems to conform to this local topography. Houses continued to be built mainly of perishable materials deep into the 3rd c. B.C. 92 Only during the 2nd c. B.C. does a notable urban change become visible, and the influence of Rome is visible in the Hellenistic shift from Rusticitas to Urbanitas.93 In the more backward regions of inner Picenum, farther away from the Via Flaminia, the viritane colonists were the driving force behind the cultural contact with the conqueror and the development of new settlement systems. As the destiny of these settlements were linked to the developing secondary road system, the diverticula linking the main Flaminian artery to the coast, it is in the valley floors that their presence and the connected destructive effect on indigenous society must be sought. In these lower, fertile areas the strong presence of viritane colonists went hand-in-hand with the concession of citizenship to local populations, and with the creation of a system of praefecturae.94 At the same time, vici and conciliabula developed where more unofficial contacts between new and old inhabitants existed, and where the majority of colonists were probably organized in an old Roman-style pagus system.95 In this first colonial phase, the Picene hill sites surely continued to exist, but they started to create satellite settlements in the nearby valley floors where commercial activities and contacts with new cultural and religious influences were established. This is also the period in which the first Latin inscriptions appear in Picenum,96 all in the religious sphere, involving local magistrates (magistri, quaestores) who played a key role in religiously inspired acculturation processes.97 It is clear that in the inner ager Gallicus, in such

90 91 92 93



Paci 2008; Sisani 2007. Bandelli 2007, 14; Sisani, 2007, 184. Biocco 2000; Salvini 2002. Patterson 2006, 125; Sisani 2007, 232.

94 95 96 97

Ibid., 231. Capogrossi Colognesi 2000; Tarpin 2002. Paci 2008, 340. Stek 2009.


rural sanctuaries as Montefortino di Arcevia, teeming with ex votos dated to the 3rd and 2nd c. B.C.,98 and in the ager Picenus with the more Hellenistic-style sanctuary of Monte Rinaldo,99 religious centers were essential in this first phase of transition and contact with the Roman world.100 If the discovery of another major early religious site at Septempeda is confirmed (cf. above), it seems plausible that it functioned as a similar place of contact between the newcomers and the local population. Tentatively, also the fragile evidence for early settlement nuclei at Trea and Septempeda may be interpreted as the remnants of antenna settlements belonging to the indigenous central place (i.e. Monte Pitino), which fell gradually under the control of the Roman colonists and merchants in the course of the 3rd and 2nd c. B.C., possibly as part of a praefectura. In Septempeda, it is even possible to ask if the opus quadratum city wall had not already been built by the 2nd c. B.C. as an early ‘invitation to urbanism’, and that it, as is attested clearly for other places, took several generations to be filled with houses and public architecture.101 As can be seem in Trea and Ricina, the full urbanization of such roadside settlements only occurred in the course of the 1st c. B.C., generations after the urban developments in the coastal area. Between 184-174 B.C., when a Roman colony was founded and created at Potentia, a cautious but nevertheless incisive Roman impact on the coastal zone can be witnessed.102 This colony was founded in what was still a largely deserted wetland area where some early commercial activity concentrated near the Potenza river estuary. There must have been little resistance to Roman rule in this area, as it was geographically located between the larger Latin colony of Firmum (264 B.C.) to the south and the allied town of Ancona to the north. It served (as its sister town Pisaurum in the ager Gallicus) probably as a coastal defense and as a point of reference for the viritane colonists already settled in the interior shortly after the conquest, and in particular after the initiatives of Flaminius in the late 3rd c. B.C. Acculturation processes in the immediate hinterland of Potentia must have accelerated in this period as new commercial horizons via the sea were opened up. Much more epigraphic material becomes available during the following decades – some forty 2nd c. B.C. texts are so far known in the Marche regio – a corpus now also enlarged by commercial inscriptions (on containers and instrumentum) and architectural inscriptions which are linked to the early urbanization process.103 From the late 2nd c. B.C. onwards there is a more profound romanization of the coastal area steered by the investment of rich landowners from Rome and Latium (e.g. the gens Oppia) in wine- producing estates on hill slopes near the colony, as is testified by a handful of located amphora workshops and some inscriptions.104 This phenomenon was accompanied by an embellishment of the cult places of the city of Potentia, such as is attested, for example, by the terracotta decorations dating to this period found in the excavated temple105 – and also with an unmistakable demographic boom in this town.106 The insertion of local elites into the wider economic networks of the Adriatic and the East inevitably led to frenetic euergetism and to the monumentalization of the town, to ensure that the local politico-administrative organization conformed to that of Rome. In this period, Roman activity also seems to increase in 0the roadside settlements of the interior which developed around the new economic arteries in the valley floors, further weakening the importance of the indigenous hilltop centers. After 90 B.C. especially there was a strong impulse for urbanization in the interior with a marked shift of rural settlements to the valley bottom, towards the new centers. Some of the interior cities in Picenum originated from older vici or praefecturae which had grown up around sanctuaries, and which had still been in a proto-urban phase in the 2nd c. B.C.107 Thanks to the expansion of the road

98 99 100 101 102 103

Landolfi 1997. De Marinis and Paci 2000. Bandelli 2007. Cf. Interamna: Sisani 2007, 242. Salmon 1969, 178; Bandelli 2007, 17; Harvey Jr. 2006, 129. Paci 2008, 344.

104 105 106 107

Monsieur forthcoming. Percossi Serenelli 2009. Vermeulen 2008; id. 2012c. Cf. Cupra Maritima and Cupra Montana dedicated to the goddess Cupra, Pausulae to Apollo, probably also the patron of Septempeda.




network in this period (the first milliari date from this period) these centers could flourish.108 Nevertheless, apart from some exceptions in the Umbrian contact zone (e.g. Matilicum and Camerinum) inscriptions attesting to municipal foundations are rare before the mid-1st c. B.C.109 The most important period of municipalization was shortly after 49 B.C., when Caesar visited the area. At that moment, in the whole Marche region, which before that time was chiefly administered via praefecturae, some twenty municipia were created. In the period from the Triumvirate to Augustus this municipalization was followed by new assignations of land to veterans in the valley floors, which led to an economic boom and strong impulses towards urbanization supported by euergetism. The settling of large groups of veterans, between approximately 50 and 30 B.C., stimulated towns to develop a complete urban infrastructure.110 This process culminated in Augustan times with an even more vehement emphasis on politico-ideological messages and forms of communication, which can be read in some of the architectural improvements in the urban centers of the Potenza Valley. In this period, the new municipia of Septempeda, Trea and Ricina, as well as the colony of Potentia became places in which large flows of people and substances gathered. In this sense, a town can be seen as a station where activities cluster and where possibilities were created for other activities to occur in the close vicinity. Therefore, these towns not only provided functions and facilities to their inhabitants, they also structured the pattern of activities and the material residue it left in the wider landscape. Cities became recognized as central features in the Roman Empire and the preferred form of social, political and administrative form of organization. They provided Rome with a hierarchical governing framework,111 and with the location in which civilized people, especially those adopting a ‘Roman way of life’, could dwell.112 The growing wealth of the Picene elite and the gradual emergence of the towns as significant centers ensured the almost unbroken prosperity of municipal life during the 1st and large part of the 2nd c. A.D. From Augustan times onwards the towns in Picenum, colonies and municipia alike, were embellished with theaters, amphitheaters, Capitolia and other temples, basilicae, aqueducts, sewerage systems, cisterns, public baths and monumental gates.113 Originally created as rudimentary administrative centers, they grew into flourishing centers of population, and hence of marketing and exchange.

Acknowledgements The author wishes to thank the organizers and participants of the Nijmegen workshop for some stimulating discussions. I am also grateful to the two reviewers who made many useful suggestions about how this paper could be improved and to the many collaborators in the PVS team who have worked many years by my side gathering the new field data that form the basis of the various thoughts expressed here. The PVS research was primarily financed by the Belgian Science Policy (IAP programe), the Fund for Scientific Research – Flanders and Ghent University.

108 109 110



Paci 2008, 348. CIL XI 815. Keppie 1983.

111 112 113

Perkins and Nevett 2000. Whittaker 1997; Lomas 1998; Laurence 1998. Paci 1998; Luni 2004.


References Baldelli, G. 2004. “Civiltà picena e gallica,” in M. Luni (ed.), Archeologia nelle Marche dalla preistoria all’Età Tardoantica (Florence) 43-56. Bandelli, G. 2007. “Considerazioni sulla romanizzazione del Piceno (III-I secolo a.C.),” in Il Piceno romano dal III sec. a.C. al III sec. d.C. (Macerata) 1-26. Biocco, E. 2000. “Matelica,” in Atlante Tematico di Topografia Antica, VI Suppl., Città Romane, 1 (Rome). Blume, F., K. Lachmann and A. Rudorff 1848. Die Schriften der römischen Feldmesser (Berlin). Bonomi Ponzi, L. 1992. “Occupazione del territorio e modelli insediativi nel territorio Plestino e Camerte in età protostorica,” in La civiltà picena nelle Marche (Ripatransone) 204-41. Boullart, C. 2003. “Piceni Settlements: Untraceable or Neglected?,” Picus 23, 155-88. Boullart, C. and F. Vermeulen 2005. “The Potenza Valley Survey (Italy): in search of Piceni settlements,” in Proceedings of the UISPP-congres Liège 2001, Section 12 - The Iron Age in Europe (Oxford) 53-60. Branchesi, F. 2007. “Assetti fondiari del Piceno romano: dal III sec. a.C. al III sec. d.C.,” in Il Piceno romano dal III sec. a.C. al III sec. d.C. (Macerata) 183-237. Capogrossi Colognesi, L. 2000. Cittadini e territorio: consolidamento e trasformazione della ‘civitas Romana’ (Rome). Coarelli, F. 1988. “Colonizzazione romana e viabilità,” DialArch 6, 35-48. Conati Barbaro, C., A. Manfredini and M. Silvestrini 2003. “Dal Paleolitico all’età del Bronzo,” in M. Luni (ed.), Archeologia nelle Marche. Dalla preistoria all’età Tardoantica (Florence) 13-42. Corsi, C. 2008. “La centuriazione romana di Potentia nel Piceno. Nuovi approcci per una revisione critica e per una comprensione diacronica,” Agri Centuriati 5, 107-26. Corsi, C. and F. Vermeulen 2010. “Il contributo della cartografia storica per lo studio delle divisioni agrarie nella bassa valle del Potenza nel Piceno,” Agri Centuriati 7, 227-45. Dall’Aglio, P. L. and S. De Maria 2010. “Il territorio delle Marche e l’Adriatico in età romana,” BdA

Online 1, 39-48. Dall’Aglio, P. L. and P. Castagnoli (edd.) 2002. Sulle tracce del passato. Percorsi archeologici nella provincia di Pesaro e Urbino (Pesaro). Dall’Aglio, P. L., S. De Maria and A. Mariotti (edd.) 1991. Archeologia delle valli marchigiane Misa, Nevola e Cesano (Perugia). De Ligt, L. 2012. “Urban archaeology, urban networks and population dynamics in Roman Italy,” in F. Vermeulen et al. (edd.), Urban Landscape Survey in Italy and the Mediterranean (Oxford) 183-96. De Marinis, G. and G. Paci (edd.) 2000. Atlante dei beni archeologici dei territorio di Ascoli Piceno e Fermo (Cinisello Balsamo). De Marinis, G. and M. Silvestrini 2005. “L’Orientalizzante di Matelica,” in G. De Marinis et al. (edd.), Archeologia nel maceratese. Nuove acquisizioni (Macerata) 135-43. Delplace, C. 1993. La Romanisation du Picenum. L’exemple d’Urbs Salvia (Rome). Di Filippo Balestrazzi, E. 1999. “Le origine di Iulia Concordia,” in G. Cresci Marone and M. Tirelli (edd.), Vigilia di Romanizzazione. Altino e il Veneto orientale tre II e I sec. a.C. (Rome) 229-57. Fabrini, G. M. 2012. “Urbs Salvia dalle origini all’età augustea,” in G. de Marinis et al. (edd.), I processi formativi ed evolutivi della città in area adriatica (Oxford) 281-309. Giorgi, E. and G. Lepore (edd.) 2010. Archeologia nella valle del Cesano da Suasa a Santa Maria in Portuno (Bologna). Goethals, G., M. De Dapper and F. Vermeulen 2009. “Geo-archaeological implications of river and coastal dynamics at the Potenza river mouth,” in M. De Dapper et al. (edd.), Ol’Man River. Geoarchaeological aspects of rivers and river plains (Ghent) 407-38. Golvin, J. C. 1988. L’amphithéâtre romain (Paris). Gros, P. and M. Torelli 2007. Storia dell’urbanistica. Il mondo romano (Rome). Guidobaldi, M. P. 1995. La romanizzazione dell’ager Praetutianus (secoli III-I a.C.) (Naples). Harvey Jr., P. B. 2006. “Religion and memory of Pisaurum,” in C. E. Schultz and P. B. Harvey Jr. (edd.), Religion in republican Italy (Cambridge) 117-36.




Keppie, L. 1983. Colonization and veteran settlement in Italy, 47-14 B.C. (London). Laffi, U. 2001. “Linee di sviluppo della colonizzazione romana della fine della guerra annibalica all’età graccana,” in Iuris vincula (Naples) 519-30. Landolfi, M. 1997. “Montefortino di Arcevia,” in M. Pacciarelli (ed.), Acque, grotte e dei (Ravenna) 172-79. Landolfi, M. 2003. Il Museo Civico Archeologico di San Severino Marche (San Severino Marche). Laurence, R. 1998. “Territory, ethnonyms and geography. The construction of identity in Roman Italy,” in R. Laurence and J. Berry (edd.), Cultural identity in the Roman empire (London) 95-110. Letta, C. 1992. “I santuari rurali nell’Italia centroappenninica: valori religiosi e funzione aggregativa,” MÉFRA 104, 109-24. Lollini, D. G. 1958. “Notiziario. Scoperte e scavi preistorici in Italia durante il 1958. Marche,” Rivista di Scienze Preistoriche 8, 204-205. Lollini, D. G. 1979. “Il Bronzo Finale nelle Marche,” Rivista di Scienze Preistoriche 34, 179-215. Lomas, K. 1998. “Roman imperialism and the city in Italy,” in R. Laurence and J. Berry (edd.), Cultural identity in the Roman empire (London) 64-78. Luni, M. 2004. Archeologia nelle Marche dalla preistoria all’età tardoantica (Florence). Maioli, M. G., L. Malnati and M. Miari 2012. “La Romagna adriatica tra VI e II secolo a.C.,” in G. De Martinis et al. (edd.), I processi formativi ed evolutivi della città in area adriatica (Oxford) 71-83. Marengo, S. M. and G. Paci 2004. “Recenti acquisizioni storico-epigrafiche nel Maceratese,” Studi Maceratesi 38, 297-319. Mercando, L. 1971. “Macerata, frazione Villa Potenza - Rinvenimento in proprietà AGIP,” NSc 9, 402-17. Mercando, L. 1979. “Marche. Rinvenimenti di insediamenti rurali,” NSc 33, 89-296. Mercando, L., L. Bachielli and G. Paci 1984. “Monumenti funerari di Ricina,” BdA 28, 11-52. Monsieur, P. 2011. Amphores adriatiques et tyrrhéniennes des époques tardo-républicaine et impériale précoce découvertes à Pessinonte (Oxford). Monsieur, P. forthcoming. “New Light on the production of Central Adriatic Amphorae. The Results of Survey and Excavation in the Ager Potentinus (Marche, Italy),” JFA.



Naso, A. 2000. I Piceni. Storia e archeologia delle Marche in epoca preromana (Milan). Paci, G. 1998. “Sistemazione dei veterani ed attività edilizia nelle Marche in età triumvirale-augustea,” Memorie dell’Accademia Marchigiana di Scienze Lettere ed Arti di Ancona 33, 209-44. Paci, G. 2008. Ricerche di storia e di epigrafia romana delle Marche (Tivoli). Patterson, J. R. 1987. “Crisis. What crisis? Rural change and urban development in Imperial Appenine Italy,” PBSR 55, 115-46. Patterson, J. R. 2006. Landscapes and cities: rural settlement and civic transformation in early imperial Italy (Oxford). Percossi, E., G. Pignocchi and F. Vermeulen (edd.) 2006. I siti archeologici della Vallata del Potenza (Ancona). Percossi Serenelli, E. 1989. “Rinvenimenti ed emergenze archeologiche nel territorio dell’antica Ricina,” Picus 9, 65-117. Percossi Serenelli, E. (ed.) 1999. Il territorio di Recanati dalla preistoria all’età romana (Loreto). Percossi Serenelli, E. (ed.) 2001. Potentia. Quando poi scese il silenzio…: Rito e società in una colonia romana del Piceno fra Repubblica et tardo Impero (Porto Recanati). Percossi Serenelli, E. 2003. “Le necropoli di Recanati e Pollenza (VII-IV sec. a.C.) e il popolamento della vallata del Potenza,” in I Piceni e l’Italia medio-adriatica: atti del 22. Convegno di studi etruschi ed italici: Ascoli Piceno, Teramo, Ancona, 9-13 aprile 2000 (Pisa/Roma) 605-35. Percossi Serenelli, E. 2007. “La necropoli di Potentia. Nuovi rinvenimenti,” Studi Maceratesi 61, 547-72. Percossi Serenelli, E. 2009. “Su alcuni tipi di terrecotte architettoniche da Potentia,” in G. De Marinis and G. Paci (edd.), Omaggio a Nereo Alfieri. Contributi all’archeologia marchigiana (Tivoli) 439-90. Percossi Serenelli, E. 2012. “Le fasi repubblicane di Potentia,” in G. De Martinis et al. (edd.), I processi formativi ed evolutivi della città in area adriatica (Oxford) 309-31. Perkins, P. and L. Nevett 2000. “Urbanism and urbanisation in the Roman world,” in J. Huskinson (ed.), Experiencing Rome. Culture, identity and power in the Roman empire (London) 213-44.


Perna, R. 2006. Urbs Salvia (Rome). Perna, R. 2012. “Nascita e sviluppo della forma urbana in età romana nelle città del Piceno e dell’Umbria adriatica,” in G. De Martinis et al. (edd.), I processi formativi ed evolutivi della città in area adriatica (Oxford) 375-413. Salmon, E. T. 1969. Roman colonization under the Republic (London). Salvini, M. 2002. Il museo civico archeologico di Camerino (Pescara). Sisani, S. 2007. Fenomenologia della conquista. La romanizzazione dell’Umbria tra il IV sec. a.C. e la guerra sociale (Rome). Sommella, P. 1988. Italia antica. L’urbanistica romana (Rome). Stek, T. D. 2009. Cult places and cultural change in Republican Italy. A contextual approach to religious aspects of rural society after the Roman conquest (Amsterdam). Stek, T. D. and J. Pelgrom 2005. “Samnite sanctuaries surveyed. Preliminary report of the sacred landscape project 2004,” BABesch 80, 65-71. Tarpin, M. 2002. Vici et pagi dans l’Occident romain (Rome). Vanzetti, A. 2004. “Risultati e problemi di alcune attuali prospettive di studio della centralizzazione e urbanizzazione di fase protostorica in Italia,” in P. A. J. Attema (ed.), Centralization, early Urbanisation and Colonization in First Millennium B.C. Italy and Greece. Part 1: Italy (Louvain) 1-28. Vermeulen, F. 2005. “La media valle del Potenza in età romana: da Trea a Helvia Ricina,” in G. De Marinis et al. (edd.), Archeologia nel maceratese. Nuove acquisizioni (Macerata) 180-88. Vermeulen, F. 2008. “Functional zoning and changes in the use of space in the Roman town of Potentia. An integrated survey approach,” in H. Vanhaverbeke et al. (edd.), Thinking about Space. The Potential of Surface Survey and Contextual Analysis in the Definition of Space in Roman Times (Turnhout) 233-49. Vermeulen, F. 2012a. “Integration of survey, excavation and historical data in Northern Picenum,” in P. A. J. Attema and G. Schörner (edd.), Comparative issues in the archaeology of the Roman rural landscape. Site classification between survey, excavation and historical categories (Portsmouth, RI) 43-55.

Vermeulen, F. 2012b. “Potentia: a lost new town,” in N. Christie and A. Augenti (edd.), Urbes Extinctae. Archaeologies of abandoned classical towns (Farnham) 77-95. Vermeulen, F. 2012c. “Topografia e processi evolutivi delle città romane della valle del Potenza (Picenum),” in G. De Martinis et al. (edd.), I processi formativi ed evolutivi della città in area adriatica (Oxford) 331-45. Vermeulen, F. and F. Carboni 2010. “Measuring urbanisation in the Late Republican and Early Imperial landscape of Adriatic Central Italy,” BdA Online 1, 80-90. Vermeulen, F. et al. 2003. “The Potenza Valley Survey: preliminary report on field campaign 2002,” BABesch 78, 71-106. Vermeulen, F. et al. 2009. “Investigating the impact of Roman urbanisation on the landscape of the Potenza Valley. A report on fieldwork in 2007,” BABesch, 85-110. Vermeulen, F. et al. 2011. “Excavations of the West gate of Potentia: a preliminary report,” Picus 31, 169-205. Vermeulen, F. and D. Mlekuz 2012. “Surveying an Adriatic Valley: a wide area view on early urbanization processes in northern Picenum,” in F. Vermeulen et al. (edd.), Urban landscape survey in Italy and the Mediterranean (Oxford) 207-22. Vermeulen, F., B. Slapšak and D. Mlekuz 2012. “Surveying the townscape of Roman Trea (Picenum),” in P. Johnson and M. Millett (edd.), Archaeological Survey and the City (Oxford) 26182. Vermeulen, F. and G. Verhoeven 2004. “The contribution of aerial photography and field survey to the study of urbanisation in the Potenza valley (Picenum),” JRA 17, 57-82. Whittaker, C. R. 1997. “Imperialism and culture. The Roman initiative,” in D. J. Mattingly (ed.), Dialogues in Roman imperialism. Power, discourse, and discrepant experience in the Roman empire (Portsmouth, RI) 143-63. Williams, J. H. C. 2001. “Roman intentions and Romanization. Republican northern Italy, c. 200 - 100 BC,” in S. Keay and N. Terrenato (edd.), Italy and the West. Comparative issues in Romanization (Oxford) 91-101.




FIG. 1. A pertica (highly) schematized. Vatican, Palatinus 1564. Fig. 136a Th, 197a L (drawing by M. Tarpin).


Strangers in Paradise. Latins (and other non-Romans) in colonial context: a short story of territorial complexity Michel Tarpin

Introduction The treatises composed by the Roman land surveyors present the reader with a detailed view of the territorial and legal organization of Roman colonies. Regrettably, these texts are so difficult to understand that, rather than clarifying matters, they often complicate and confuse our understanding of Roman colonial territorial organization. Some scholars even doubt whether they reflect any administrative reality at all. To some extent, one can blame the medieval copyists for the inaccuracies and errors in the manuscripts as they did not always understand what they were copying. Recently, Gérard Chouquer argued convincingly that the confusion is primarily the result of Roman land surveyors trying to combine legal concepts, historical facts and local metrology.1 The famous picture of a colonial pertica in the Palatinus 1564 manuscript (fig. 1) is a good example of this complex process. What is depicted is an attempt to capture, in one single image, the long history of a colonial territory. The unfeasibility of such an attempt becomes clear if one understands the turbulent and dynamic history of such colonized landscapes.2 To give one example, according to Pliny, the Ligures Ingauni, a small community in Northern Italy, was granted land thirty times, a situation impossible to record on a single map.3 Once one understands the long, complex legal and social history of the colonial lands, one understands the difficult tasks the Roman land surveyors had and starts to appreciate the attempts they made. This article gives a few examples showing how migration and the political integration of diverse groups or individuals might have contributed to the wide variety of situations the land surveyors tried to describe and that, in essence, these situations reflect pragmatic choices made by the Senate when confronted with different circumstances. The focus will mainly be on Latin migration in colonial contexts, as well as on individual naturalization. Afterwards, returning to the Palatinus map, it can be seen that the map is a simplification of what the real situation must have been.

The Latins’ ‘privileged relationship’ with Rome Latin colonists of either Roman or Latin origin were formally peregrines, citizens of communities linked to Rome probably by a foedus. Their legal relation to Rome was probably stipulated in the colonial foundation laws.4 Such laws may also have simplified political integration into the Roman citizen body. In

1

2

3

Chouquer 2008, 204. See also http://www.archeogeographie.org/index.php?rub=arpentage/romain/orange/interphisto. On this see Roselaar 2010, 298. More recently, Lorenzo Gagliardi (Gagliardi 2011; 2014) has focused on the juxtaposition of statuses that can be found in the same colonial territory. Plin., HN 3.5.46.

4

Contra Chastagnol 1995, passim and Gascou 1999, 294, note 4 (with complementary bibliography), who see Latins as normal peregrines. Of course, as Latium was not a city, there was no unique ‘Latin citizenship’. Nevertheless, legal texts that distinguish between ‘normal’ peregrines and Latins do exist. After the lex Iunia was passed, a new distinction was added between Junian Latins (i.e., freedmen e lege Iunia) and Latini coloniarii, who were freeborn. Cf.




modern literature usually three privileges are supposed to have facilitated, if not full political integration, at least their legal relationship with Rome: commercium, conubium and the so-called ius migrandi. These are all hotly debated topics for which very little reliable historical information exists.5 According to Gaius, in the Republican period, conubium allowed a male Roman citizen to ensure that his children would receive full citizenship if their mother was a Latin or peregrine from a community with a lex or a clause in a foedus that stipulated conubium. The children also fell under the potestas of the father and could therefore be legal heirs.6 According to the sources, this traditional right of intermarriage with Latins can be traced back to the Regal period; an indication that it was considered a legitimate and ancient right. Perhaps it was even one of the clauses in the treaties of the Latin League or in the foedus Cassianum.7 Certainly conubium had once existed between the Latin cities and was abolished by Rome in 338 B.C. (only conubium with Rome remained).8 In the same way Aemilius Paullus deprived the four new Macedonian regions of conubium et commercium inter se in 167 B.C.9 Commercium is a much-debated privilege.10 It certainly included the right to transfer full ownership by

5 6



Gai., Inst. 1.29, 2.56: “cives Romani ingenui qui ex urbe Roma in Latinas colonias deducti Latini coloniarii esse coeperunt”. Gaius (Gai., Inst. 1.79) also contrasts “qui Latini nominantur” (Junian Latins) and “alios Latinos […] qui proprios populos propriasque civitates habebant et erant peregrinorum numero” (“colonial Latins”). Cf. Cod. Iust., 7.6.1: “Cum enim Latini liberti ad similitudinem antiquae Latinitatis, quae in coloniis missa est, videntur esse introducti”. And: Gai., Inst. 3.56: “Postea uero per legem Iuniam eos omnes, quos praetor in libertate tuebatur, liberos esse coepisse et appellatos esse Latinos Iunianos: Latinos ideo, quia lex eos liberos proinde esse uoluit, atque si essent ciues Romani ingenui, qui ex urbe Roma in Latinas colonias deducti Latini coloniarii esse coeperunt; Iunianos ideo, quia per legem Iuniam liberi facti sunt, etiamsi non essent ciues Romani.” See Broadhead, especially 2001 and 2004. Cf. Gai., Inst. 1.56: “si ciues Romanas uxores duxerint uel etiam Latinas peregrinasue, cum quibus conubium habeant: cum enim conubium id efficiat, ut liberi patris condicionem sequantur, euenit, ut non solum ciues Romani fiant, sed et in potestate patris sint”. (“Roman citizens who have contracted marriage according to the Civil Law and who have produced children are understood to have those children in their power if they marry Roman citizens, or even Latins or foreigners whom they have the right to marry; for the result of legal marriage is that the children follow the condition of the father and not only are Roman citizens by birth, but also become subject to paternal authority [patria potestas].”) (transl. De Zulueta 1946). Gai., Inst. 1.76-77. Even under the Empire, the edict of Claudius in favour of Volubilis shows that conubium with peregrines was

7

8

9

10

a feature of the Roman municipium. ILM 116: “M. Val(erio) Bostaris | f(ilio), Gal(eria), Seuero […]. Huic ordo municipii Volub(ilitani) ob merita | erga rem pub(licam) et legationem |bene gestam, qua ab divo | Claudio civitatem Romanam | et conubium cum peregrinis | mulieribus, immunitatem | annor(um) X, incolas, bona civium bello interfectorum, | quorum heredes | non extabant, suis impetrauit. […]”. Coşkun 2009, 35-39. Initially, conubium might have been limited to the XII coloniae. After the lex Minicia, sons inherited the lesser right (Paulus, Sent. 5.8). See discussion in Gai., Inst. 1.78-79. Broadhead 2001, 81. Liv. 8.14.10: “ceteris Latinis populis conubia commerciaque et concilia inter se ademerunt”. However, the Hernici, namely Aletrium, Verulanum, and Ferentinum, later obtained the right to maintain its old laws and conubium inter ipsos (Liv. 9.43.23-24) as a privilege. Liv. 45.29.10: “pronuntiavit deinde neque conubium neque commercium agrorum aedificiorumque inter se placere cuiquam extra fines regionis suae esse”. Hartmut Galsterer (Galsterer 1995, 90) sensibly notes that conubium was a prerequisite for the Roman citizenship granted to a local Latin government (ius civitatis adipiscendae per magistratum). Otherwise a number of marriages in the Latin cities would have been null and void. Coşkun (2009, 39-58) thinks that the ius commercii is a modern concept, defined by the verbs ‘to give’ or ‘to remove’ attached to commercium. As usual, because of the lack of Republican sources, reference has to be made to Imperial-era lawyers. Nonetheless, the rules governing commercium emphasize the co-existence of communities of different status, a situation better suited to the Italian Re-


mancipatio (originally per aes et libram). Commercium still appears in Imperial legal texts after formal mancipatio seems to have disappeared. For example, a prodigus (spendthrift) was deprived of commercium and therefore could not sell his slaves by mancipatio, which is definitely a restriction on his freedom as a citizen.11 Mancipatio apparently remained important as far as wills were concerned. It is possible that Rome granted the Latins and some allies not only the freedom to practice a trade protected by law in Rome, but also the right to inherit from a Roman or appoint a Roman as heir.12 Umberto Laffi draws attention to the fact that commercium was an essential precondition for the fraudulent naturalization mentioned by Livy in 177 B.C.13 It is usually assumed that the Latins formally lost commercium inter se in 338 B.C. which would imply they lived for over two centuries without being able to make legal sales contracts that awarded full ownership.14 Will Broadhead convincingly explains this harsh measure as a Roman strategy to monopolize commerce. Rome had a financial interest in forbidding commercium between Latin cities, while maintaining unilaterally its own commercium with each city.15 The right of commercium might also occasionally be granted to allow peregrines to make legal purchases in Rome. For example, in the early 2nd c. B.C., in addition to the usual diplomatic gifts, the Senate granted some Gallic ambassadors temporary commercium for specific products they wanted to buy in Rome.16 Cicero states that Sulla allowed the cities he had deprived of Roman citizenship to retain the legal categories of nexum and hereditates,17 which were probably the most essential elements of commercium. No doubt it would have been impossible to abolish these rights without creating widespread confusion throughout Italy. Admittedly nexus (or nexum) is a rather shadowy term that is difficult to define, but it certainly was not the archaic right to enslave a debtor, which was probably only one aspect of nexus. Altay Coşkun rightly supposes that its meaning may have extended to a form of personal guarantee on contracts.18 Mancipatio might have been a special category of nexus.19

11

public. The simplest definition given by the Tit. Ulp., 19.5: “Commercium est emendi vendundique invicem ius” (“commercium is the right to buy and sell reciprocally”). Commercium is directly related to mancipatio, the definition of which is less obvious (cf. http://remacle.org/bloodwolf/Gracques/mancipium.htm); Tit. Ulp., 19.3: “Omnes res aut mancipi sunt aut nec mancipi. Mancipi res sunt praedia in Italico solo, tam rustica, qualis est fundus, quam urbana, qualis domus; item iura praediorum rusticorum, velut via, iter, actus, aquaeductus; item servi et quadrupedes, quae dorso collove domantur, velut boves, muli, equi, asini; ceterae res nec mancipi sunt. Elefanti et cameli, quamvis collo dorsove domentur, nec mancipi sunt, quoniam bestiarum numero sunt”. There is endless debate about the question of the “rights of the 12 colonies” regarding commercium. At the moment, it adds nothing to the discussion of the nature of commercial relations between Latins and Romans. Tit. Ulp., 20.13. Insofar as a person with no right of mancipatio might have had the opportunity to perform in iure cessio, it must be inferred that some sales had full legal validity only if they were made by mancipatio. Had this not been so, the restriction placed on the prodigus would have been meaningless. See what happened to the son

12 13 14

15 16

17 18

19

of Q. Fabius Maximus Allobrogicus (Val. Max. 3.5.2). Contra Coşkun 2009, 62. Infra, pp. 169-70; Laffi 1995, 54. Cf. Coşkun 2009, 40. He argues that the absence of references to either commercium or mancipatio in the Carthaginian treaties suggests that this right did not exist at that time. However, these treaties regulated very specific circumstances: the Carthaginians were not Italians and had no hierarchical relationship with Rome or conubium. Above all, the treaties with Carthage were intended to prevent interaction between the two cities; in fact their purpose is precisely the opposite of commercial treaties. Broadhead 2001, 81. Liv. 43.5.9: “illa petentibus data, ut denorum equorum iis commercium esset educendique ex Italia potestas fieret”. The combination of commercium and the right to export the horses is notable. Cic., Caecin. 35.102. Coşkun 2009, 42-43; Festus, Gloss. Lat. 426-28, 474, 74, 91. On nexus in the Law of the XII Tabulae 1.5, see Crawford 1996, 555, no. 40. See http://remacle.org/bloodwolf/Gracques/mancipium. htm#texte. Cic., Top. 5.28: “Abalienatio est eius rei quae mancipi est aut traditio alteri nexu aut in iure cessio




Roman foedera seem to have included a status called exilium, the right to give refuge to a Roman criminal without fear of prosecution. Polybius refers to two major Latin cities, Praeneste and Tibur (see fig. 3. in the contribution of Terrenato in this volume) and to a Greek one, Naples, and writes that all allied cities could offer refuge to an exile.20 One can only agree with Edward Bispham that exilium represented a significant gesture of trust toward a city that had the right to host a Roman criminal.21

On the supposed collective change from Latin to Roman citizenship Relying primarily on indirect references in Livy, most ancient historians accept that it was possible for Latins to acquire civitas Romana by residing in Rome. In modern scholarship, this ‘right’ is called the ius migrandi or ius migrationis. Some historians doubt that a genuine right – which indeed has never been verified22 – existed, while others argue that it was a very ancient custom23 that was, in later times, occasionally abused.24 Still others believe that Latins may have benefited from temporary flexibility in the Roman census.25 The ius migrandi would have given a Latin the right to acquire civitas Romana by settling and registering in Rome. The ius migrandi has sometimes been associated with the so-called right ‘of Ariminum’, about which little is known.26 The ius migrandi must have differed in some way from the ius suffragii.27 Even if ever fewer scholars believe that there actually was a ius migrandi, they cannot deny that there were significant waves of migration of Latins to Rome in the first quarter of the 2nd c. B.C. Therefore, the best way to confirm or refute the hypothesis that such a specific right existed would be to examine in chronological order episodes of population movement between the capture of Capua – and the condemnation and exile of the Campani – and the last book of Livy. In 210 B.C., after a plebiscite, the Roman Senate decreed the restoration of the freedom of the Campani Atellanes, Calatini, and Sabatini with the exception of those who had been condemned and with the stipulation that they could not become Roman or Latin citizens.28 Since these Campanians were not sent to colonies but

20 21

22

23

24 25



inter quos ea iure civili fieri possunt.” (“Alienation is the surrender of anything which is a man’s private property, or a legal cession of it to men who are able by law to avail themselves of such cession”) (transl. Yonge, 1856). Cic., Har. Resp. 7. One simple definition of nexus, attributed to Manilius by Varro (Varro, Ling. 7.105) is: “Nexum Manilius scribit omne quod per libram et aes geritur, in quo sint mancipia”. See Agen. Urb., De contr. 62 L., 23 Th: “qui agri (provinciales) nexum non habent”. Cf. Saumagne 1965, 80-85. Polyb. 6.14.8. Bispham 2009, 133; see also the detailed analysis of Coşkun 2009, 74-82. Broadhead 2001; id. 2004; Coşkun 2009, passim, does not reach a clear conclusion. E.g., Martin 2001, 74, who quotes the precedents of Kings Tarquinius and Servius. Galsterer 1995, 82. Laffi 1995, 51-52, 77; Ferrary 2003. Laffi proposes to limit the right of integration into Rome to Latins of Roman origin (Laffi 1995, 52), but there is no evidence of such a criterion. In contrast to the general hypoth-

26

27

28

esis of Kremer 2006, I am not sure that the Senate made a legal distinction between Latins of Roman origin and Latins of Latin (or peregrine) origin. See the critical analysis of Broadhead 2001. Cf. Salmon 1936, 56-57. Contra: Laffi 1995, 51. These twelve colonies are mentioned by Cicero (Caecin. 102), but he does not say what the ius duodecim coloniarum was. Another problem is that one of the colonies is Ariminum: more than eleven colonies preceded Ariminum and there should have been twelve after it, if Luca was a Latin colony, which is not always accepted. See Broadhead 2001, 76-77; Coşkun 2009, 64-69. App., B Civ. 1.23. This practice was imitated later when incolae were allowed to establish themselves in a colony to vote in a curia. E.g., the lex municipii Malacitani, § 53. Liv., 26.34.6-7: “Campanos omnes Atellanos Calatinos Sabatinos, extra quam qui eorum aut ipsi aut parentes eorum apud hostes essent, liberos esse iusserunt, ita ut nemo eorum civis Romanus aut Latini nominis esset, neve quis eorum qui Capuae fuisset dum portae clausae essent in urbe agrove Campano intra certam


were required to reside under compulsory residence orders in different places on the ager Romanus, the most obvious conclusion that can be drawn from this passage is that being ‘Latin’ implied specific individual rights, regardless whether one belonged to a specific city. We may also presume that the Roman magistrates sent a copy of the decree to all the Latin communities and subsequently compiled lists of people convicted in Rome, so that the convicted would have been prevented from enrolling in a colony. Since before 204 B.C., the Latin census was archived in each city according to their local rules, the Senate had no direct control over whether the allies and Latins incorporated individual Campanians in their cities before this time. The next relevant thing in this context regards the problems the colonies of Placentia and Cremona experienced in maintaining stable population levels. Founded on the Gallic frontier in 218 B.C., these Latin colonies were immediate targets of the Gauls. In 206 B.C., ambassadors from these cities complained that their citizens had been leaving and hence the cities were becoming deserted. The Senate accordingly ordered the colonists to return to their colonies, now protected by a Roman army.29 This measure applied only to Placentini and Cremonenses who had remained free and fled to safer places.30 Others captured by the Gauls, once they were brought to the enemy’s praesidia, were no longer citizens of their former colonies.31 It took Rome many years to finish the task of freeing the captives and to round up the fugitive colonists. In 200 B.C., the praetor M. Furius Purpureo crushed the Gauls who had taken Placentia and returned the captive colonists to their colonies.32 Still, two years later, Sextius Aelius spent most of the year bringing back dispersed colonists.33 In my opinion they must have been free colonists because he had to force them to return. It is furthermore recorded that in 197 B.C., the citizens of the two colonies showed their support for Cornelius Cethegus during his triumph, following his chariot pilleati as freedmen.34 Although there is some confusion about whether Cethegus had been their liberator (the alternative candidate is Purpureo),35 these ex-

29

30

diem maneret; locus ubi habitarent trans Tiberim qui non contingeret Tiberim daretur” (“[The senate] ordered that all Campani, Atellani, Calatini, Sabatini, except those who had sided with the enemy, either themselves or their fathers, were to be free men, with the reservation that no one of them should be a Roman citizen or reckoned a Latin, and that no one of them who had been at Capua while the gates were closed should remain in the city or in the territory of Capua beyond a certain date; that a region across the Tiber, but not touching the Tiber, be given them as a dwelling-place”) (transl. Moore 1934, with modifications). This clause is probably very different to that for the Insubres, Cenomani, and Helvetii, because it does not offer foreign communities protection and gives no access to civitas Romana (Ferrary 2003, 110; Sanchez 2007). Liv. 28.11.11. The citizens had to return to their colonies before a certain day, exactly the same rule applying to soldiers at the beginning of a campaign. As quoted by Coşkun, the colonists fled not only to Rome but undique, and nobody says they became Romans. The Placentini and Cremonenses therefore are not an example of people subject to the ius migrandi (Coşkun 2009, 159).

31

32

33

34

35

Admittedly, such reconstruction of events is based on Imperial law (Cf. Pompon., Dig. 49.15.5), but seems to be supported by the provisions laid down in the second Romano-Carthaginian treaty which indicate that the soil played an important role in the loss or recovery of citizenship (cf. Polyb. 3.24.6). Liv. 31.21.18: “Placentini captivi ad duo milia liberorum capitum redditi colonis”. Liv. 32.26.3: “[Sex. Aelius] totum prope annum Cremonensibus Placentinisque cogendis redire in colonias, unde belli casibus dissipati erant, consumpsit.” Liv. 33.23.6: “ceterum magis in se convertit oculos Cremonensium Placentinorumque colonorum turba, pilleatorum currum sequentium.” Livy does not mention the release of captives after Cethegus’ victory in 198 B.C. (Liv. 32, 30). Cornelius Cethegus did not free Placentia and Cremona, which would had been liberated previously by Purpureo, Their populations would have been restored in 199-98 B.C. Moreover, Hamilcar, who died facing Purpureo, was alive the next year when he walked before Cethegus’ chariot. Livy indicates that the sources for the first years of the 2nd c. B.C. in Gaul are not unanimous (cf. Liv. 33.36.15).




captives were exhibited in Rome in honor of the general who rescued them, wearing freedmen’s caps. These former citizens of Cremona and Placentia probably returned to their colonies after the parade, under the terms of the above discussed senatorial decree. I believe that their wearing a pilleus during the triumphal procession was more than just a symbolic gesture: it was a demonstration pregnant with meaning. The exact terms of postliminium under the Republic are unknown but it is plausible that some of the Imperial definitions given by Paulus and Pomponius could date to the Republic.36 No text gives the procedure for reinstatement of a released captive in a city, but there are reasonable grounds for conjecturing that the former colonists, who may have spent some time as slaves of the Gauls, were brought to the free Latin colonists of Placentia and were immediately freed.37 The author’s view is that, except in the situation of a reditus in eodem bello, a man captured and enslaved needed to be freed to regain his citizenship.38 As Paulus says, only those who were captured by pirates or bandits – who were not considered real hostes and had no real praesidia – retained their free status (the young Caesar in Cilicia, for instance). In other words, they were not considered to have been taken captive.39 It is clear, however, that only a limited number of captured colonists were freed because, in 190 B.C., the two cities asked for the dispatch of a new contingent of colonists. The consul C. Laelius enrolled six thousand families to settle there.40 It can be reconstructed from these events that in the period between 206 and 190 B.C., Cremona and Placentia were repopulated by people with different backgrounds including old colonists who had remained free during the war, by colonists who had been reintegrated after their captivity and, finally, by new colonists. In 199 B.C., at the beginning of the censorship of Scipio Africanus and P. Aelius Paetus, two cities asked that their populations be increased. The request of Cosa was refused, the city having to wait until 197 B.C.,41

36

37



Pompon., Dig. 49.15.5.2: “In bello, cum hi, qui nobis hostes sunt, aliquem ex nostris ceperunt et intra praesidia sua perduxerunt: nam si eodem bello is reversus fuerit, postliminium habet, id est perinde omnia restituuntur ei iura, ac si captus ab hostibus non esset. Antequam in praesidia perducatur hostium, manet civis. Tunc autem reversus intellegitur, si aut ad amicos nostros perveniat aut intra praesidia nostra esse coepit.” (“In war, when those who are our enemies seize one of us, and take him within their fortifications, for if he returns during the same war, he will have the right of postliminium; that is to say, all his rights will be restored to him, just as if he had not been captured. Before he is taken into the fortifications of the enemy, he remains a citizen, and he is understood to have returned if he comes to our friends, or within our defences”) (transl. Scott 1932). Repeated by Paulus, Dig. 49.15.19.pr. Postliminium was a legal answer to a very old question: what should a city do with people and things returning from an alien country? But note the precision of the clause si eodem bello is reversus fuerit. The Placentini and Cremonenses were not freed “in the same war” in which they had been captured. The procedure of postliminium under the Republic is not described by the jurists, but a parallel can be drawn with the fictitious sale of emancipation (Inst. Iust. 1.12.6). Pertinent in this instance are the thousands of Roman

38 39

40 41

captives sold in Greece by Hannibal and bought back to be freed by the Greeks allies, mostly the Achaeans, on the orders of Titus Flamininus (Liv. 34.52.12). Their heads shaven, they paraded behind the triumphal chariot. The Placentini and Cremonenses, too, followed the chariot like soldiers, an indication that they were freed during the war and recruited (as allies?). The terms of the second Romano-Carthaginian treaty state that, if an Italian captured by the Carthaginians entered a Roman harbour and if a Roman took him under his protection, he could return home (Polyb. 3.24.6). ἐπιλαμβάνω in Polybius is commonly interpreted as a kind of manumissio (contra Coşkun 2009, 85). See the Romano-Lycian treaty (R. Pintaudi, ed., Papyri Graecae Schoyen I. Papyrologica Florentina XXXV, Firenze, 2005, 161-258): “ἐάν Λὺκιος ἐκ πολεμίων ἀνασωθεὶς εἰς Ῥώμην παραγένεται ἐλεύθερος ἔστω, ὡς ὁμοίως δὲ καὶ Ῥωμαίοι τὸ αὐτὸ δίκαιον ἐν Λυκίαι ἔστωι”. Cf. Coşkun 2009, 100. Paulus, Dig. 49.15.19.2: “a piratis aut latronibus capti liberi permanent”. Cf. also Ulp., Dig. 49.15.24. On postliminium, see Ando 2008, 503-4. Liv. 37.46.9-10. Liv. 33.24.8-9: “Cosanis eodem anno postulantibus, ut sibi colonorum numerus augeretur, mille adscribi iussi, dum ne quis in eo numero esset, qui post P. Cornelium et Ti. Sempronium consules (204 B.C.) hostis fuisset.”


but the request of Narnia was granted.42 Deryck J. Piper argues that the reluctance of the Roman Senate to grant both cities their request is probably the result of the war and of extensive colonial movement which made it too difficult to recruit enough new colonists for both cities.43 Another explanation, however, might be more convincing. Narnia was one of the rebel colonies during the war and had been punished. As a result, in 204 B.C., its census was taken over by the Roman censors44 who would immediately see if the number of colonists in the city was really dwindling. Conversely, in Cosa, a ‘loyal’ colony, the census was conducted by local censors and conveyed to Rome after the conclusion of the Roman census which happened precisely in 197 B.C. Furthermore, there were apparently no triumvirs for Cosa and a condition was imposed on the registration of new settlers: those who had been traitors during the Hannibalic War were explicitly excluded. These two points suggest that the recruitment of new colonists was conducted by the local censors. Since the Campanian traitors were settled on the right bank of the Tiber, tactically speaking, it would have been better to prevent the magistrates of Cosa from welcoming them. Narnia was an entirely different case. Although the colony suffered a shortage of colonists, it was also home to a foreign population that was acting ‘like colonists’. Livy does not say whether the triumvirs were content to integrate these non-colonist inhabitants or whether they sought and found new settlers in Rome. The expression pro colonis se gerere recalls the Pro Archia, in which Cicero says that Archias could be recognized as a Roman because se iam tum gessisse pro cive.45 From a lawyer’s perspective this is reasonably good evidence of citizenship; it arouses the suspicion that the triumvirs enrolled at least some peregrini, the majority of them certainly Italians.46 The following complicated case further illustrates that there was considerable migratory movement and integration at the beginning of the 2nd c. B.C. In 194 B.C., shortly before the definitive dispatch of colonists to Puteoli, Salernum and Buxentum, the Senate was confronted with a petition from ‘Ferentinates’ demanding that Latins who had signed up for a Roman colony were immediately granted Roman citizenship.47 It cannot

42

(“The same year, the people of Cosa requested that the number of their colonists be increased; one thousand were ordered to be enrolled, with the provisio that no one should be included in the number who had been an enemy since the consulship of Publius Cornelius and Tiberius Sempronius.”) (transl. Sage 1935, modified). Liv. 32.2.6-7: “Et Narnienses legatis querentibus ad numerum sibi colonos non esse et immixtos quosdam non sui generis pro colonis se gerere, earum rerum causa tresviros creare L. Cornelius consul iussus. […] Quod Narniensibus datum, ut colonorum numerus augeretur, id Cosanis petentes non impetraverunt”. (“Deputies from Narnia, complaining that they had not their due number of colonists, and that several who were not of their community were conducting themselves as colonists, Lucius Cornelius, the consul, was ordered to appoint three commissioners to adjust those matters. […] The favor granted to the Narnians, of filling up their number of colonists, was refused to the people of Cosa, who asked for it.”) (transl. Sage 1935). It is not said that the foreigners had no right to act like the colonists. The problem may be that those foreigners did not provide soldiers with the Narnienses.

43 44 45 46

47

Piper 1987, 50. Liv. 29.15. Cic., Arch. 11. Se pro colonis gerere doubtless means that they were landowners or possessores; that they were participating in cults and activities of the colony. Perhaps they even paid taxes and made wills. Liv. 34.42.5: “Novum ius eo anno a Ferentinatibus temptatum, ut Latini qui in coloniam Romanam nomina dedissent [Weissenborn Müller; E. T. Sage; dederant: Oxford, followed by Piper] cives Romani essent: Puteolos Salernumque et Buxentum adscripti coloni qui nomina dederant, et, cum ob id se pro civibus Romanis ferrent, senatus iudicavit non esse eos cives Romanos.” (“This same year, the “Ferentinates”, claim to a new privilege; that Latins who have given their names for a Roman colony, should be (immediately?) Roman citizens. As some colonists, who had given in their names for Puteoli, Salernum, and Buxentum, assumed, on that ground, the character of Roman citizens, the senate determined that they were not.”) (transl. D. Spillan and C. Edmonds 1868, modified). Cf. Piper, 1987.




be a coincidence that the decision to establish the three settlements dated back to 197 B.C. and three triumvirs had been appointed over all the colonies for this purpose – before the Ferentinates made their demand.48 The triumvirs were also responsible for the foundation of Volturnum and Liternum later in 194 B.C. on the authority of Sempronius Longus who was then consul.49 The interpretation of these related events depends on the exact legal status of Ferentinum.50 Livy says that Ferentinum, faithful to Rome in 306 B.C., was granted the right to retain its institutions aliquamdiu.51 Therefore it was probably a federated city. However, the question is: did it keep this status until the Social War? The sources are absolutely silent on this matter. It seems logical to agree with Michel Humbert that what the Ferentinates wanted in 194 B.C. would, in general measure, apply to all Latins. The only stumbling block is the question of why Ferentinum, a Hernician city, would have made this request in the name of all the Latins. Explanations proposed are: I) many Ferentinates subscribed to the petition; II) the city was perhaps in some way assimilated as a Latin city; III) or, very hypothetically, the word Ferentinates does not derive from Ferentinum but from Ferentina. It would in that case have been a collective demand made by the old Latins through their common cult place.52 We know that Latins registered their names for Roman colonies and were enrolled without objection. However, the fact that the Ferentinates asked that Roman citizenship be granted at the moment they signed up for colonization (thus a couple of years before the colonists departed to their new homes) seems to imply that being registered on the lists of future Roman colonists was not sufficient grounds for claiming Roman citizenship. This is in line with the theory of Richard Edwin Smith who pointed out that colonists were granted citizenship in their colony only when the census was taken, not during the enrollment of ‘volunteers’ (adscriptio).53 Furthermore, Claudia Moatti observes that Appian also distinguishes between enrollment (καταλέγειν) and distribution to different colonies (ἐπινέμειν).54 Indeed, there was a real risk that persons registered on the lists might not actually settle in a colony and would unrightfully claim Roman citizenship. This was a problem that recurred in Novum Comum a century and a half later.55 Moreover, as

48 49 50



Liv. 32.29.4; Ferrary 2003, 111. Liv. 34.45.1-2. Humbert 1978, 108, note 71; ibid., 213. On account of this very text, it is assumed that Ferentinum became a municipium only after 195 B.C., but this is circular reasoning. Cf. RE 6, 1909, 2208. More recently, because of this same passage in Livy, it has been assumed that the city had received Roman legal rights in 195 B.C. (DNP 4, 473, Ferentinum). Practically nothing is known about the city after 306 B.C., except that it was, as was Teanum Sidicinum, the scene of a scandal, probably in the years 130-20 B.C.: a Roman praetor is said to have flogged a local quaestor because the city baths was not to his liking. The other quaestor committed suicide (Gell., NA 10.3.3). This would have been a surprising event in a municipium, but more acceptable in a civitas foederata. In a disputed story, Florus (Flor. 1.13.7-8) relates that a certain Obsidius, commander of the Feren-

51 52

53 54

55

tine cavalry, tried to kill King Pyrrhus near Heraclea (280 B.C.). If Florus is right, Ferentinum was then a civitas foederata. Liv. 9.43.23. The hypothesis is bold, since the dissolution of the Latin League in 338 B.C. would have ended military gatherings at the outlet of Lago Albano, near the “laghetto di Turno”. However, it is impossible to exclude some form of continuity of worship – because a ritual would or could not have been terminated abruptly. Sharing sacrificial animals at Mount Alban was still a well-established ritual in 199 B.C. (Liv. 32.1.9). On the Ferentina aqua, see Grandazzi 1996. Smith 1954, 19. Moatti 1993, 11-12. From my point of view, the difference between the two stages might have been more than formal, as Moatti assumes. For Novum Comum, see Coşkun forthcoming.


Bispham observes, a similar distinction can be found in the lex agraria of 111 B.C. which records the presence of persons who were ‘enrolled in the colony’ alongside the colonists.56 In short, this law distinguished between actual colonists and ‘potential colonists’. In light of this evidence, the case of the Ferentinates should not be taken as evidence of the existence of sanctions that prevented Latins from enrolling in a Roman colony and, potentially becoming Romans. During the 194 B.C. census, besides the above mentioned Roman colonies, it was decided to create two new Latin colonies (one in Bruttium and the other in the territory of Thurium).57 By this time, the census of the Latins had certainly been archived in Rome. This is deduced from the fact that in 193 B.C. the consul Q. Minucius went to the Capitol with the magistrates of the Latin cities to enlist troops according to the proportion of iuvenes in each city.58 However, the urgency of the military situation (i.e. the rising of the Ligurians) led the Senate to declare a tumultus and enlist the Latins already under arms and allow the consul to conscript all the men he wanted from the cities and countryside.59 This measure certainly encouraged Latins to migrate to Rome in order to escape arbitrary conscription by the magistrates.60 That same year, Rome imposed Roman money lending regulations on the Latins in an effort to eradicate the questionable practice of asking an illegally high usury rate by using Latin straw men as lenders.61 Imposing this rule meant that most, if not all, Latins now had the right to conclude a contract with a Roman (was it nexus?). The decision was made shortly before the completion of the census. The logical conclusion is that the Latin cities had their own laws on interest rates (which apparently in some cases were higher) and that the Roman Senate had the power to impose an agreement by which debtors could choose under which law credit would be granted. To complicate things even further, Plutarch says that during the following censorship (189-88 B.C.) the tribune Terentius Culleo passed a plebiscite requiring the censors (Flamininus and Marcellus) to register citizens according to a new rule.62 This regulation can be interpreted as either the registration of all the sons of free men as citizens, which would be a rather surprising measure, or a simpler solution would be to assume that the censors were compelled to register all the freeborn citizens (πολίτας ἀπογραφομένους πάντας). This could be interpreted as a measure justified by gaps in the previous census. Indeed, the figure for the census for 189-88 B.C. is 258,318,63 whereas in 194 B.C. only 143,704 citizens were registered,64 an absurdly low figure.65 Furthermore, Livy mentions a decree of the Senate for the year 189 B.C. that required the

56

57 58 59 60

61 62

Lex agraria, passim: “quoi colono eive quei in colonei numero scriptus est” (Crawford 1996, 188). Bispham 2009, 84. Liv. 34.53.1-2. Liv. 34.56.5-7. Liv. 34.56.12-13. Being a Latin resident in Rome was perhaps a unique solution to escape conscription in the army. Indeed, the allies were usually rounded up by force and the urban legions were called to take up arms without any possible exemption (Liv. 34.56.11). Liv. 35.7.2-5. Plut., Vit. Flam. 18.1: “Μετὰ δὲ τὰς Ἑλληνικὰς πράξεις καὶ τὸν Ἀντιοχικὸν πόλεμον ἀπεδείχθη τιμητής, ἥτις ἐστὶν ἀρχὴ μεγίστη καὶ τρόπον τινὰ τῆς πολιτείας ἐπιτελείωσις. καὶ συνῆρχε μὲν αὐτῷ Μαρκέλλου τοῦ πεντάκις ὑπατεύσαντος υἱός, ἐξέβαλον δὲ τῆς βουλῆς τῶν οὐκ ἄγαν ἐπιφανῶν τέσσαρας, προσεδέξαντο δὲ πολίτας ἀπογραφομένους πάντας,

63 64 65

ὅσοι γονέων ἐλευθέρων ἦσαν, ἀναγκασθέντες ὑπὸ τοῦ δημάρχου Τερεντίου Κουλέωνος, ὃς ἐπηρεάζων τοῖς ἀριστοκρατικοῖς ἔπεισε τὸν δῆμον ταῦτα ψηφίσασθαι.” (“After his achievements in Greece and the war with Antiochus, Titus was appointed censor. This is the highest office at Rome, and in a manner the culmination of a political career. Titus had as colleague in this office a son of the Marcellus who had been five times consul, and the two censors ejected from the Senate four men of lesser note, and received into citizenship all who offered themselves for enrollment, provided they were born of free parents. To this step they were forced by the tribune Terentius Culeo, who wanted to spite the nobility and so persuaded the people to vote the measure”) (transl. Perrin 1921). Cf. Ferrary 2003, 120. Liv. 38.36.10. Liv. 36.9.2. For the census figures, see Lo Cascio 2008.




Campanians to register in Rome. Previously it was uncertain where they should be registered. They gained what must be described as a conubium.66 Therefore it is tempting to regard the senatus consultum or emergency decree and the plebiscite responsible for the high figure of capita civium registered in 189-88 B.C. as removing the need to suppose this was the result of many Latins being registered as Romans. However, bearing in mind that the census of 204 B.C. had already registered 214,000 capita civium, caution is advisable.67 For the first time, says Livy, soldiers were registered in the Roman census.68 As far as Livy’s figures are reliable, the conclusion is twofold: on the one hand, the census of 194 B.C. had certainly left out many people; on the other, the figures in the census of 189 B.C. had risen probably because it was more complete, not because many Latins had come to Rome. One point has irrefutably emerged: in those years Rome was heavily involved in controlling the census and various reforms were undertaken. However, nothing indicates that the Senate contemplated including Latin migrants in the citizen lists, as has sometimes been inferred from two particularly famous episodes. In 187 and 177 B.C., the Senate decided to send many Latins settled in Rome back to their cities. Both events have been extensively debated. The following discusses them briefly and refers to recent work.69 In 187 B.C., ambassadors from the Latin cities of Latium,70 waiting at the doors of the Senate, complained about the loss of many of their citizens who had been registered in Rome: Then the Senate granted an audience to embassies from the allies of the Latin name, who had gathered in large numbers from everywhere in all Latium. When these complained that a very great number of their citizens had migrated to Rome and had been registered in the census there, the task was given to the praetor Q. Terentius Culleo of searching these migrants out and of forcing to return to the place where they had been registered any whom, or the father of whom, the allies proved to have been registered in one of their cities in or after the censorship of C. Claudius and M. Livius (204 B.C.). As a result of this investigation, 12,000 Latins returned home: already then a great number of foreigners were burdening the city (Liv. 39.3).71

66

67 68



Liv. 38.36.5-6: “Campani, cum eos ex senatus consulto, quod factum erat priore anno, censores Romae censeri coegissent (nam antea incertum fuerat, ubi censerentur), petierunt ut sibi ciues Romanas ducere uxores liceret, et si qui prius duxissent, ut habere eas, et ante eam diem nati, ubi iusti sibi liberi haeredesque essent; utraque res impetrata.” (“The Campanians, since, according to the decree which had been passed the year before, the censors compelled them to be assessed at Rome – for – requested that they should be permitted to take Roman citizens as wives, that any who had already married Roman citizens should be allowed to keep them, and that children born before this day should be legitimate, and capable of inheriting from their fathers. Both requests were granted.”) (transl. Sage 1936). Liv. 29.37.6. Cf. Lo Cascio 2008, 246. I agree with Lo Cascio’s argument that the census figures include all citizens (see also Hin 2008, 189). Cicero tries to explain the non-registration of men-at-arms by saying that it was normal. Archias was not registered, because he was campaigning as a member of Lucullus’ general staff. Cicero pretends not to know

69

70

71

that the registration of soldiers had commenced in 204 B.C. Furthermore, Archias was not a “citizen at arms”, but merely a foreign professional flatterer (Cic., Arch. 11). Primarily Coşkun 2009, and Broadhead 2001; id. 2002; id. 2003; id. 2004. Undique ex Latio means either the entire area of Latium or all the Latin cities (this is the position of Kremer 2006, 33). But the text virtually imposes the first solution (socii Latinis nominis ex Latio), as noted by Weissenborn (Teubner edition) in his commentary. See Laffi 1995, 43, 45 and 49, who tends towards a global description of Latins and their allies. See the discussion in Coşkun 2009, 161-87, without any positive conclusion. Transl. Broadhead 2003, 132: “(3) Legatis deinde sociorum Latini nominis, qui toto undique ex Latio frequentes convenerant, senatus datus est. (4) His querentibus magnam multitudinem civium suorum Romam commigrasse et ibi censos esse, (5) Q. Terentio Culleoni praetori negotium datum est, ut eos conquireret, et quem C. Claudio M. Livio censoribus [204-3 B.C.] postve eos censores ipsum parentemve eius apud se censum esse probassent


The ambassadors had to wait, probably until the census was complete and all the Latin records were collected on the Capitolium. Only then could the Senate have a proper understanding of the situation by comparing these new statistics with those of the censuses of 204, 199, and 194 B.C. The choice of 204 B.C. as a reference point is explained because, from that specific date, the Latins had to conduct their census according to Roman stipulations and at the same time.72 To identify Latins qui (Romae) censi erant the magistrates had to compare the lists, with many names erased, of Latin cities (apud se censi)73 to those of Rome (which lists precisely are not known) containing many new names. The episode attests clearly of a Latin migration to Rome, perhaps resulting from the resumption of various wars beginning in 192 B.C. when the Latins anxiously tried to escape by evading conscription and the heavy duties levied on Latin cities. Moreover, they must have also been attracted to migration in order to share in the economic prosperity of Rome.74 But, if the problem of 187 was that Latins had actually been enrolled as Roman citizens en masse, as has been assumed,75 the Roman magistrates should have been in a position to withdraw their citizenship without resorting to a specific law.76 It would actually have been easier to send them back if they had registered in the Roman records as Latin residents (like Imperial incolae) without enrollment in a tribe. Curiously, Livy does not allude to any problem of citizenship, but only of demography. In 177 B.C., the Latin cities brought another complaint before the Senate. Livy begins by summarizing the demographic problem: The Senate was also moved by embassies from the allies of the Latin name, who had wearied both the censors and the previous consuls, and had finally been brought in to the Senate. The point of their complaints was that a great number of their citizens had migrated to Rome and had been registered at Rome; and that, if this trend were allowed to continue, within a few lustra, their deserted towns and deserted territories would not be able to produce a single soldier. Samnites and Paelignians were also complaining that 4,000 families from their territory had gone over to Fregellae, and that neither of them as a result of this emigration furnished any fewer soldier in the levy.

He goes on to explain how some Latins had managed to acquire Roman citizenship clandestinely by committing fraud: Moreover, two kinds of fraud had been practised to secure individual transfers of citizenship. The law granted to any persons among the allies of the Latin confederacy, who should leave in their home towns offspring of their loins, the privilege of becoming Roman citizens. By the abuse of this law some were injuring the allies, some the Roman people. For in the first place, in order to evade the requirement that they should leave offspring at home, they would give their sons to any Romans whatsoever in slavery, on the condition that they should be manumitted and thus become citizens of freedman condition; in the second place, those who had no offspring to leave behind, in order to become Roman citizens adopted children. Later, disdaining even these pretenses of obedience to law, just as they pleased, with no regard to the statute or to the requirement of offspring, they would transfer to the Roman citizenship by migration and recognition in the census. In order that these things might not occur in future,

72

73

socii, ut redire eo cogeret, ubi censi essent. (6) Hac conquisitione duodecim milia Latinorum domos redierunt, iam tum multitudine alienigenarum urbem on<e>rante.” Liv. 29.15.9-10. Cf. 29.37.7-8 [204 B.C.]. This is probably one reason why the Latin magistrates had to wait to submit their complaint. Cf. Kremer 2006, 82-85. Laffi 1995, 44.

74

75 76

McDonald 1944, 22, quoting Liv. 33.42.8 and 34.1 (lex Oppia). The latter citation is consistent with this hypothesis, which cannot be said of the first because it is not known whether recent immigrants were eligible for the public distributions. Cf. Coşkun 2009, 162. Moreover, there is no mention of any tribunician intercessio.




the ambassadors requested first, that the Senate should direct allies to return to their cities; second, that a law should be passed providing that no one should acquire a son or dispose of one for the purpose of changing his citizenship; third, that if anyone had thus become a Roman citizen, he should not be a Roman citizen. The Senate granted these petitions (Liv. 41.8.6-12). 77

It is noteworthy that the decision to grant Roman citizenship was made individually, viritim, and was far from fraud-proof. Tired of the confusion, the Latin cities had apparently asked for definitive measures. The law requiring Latin citizens who wanted to become Roman to leave a son behind in their home cities is the source of endless debate.78 Scholars agree that the law in question was a Roman law intended to protect the Italian cities from depopulation (lex dabat sociis). Broadhead says that it might have been an amended part of the colonial lex data.79 Ultimate proof that a son had been left behind could only be found in colonial archives, not in any Roman record. Today it is difficult to see how the fraud would have worked.80 If a Latin went to Rome and became Roman, how could he sell his son who had remained in his native Latin city, to a Roman? If he sold his son before he left, he could not become Roman because he had not left a legal son behind. Or should it be assumed that, once such a man had become a Roman, he still retained patria potestas over his Latin son? As Livy says, however, most people came to Rome sine lege, outside any legal framework. One important point that has often been made is that the protest came from cities that had suffered a serious drain on their manpower (Latins and Samnites); those cities which had stable or growing populations did not complain. This raises the question of whether the whole affair was really about Roman naturalization or,81 more simply, a demographic problem caused by mass migration. The Senate agreed to send the Latins back to their cities and took firm measures to restore the situation. However, it seems likely that, as Coşkun argued, the (few) Latins who had gained full Roman citizenship – meaning they had actually left one adult

77



Transl. Broadhead 2008: “Moverunt senatum et legationes socium nominis Latini, quae et censores et priores consules fatigaverant, tandem in senatum introductae. Summa querellarum erat, cives suos Romae censos plerosque Romam commigrasse; quod si permittatur, perpaucis lustris futurum, ut deserta oppida, deserti agri nullum militem dare possint. Fregellas quoque milia quattuor familiarum transisse ab se Samnites Paelignique querebantur, neque eo minus aut hos aut illos in dilectu militum dare. Genera autem fraudis duo mutandae viritim civitatis inducta erant. Lex sociis [ac] nominis Latini, qui stirpem ex sese domi relinquerent, dabat, ut cives Romani fierent. Ea lege male utendo alii sociis, alii populo Romano iniuriam faciebant. Nam et ne stirpem domi relinquerent, liberos suos quibuslibet Romanis in eam condicionem, ut manu mitterentur, mancipio dabant, libertinique cives essent; et quibus stirps deesset, quam relinquerent, ut cives Romani * * fiebant. Postea his quoque imaginibus iuris spretis, promiscue sine lege, sine stirpe in civitatem Romanam per migrationem et censum transibant. Haec ne postea fierent, petebant legati, et ut redire in civitates iuberent socios; deinde ut

78

79 80 81

lege cauerent, ne quis quem civitatis mutandae causa suum faceret neue alienaret; et si quis ita civis Romanus factus esset, <civis ne esset>. Haec impetrata ab senatu.” See Laffi 1995, 54-57; Coşkun 2009, 107-11, who provides a parallel with SIG2, I, no. 47 (founding of Naupactus by the Locrians). Discussion in Ferrary, 2003 115; Broadhead 2003, 135; Kremer 2006, 36-8. No text is conclusive. This problem was not raised in 187 B.C. It might have been a condition imposed by the Latin cities to limit the exodus, but it might also have been intended to increase the population of Roman adsidui. On the other hand, neither in 187 nor in 177 B.C. it is said that the Latins who had actually left a son in their cities were excluded from eviction from Rome. The answer might be that this rule was very old (perhaps from the time of the foedus Cassianum, or from 338 B.C.?), obsolete, or rarely used, but with new life breathed into it to cope with the Latins’ situation after the war. Broadhead 2004, 316 and 325; id. 2008, 454-55. See Broadhead 2004, 320-22; Coşkun 2009, 178-84. A point that seems important in the modern debate because this ‘law’ has been identified with the socalled ius migrandi.


son in their cities and had been registered in a Roman tribus – were not expelled. This senatorial decision is generally considered to have been the end of the ius migrandi, if it ever really existed, but such an assumption is perhaps excessively hypothetical.82 However, the measures did not stop the immigration problem and a lex Claudia was passed in 177 B.C., at the end of the census, to prevent the registration of Latins in the Roman census. This law was applied for the first time during the next census, in 173 B.C.: This year, the census was closed. The censors were Quintus Fulvius Flaccus and Aulus Postumius Albinus, the latter of whom performed the ceremony. In this survey were rated 269,015 Roman citizens. The number was considerably less, because the consul Lucius Postumius gave public orders in assembly that none of the Latin allies (who, according to the edict of the consul, Caius Claudius, ought to have gone home) should be registered at Rome, but all of them in their respective cities (Liv. 42.10.1-3).83

The result of this new law, according to Livy, was that the census of 174-73 B.C. was lower than that of 179-78 B.C. This assertion is contradicted by his own figures which actually suggest an increase of more than 10,000 citizens.84 The upshot was that only a few Latins were actually registered in the capita civium in 177 B.C. The law raises no questions about legal or illegal naturalization or an obligation to leave a son behind. Its sole concern is censorial procedure. Moreover, there is no indication that the Latins were forced to return to their cities permanently. Once registered in their place of birth, they were probably supposed to remain domiciled there, even if they chose to reside in Rome most of the year. If they chose to do so, they would have had to pay taxes and serve in the army in the contingent of their cities. It seems strange that the law was proclaimed in the second year of the census and not right at the outset, unless Livy has confused, L. Postumius Albinus (cos. 173 B.C.) and Sp. Postumius Albinus (cos. 174 B.C.). In fact, the actual censor was Aulus Postumius Albinus. Therefore, it seems safe to conjecture that the contio, during which the consul ordered the Latins to return home, was held at the opening of the census. Perhaps Latins who refused to go back home were classified as incensi.85 In short, the events discussed above suggest that the lex Claudia was not a lex de civitate, but a lex de censu. The Roman solution to the problem of emigration from and consequent weakening of the Latin allies was, in the first instance, to send Latin citizens back to their cities after the census in which they had tried to enroll (either as Roman citizens, or as Latin incolae). In the second instance, it was to force them, right from the start, to return to their home cities to register. Nothing in these texts indicates that Latin immigrants had

82

83

One probably related fact is however, that after these events Rome suspended the creation of Latin colonies in Italy. The only possible exception would have been Luca, much discussed on the basis of a difficult passage in Vell. Pat. 1.15.2. Transl. D. Spillan, C. R. Edmonds and W. A. McDevitte 1850-1856, modified: “Eo anno lustrum conditum est; censores erant Q. Fulvius <Flaccus A. Postumius> Albinus; Postumius condidit. Censa sunt civium Romanorum capita ducenta sexaginta novem milia et quindecim, minor aliquanto numerus, quia L. Postumius consul pro contione edixerat, qui socium Latini nominis ex edicto C. Claudi consulis redire in civitates suas debuissent, ne quis eorum Romae, et omnes in suis civitatibus censerentur”. Cf. Laffi 1995, 72-74.

84 85

Compare with Liv. Per. 41. Cf. Lo Cascio 2008, 244. Broadhead 2003, 133; id. 2004, 322. It seems that an incensus was no longer a citizen of any city (a good reason to prefer exilium). See: Cic., Caecin. 99; Gai., Inst. 1.160; Tit. Ulp., 11.11; Zonar, 7.19; Dion. Hal., Ant. Rom. 4.15.6; Liv. 1.44.1. Vell. Pat. (2.14) records that the Roman citizens living outside Italy had to come back to Rome at each census to register. An exception was made after 204 B.C. to cover serving soldiers. Lo Cascio (2008, 250-1) supposes that the punishments meted out to incensi had become obsolete by the 2nd c. B.C. and that proletarii had lost interest in registering at each lustrum.




actually become Romans, not even in 187 B.C., when they were expelled after the conclusion of the census. They were still registered in the records of their Latin home cities, since Roman citizenship under normal circumstances could not be held concurrently with another, implying that they formally were still citizens of their original cities.86 By returning home, even for a short time, Latins forfeited all right to become Roman. This explains, for example, why the Menander mentioned by Cicero needed a special law to secure that he could remain Roman after taking part in a delegation to his native city, where he probably was still regarded as a local citizen.87 The scandal of 177 B.C. was not only about Latins, but also about the Samnite and Paelignian migration. These people were complaining that four thousand of their families had emigrated to Fregellae. Their problems were the same as those of the Latins: the Samnites could no longer muster the contingent of soldiers demanded by Rome.88 What is unclear is whether these Paeligni and Samnites had been registered in the local census and whether they had had the opportunity to become Latin citizens of Fregellae which would have opened the door to Roman citizenship by a two-step procedure.89 The parallel with the Latins in Rome is significant and arouses the suspicion that they were perhaps registered as incolae, in the same way the Samnites were incolae of Aesernia or the Salassi incolae of Aosta. In any case, Fregellae did not ask for anything and on this occasion the Senate does not seem to have interfered in its affairs. The inevitable conclusion is that the cases cited above give no evidence of massive integration of Latins into the Roman citizen body. To quote Jean-Louis Ferrary, the naturalization of individuals or groups required a law.90 If such a law was needed to reward meritorious (Latin) soldiers, it is inconceivable that a Latin could have received precious Roman citizenship just by taking up residence in Rome. Even registration in the census,91 as Cicero records, did not guarantee citizenship.92

86

87

88



Cic., Balb. 11.28: “Duarum civitatum civis noster esse iure civili nemo potest: non esse huius civitatis qui se alii civitati dicarit potest”. Cic., Balb. 11.28: “sed etiam postliminio potest civitatis fieri mutatio. Neque enim sine causa de Cn. Publicio Menandro, libertino homine, quem apud maiores legati nostri in Graeciam proficiscentes interpretem secum habere voluerunt, ad populum latum <est> ut is Publicius, si domum revenisset et inde Romam redisset, ne minus civis esset”. Pomponius (Paulus, Dig. 49.15.5.3) develops the principle suggested by Cicero in Pro Balbo and considers that only individual choice could determine citizenship, which was certainly wrong, at least at the time of Menander. Otherwise the People would not have had to vote. Coşkun (2009, 89-91) considers this episode unclear. On the contrary, I think that the example of Menander helps explain the measures taken in Rome. By returning to his former homeland, he fell under postliminium. Cf. about Balbus, Cic., Balb. 12.29: “Quod si civi Romano licet esse Gaditanum sive exsilio sive postliminio sive reiectione huius civitatis […]?” The sharing of the Samnite ager publicus, distributed in 201 B.C. to the veterans of Africa (Liv. 31.4.1-2), and the trientabulum, an action that consumed a large

89

90

91

92

part of it, probably also reduced allies’ and Latins’ ability to use part of the ager publicus alongside their own land. Roselaar (2010, 128) assumes that the trientabula were given as ager publicus, but with “virtually complete security of tenure”. Cf. Cic., Arch. 4. In the next chapter, Cicero explains that peregrines might have illicitly obtained civitas by registering in municipia. This would have allowed them to bypass the lex Papia. Perhaps this situation was no longer the same as in the 2nd c. B.C., because the municipia were Roman: Roman citizenship could be obtained outside Rome. Ferrary 2003, 128. He refers particularly to a law for the Campanians in Sicily and the Numidian Muttines. The absence of any such mention in the case of Praeneste can be explained by the fact that the cavalry declined the Senate’s offer; a vote on a law was therefore unnecessary. By and large, Laffi 1995 believes the Latins had no legal claim to be registered and that their removal was therefore a matter of course. However, he assumes that registration in the Roman census would have involved a transfer of citizenship. Cic., Arch. 5.11: “Census non ius civitatis confirmat ac tantum modo indicat eum, qui sit census, ita se iam tum gessisse pro cive.” Cf. Laffi 1995, 66.


The timing of these events, punctuated by census years, is interesting. As noted by Humbert, the census of 188-87 B.C. did away with the last sine suffragio municipalities by granting full citizenship to Formia, Fundi and Arpinum.93 At the end of the census the Senate sent twelve thousand Latins home. Finally, with the outlawing of the Bacchanalia, which also extended to Latin cities in 186 B.C., the same Senate instigated a witch-hunt throughout Italy, thereby asserting its authority over all Italy, even beyond the Roman territory.94 It seems that relations between Rome and the Latins were evolving fast.95 While many Latins migrated to Rome and potentially tried to acquire Roman citizenship, Rome founded several new Latin colonies and reinforced others, thereby creating thousands of new Latins.96 It would be wise to accept Broadhead’s main conclusions, namely that there was very high internal migration in Italy after the Second Punic War.97 Inevitably, this migration would have created only a few new Roman citizens, because most migrants remained Latins or Italian allies. What real proof is there that some or indeed many Latins became Roman by migration? As stated by Peter Brunt, there are only two secure cases in which foreign communities prosecuted their citizens for acquiring Roman citizenship without their consent. The first is that of Cassius or Crassus, whom the Mamertini unsuccessfully tried to reclaim as their own under the lex Papia.98 He may have been made a citizen viritim by an imperator; those seeking to prosecute him withdrew their case. The second case is that of Perperna, consul in 130 B.C., whose father allegedly was later sentenced under the lex Papia for obtaining Roman citizenship by fraudulent means. Brunt notes that Perperna’s father had died many years before the lex Papia was passed; he therefore concludes that “the testimony of Valerius Maximus (3.4.5) is a tissue of falsehoods”.99 The lex Papia, which passed a century after the lex Claudia, 100 was the last in a series of laws intended to control the acquirement of citizenship. There are references (most of them in Cicero) to a lex Licinia Mucia passed in 95 B.C., just after the censorship of L. Valerius Flaccus and M. Antonius, and to a lex Papia of 65 B.C., passed during the troubled censorship of M. Licinius Crassus and Q. Lutatius Catulus. Asconius says that the lex Licinia Mucia was a law de civibus redigendis like the lex Claudia.101 However, the real purpose of the law was to prevent allies from acting as Romans by sending them back in their cities.102

93 94

95

96

97 98 99

Humbert 1978. But see Mouritsen 2007. On the question of the authority of the Senate in Italy in the event of an emergency, I refer to Bispham 2009, 115-23 whose conclusions I adopt. See also the decree ordering Eumenes to leave Italy (Polyb. 30.19.6-8). Competition among the patricians drove the Roman Senate to close access to citizenship, just at the time the wars fueled by this competition required an increase in the number of allies, who would have cost more. See Bispham 2009, 136. Smith 1954, 20 connects the complaint of the Latin cities and the eventual decision to create a Latin colony at Aquileia, whereas the Senate had initially planned to make it a Roman colony (Liv. 39.55.5). The complaints will have persuaded the Senate not to accelerate the naturalization of the Latins, which would have weakened the colonies, which still played a strategically significant role. The early 2nd c. B.C. was undeniably a period of intense political activity and heated debate over citizenship and the relationship between Rome and its allies. Broadhead 2002, passim and 121. Cic., Balb. 23.52. Brunt 1982, 145. Brunt notes that Balbus was prose-

100 101 102

cuted not by the city of Gades but by an individual. There might be some truth in Valerius’ statement, but I fear I must plead ignorance of what the problem with Perperna’s citizenship was. Coşkun 2009, 150-55. Asc., Pis. 67-68 C. Cicero considers that, if it is normal to prevent the usurpation of citizenship, it is uncivilized to forbid foreigners to reside in Rome. Cf. Cic., Off. 3.47: “Male etiam, qui peregrinos urbibus uti prohibent eosque exterminant, ut Pennus apud patres nostros, Papius nuper. Nam esse pro cive, qui civis non sit, rectum est non licere, quam legem tulerunt sapientissimi consules Crassus et Scaevola. Usu vero urbis prohibere peregrinos, sane inhumanum est.” Asc., Pis. 68 C: “Nam cum summa cupiditate civitatis Romanae Italici populi tenerentur et ob id magna pars eorum pro civibus Romanis se gereret, necessaria lex visa est ut in suae quisque civitatis ius redigeretur. The reference to Pennus is strange, because the lex Iunia as we know it was a lex repetundatum” (Crawford 1996, vol. 1, 73). There should have been another lex issued by the same magistrate banning peregrines.




Both laws did not exclude the possibility of granting citizenship to foreigners.103 The procedure adopted under the first can be understood from literary evidence provided by Cicero, interestingly, recalling the procedure of 187 B.C. According to Cicero, Timaeus writes that Lysias, who was born in Athens and had acted like an Athenian all his life, was actually a Syracusan, as if the lex Licinia Mucia was applicable to him.104 Lysias was from a Sicilian family, so Cicero’s passage seems to imply that the lex Licinia Mucia would have required scouring old city archives to identify foreign citizens and subsequently forbid them to act like Romans.105 The worst fate that could befall them was expulsion from Rome during the census, which Cicero claims is the moment at which one should act like a Roman.106 Therefore, it seems that the lex Licinia Mucia was probably the first act of legislation to link migration and the fraudulent acquisition of citizenship. The lex Papia is less well known. Dio Cassius claims that it expelled only non-Italian foreigners from Rome.107 However, Cicero suggests that the law complemented another law on citizenship (probably here the lex Plautia Papiria) and set limits on the registration of peregrines in municipia.108 The supposition is that registration of foreigners may have offered a fraudulent way to obtain Roman citizenship. By excluding all the non-Italians from Rome, probably even from Italy, the censors would register only citizens already on the official lists and list only Italian allies not already included as citizens during the previous census of 70 B.C. This is probably why the Mamertines initially thought that they could make a bid to reclaim one of their former citizens but withdrew their complaint when they understood the real purpose of the law.109 In a nutshell, the evidence discussed above, suggests that migration emerged as a real citizenship problem only in the 1st c. B.C. or, perhaps in the second half of the 2nd c. B.C at the earliest. Previously, the laws that required foreigners to be expelled from Rome ensured that the census would be conducted in good conditions and probably stabilized the demography of the Latin cities. Should disputes or instances of fraud arise, the records of allied cities could be examined and fraudulent citizens sent home.

Individual naturalization Therefore, the best solution for a person eager to acquire Roman citizenship was individual naturalization, granted by the decision of a magistrate or the People. This type of naturalization probably often included privileges in terms of property status and taxation. There are numerous examples of this procedure110 and

103 104

105

106 107

Cic., Balb. 48 and 54. Cic., Brut. 16.63: “Catonis autem orationes non minus multae fere sunt quam Attici Lysiae, cuius arbitror plurumas esse, est enim Atticus, quoniam certe Athenis est et natus et mortuus et functus omni civium munere, quamquam Timaeus eum quasi Licinia et Mucia lege repetit Syracusas.” In 187 B.C. the investigation had risen to 204. It might be assumed that the lex Licinia Mucia allowed them to go back one generation at least. Cic., Arch. 5 (supra, note 92). Cass. Dio 37.9.5: “Κἀν τούτῳ πάντες οἱ ἐν τῇ Ῥώμῃ διατρίβοντες, πλὴν τῶν τὴν νῦν Ἰταλίαν οἰκούντων, ἐξέπεσον Γαΐου

τινὸς Παπίου δημάρχου γνώμῃ, ἐπειδὴ ἐπεπόλαζον καὶ οὐκ ἐδόκουν ἐπιτήδειοί σφισιν εἶναι συνοικεῖν.” (“Meanwhile all those who were resident aliens in Rome, except inhabitants of what is now Italy, were banished on the motion of one Gaius Papius, a tribune, because they were coming to be too numerous and were not thought fit per-



108

109

110

sons to dwell with the citizens”) (transl. Cary 1914). Cic., Arch. 10: “Quid? cum ceteri non modo post civitatem datam, sed etiam post legem Papiam aliquo modo in eorum municipiorum tabulas inrepserunt, hic, qui ne utitur quidem illis in quibus est scriptus, quod semper se Heracliensem esse voluit, reicietur?” (“What? When others were smuggling themselves onto the records of those towns (Rhegium, Locri, Naples, Tarentum), not only after the grant of citizenship but even after the Papian law was passed, is this man, who did not even make use of his citizenship of these towns, because he always wanted to be a Heraclean, really to be driven out?”) (transl. Polkovnik 2005, online). Cic., Balb. 52: “Iudices cum prae se ferrent palamque loquerentur quid essent lege Papia de M. Cassio Mamertinis repetentibus iudicaturi. Mamertini publice suscepta causa destiterunt.” Some are given by Cicero to strengthen the cases of Balbus and Archias. Cf. Sanchez 2007.


three different procedures can be identified. The oldest recorded is the naturalization of an individual or small group by the People as a reward for military service. This tradition began very early with L. Mamilius in 458 B.C.111 and with the Campanian cavalry in 340 B.C.112 In 216 B.C., the cavalry of Praeneste rejected the ‘privilege’ offered them for their courage during the siege of Casilinum.113 In 215 B.C., after the betrayal of Capua, the People passed a law or plebiscite granting Roman citizenship to three hundred loyal Campanian cavalrymen who no longer knew to what country they should be linked. The bestowal of citizenship was therefore dated retroactively to the day before the defection of Capua, and they were attached to the municipium of Cumae.114 The procedure apparently included a senatus consultum asking the People, who were competent in matters of citizenship, to validate the proposed naturalization.115 One of the latest examples is that of Minatus Magius from Aeclanum, mentioned by Velleius.116 A very special case was the naturalization of the priestess of Ceres. At the request of the praetor urbanus, the People passed a law for each new priestess nominatim.117 Another way to obtain citizenship was to take advantage of the right of a colonial foundation to incorporate peregrines in its civitas.118 It is a known fact, for instance, that Ennius was naturalized by being registered in a colony.119 Probably a peregrine was initially enrolled in the colony as a civis in the colonial tribe and then registered with all Roman citizens during the first census after his enrollment. The request made by the Ferentinates (discussed above) shows that it was possible,120 indeed probably common, to allow some Latins to register for Roman colonies. Another, more individual colonial procedure is known from the time of Marius. The accuser of C. Matrinius of Spoleto argued that Marius had not founded the colonies originally planned and that therefore the right granted to him by the lex Apuleia to make three peregrines per colony citizens had lapsed.121 This argument seems legitimate, but in spite of this, Matrinius had officially been made a Roman citizen. It can be assumed that Marius had received the privilege on the pretext of founding colonies but without the requirement to the actual founding of a colony. He did exactly what the Ferentinates had requested a century earlier, granting citizenship at the moment the colonists entered their names and not during the first census of the Roman colony. Later, Roman citizens were created individually by magistrates – no longer by the People – for outstanding merit in warfare. The examples cited throughout the Pro Balbo date no earlier than Marius but many confirm that the practice was well established, probably as early as the late 2nd c. B.C.122 It can be supposed that the

111 112 113

114

115 116 117 118

Liv. 3.29.6. Liv. 8.11.16. Liv. 23.20.2: Livy adds that he ignores the fate of the Perusian soldiers, who had apparently not been the subject of a senatus consultum. Liv. 23.31.10-11: “Et de trecentis equitibus Campanis qui in Sicilia cum fide stipendiis emeritis Romam venerant latum ad populum ut cives Romani essent; item uti municipes Cumani essent pridie quam populus Campanus a populo Romano defecisset. Maxime ut hoc ferretur moverat quod quorum hominum essent scire se ipsi negabant vetere patria relicta, in eam in quam redierant nondum adsciti.” Cf. Ferrary 2003, 109. The survivors of the Casilinum garrison were also sent to Cumae. Sanchez 2007, 236. Vell. Pat. 2.16. Cic., Balb. 55. Smith 1954, 19-20. This is one of the options considered by Gagliardi 2011, 65; id. 2014, 72-75, in the

119

120 121

122

event that natives were still established in a colonial territory. Ennius was made citizen by Q. Fulvius Nobilior (Cic., Brut. 79), perhaps during the deduction of either Pisaurum or Potentia in 184 B.C. Cf. Cic., Orat. 3.42.168. Salmon 1936, 50; Piper 1987, 40. Supra, p. 165. Cic., Balb. 21.48: “Sed cum lege Apuleia coloniae non essent deductae, qua lege Saturninus C. Mario tulerat ut in singulas colonias ternos civis Romanos facere posset, negabat hoc beneficium re ipsa sublata valere debere.” This might explain why provincials took the name of governors, such as the Domitii in Gaul. Plutarch (Vit. Mar. 28.3) records that some people regarded the granting of citizenship to a thousand soldiers from Camerinum as illegal. Marius probably had no recourse to a specific law. He might have hoped for a global ratification of his actions at the end of his consulship.




specific laws that permitted magistrates to grant citizenship were limited to rewarding bravery on the battlefield (perhaps by virtue of imperium militare). Certainly, it is recorded that Theophanes of Mytilene was granted citizenship in the camp, in front of the army.123 Some magistrates apparently did not have the benefit of such a clause in their leges de imperio. Cicero, for example, says that Catulus could not have granted Archias citizenship personally, but could have asked Metellus for this favor.124 Archias was a member of Lucullus’s army and had never served with Metellus. It can be assumed that Lucullus did not have a law authorizing him to create new citizens and that is why Archias took advantage of the lex Plautia Papiria. The Asculum decree remains the best evidence for this type of naturalization.125 Pompeius Strabo gave Roman citizenship to a group of Spanish cavalry in the turma Salvitana, citing a lex Iulia.126 The document lists the soldiers’ names prior to their naturalization. They are identified by their home cities, which would have been important for the archives of the cities that lost elite citizens. Their names are still peregrine in form (for example, Beles Umarbeles f.) with the exception of the citizens of Lerida (for example, Cn. Cornelius Nesille f.) who were most likely Latins.127 Nevertheless, the cavalrymen still had to be included in the Roman voting tribes and registered in Rome. Obviously, an imperator had no right to assign anybody a new tribe: that was a censorial privilege and one that could be exercised only during the census. The difference with the Campanian equites was that an imperator could act on a request for citizenship before a public vote was taken and it was no longer necessary to assign them to a municipium. The letter of Octavian to Seleucos of Rhosos half a century later attests the procedure explicitly and reminds us that naturalization implied some tax privileges in one’s home city that also had an impact on property rights.128 First (§ 3), by virtue of a lex Munatia Aemilia, Octavian grants Seleucos Roman citizenship with exemptions

123

124 125

126



Cic., Arch. 24: “Quid? Noster hic Magnus, qui cum virtute fortunam adaequavit, nonne Theophanem Mytilenaeum, scriptorem rerum suarum, in contione militum civitate donavit; […].” One suspects that Pompey had made a travesty of this military reward during a contio in order to formally enforce his own lex de imperio, because Theophanes was not a soldier. Cic., Arch. 26. CIL I2 709; VI 27045; ILS 8888; ILLRP 515: “[C]n Pompeius Sex. [f. imperator] virtutis caussa / equites Hispanos ceives [Romanos fecit in castr]eis apud Asculum a.d. XIV k Dec / ex lege Iulia, etc.” The lex Iulia is generally identified with the lex Iulia de civitate danda (already Stevenson 1919, 96, who supposes that Pompeius Strabo gave civitas to P. Caesius of Ravenna by virtue of the same lex Iulia). Cicero says this lex Iulia granted citizenship to entire communities if they agreed to become Romans (Cic., Balb. 21; see Brunt 1982, 144). See also App., B Civ. 1.49; Cic., Balb. 8.21; Vell. Pat. 2.16 and 20. Therefore the beneficiaries were probably the socii nominisque Latinis of the Tarentine fragment (Crawford 1996, vol. 1, 212; id. l. 12) and of the lex de provinciis praetoriis (Cnidos copy, col. II, 6-8: “οἵ τε πολλῖται Ῥωμαίων οἵ τε σύμμαχοι ὀνόματος Λατίνου”). Therefore the Asculum Decree is the only text to include a grant of civitas viritim to

127

128

Spaniards under the conditions laid down by the lex Iulia de civitate. After the example of the lex Gellia Cornelia in favor of Pompey (Cic., Balb. 8.19), it might have been assumed that the lex Iulia of Asculum was a special law giving Pompeius Strabo the right to make brave allies citizens, maybe his lex curiata de imperio. As assumed by A. Degrassi (ad loc.) following E. Païs. Cf. Sanchez 2007, 236-37. Stevenson’s 1919, 100 argument is inconsistent: if the Ilerdenses were already Romans, they had no need of the decree. Moreover, their names do not contain any tribe, whereas this is always present in the names of members of the consilium. It seems the document reflects one stage in the proceedings. The fact that some of them had names also attested in the consilium is not proof of any “pre-naturalization”, despite the possible parallel with Cic., Verr. 2.4.37: like Archias, Q. Lutatius Diodorus was under the protection of Q. Lutatius Catulus, who obtained citizenship for Diodorus from Sulla as a favor. We might conjecture that, in the normal procedure during military awards ceremonies, Roman officers would have proposed names and the soldiers, who had to choose Roman nomina, would have preferred these suggestions. In favour of Ilerda as a Latin city, see also Galsterer 1995, 86. Raggi 2004.


from tax and military service. In a second step (§ 4), he specifies that Seleucos would be a member of the tribus Cornelia, in which he would vote. The new conferral of citizenship notably includes a modification in the status of his property: “We grant citizenship and tax exemption on [all?] proper/[ty], in the same way [as those who] are [Roman] citizens with the best conditions and the best status”.129 From a land surveyor’s point of view, the question was to determine whether any piece of land bought by Seleucos automatically became tax-free. During the reign of Augustus, under the conditions laid down in the third Cyrene edict, there was a legal difference between the grants of citizenship and grants of citizenship with immunitas.130 On these grounds it can be concluded that, from the time of the letter to Seleucos and subsequently, this measure required either a law or a senatus consultum. More controversial and rather more obscure, the last method by which an individual could acquire citizenship, is what modern scholars call the ius civitatis adipiscendae per magistratum. There is little to be gained from lingering on this matter. The date at which this privilege was granted to the Latins is not known131 but it is possible to draw a parallel with the lex repetundarum, which grants a successful prosecutor Roman citizenship and exemption from military service.132 The law also conferred the right of provocatio and vacatio militiae on those who declined citizenship (cf. also the contribution of De Ligt in this volume). Whatever the details were, these rights are never mentioned in the years 204-167 B.C. At that time, provocatio had apparently become indispensable to the colonial magistrates, since Roman magistrates had grown accustomed to treating the allies as subjects. Livy traces this bad habit back to L. Postumius Albinus in 173 B.C.133 The speech in which

129

130

131

Raggi 2004, 129: “πολειτείαν καὶ ἀνεισφορίαν τῶν ὑπαρχόν/[των πάντων δί]δομεν, οὕτω[ς ὡς οἵτινες τῶ]ι ἀρίστωι νόμωι ἀρίστωι τε δικαίωι πολεῖται / [Ῥωμαῖοι ἀνείσ]φο[ρ]οί εἰσιν.” (§3, l. 20-2). FIRA I, no. 68, III (Latin translation): “Si quidam ex Cyrenaica provincia civitate honorati sint, hos nihilominus muneribus pro rata fungi in corpore Graecorum iubeo, exceptis iis quibus ex lege senatusve consulto patris mei meove decreto tributorum immunitas una cum civitate data fuerit: hosque ipsos, quibus tributorum immunitas data fuerit, harum rerum esse immunes, quas tunc habuerint, placet mihi, de postea acquisitis omnibus munera subire.” (“If any persons from the province of Cyrene have been honored with Roman citizenship I command that they nonetheless shall discharge their compulsory public services among the body of the Greeks in their proper turn, except those persons to whom by a law or by a decree of the Senate, by my father’s or my own decree, the citizenship was granted with exemption from taxation. It is my pleasure that these same persons, to whom exemption from taxation has been granted, shall be immune in respect to the property in their possession at that time, but that they shall pay taxes on all property that they later acquired”) (transl. Johnson, Coleman-Norton and Bourne 1961). Cf. Sanchez 2007, 241-43. Coşkun 2009, 134-47 notes the importance of the years 125-22 B.C., after the destruction of Fregellae, in the

132

133

process of the integration of the Latin elites. Perhaps some imperatores received the right to grant the civitas to some peregrines in the same years. In Gaul, some Domitii, a name which must be connected with Domitius Ahenobarbus (cos. 122 B.C.), are known, perhaps linked to the foundation of Narbonne. There were also Fabii (Fabius Maximus Allobrogicus, cos. 121 B.C.). On this topic: Bispham 2009, 127-31; Coşkun 2009, 134-46. Crawford 1996, vol. 1, 65-112; Lex repetundarum: “(76) vvvv de ceivitate danda. vvvvvv sei quis eor[u]m, quei ceivis Romanus non erit, ex hace lege alteri nomen [—- ad praetor]em, quoius ex hace lege quaestio erit, detolerit, et is eo {eo} iudicio hace lege condemnatus erit, tu[m eis, quei eius nomen detolerint, quoius eorum opera maxume is quoius nomen delatum erit condemnatus erit, ipse ceivis Romanus iustus esto filieique eiei gnatei, quom] / (77) ceivis Romanus ex hace lege fiet, nepotesque [d]um eiei filio gnateis ceiveis Romanei iustei sunto[inque eius tribum, quei ex h(ace) l(ege) condemnatus erit, sufragiu]m ferunto, inque ea<m> tribum censento, militiaeque eis vocatio esto, aera stipendiaque o[mnia eis merita sunto. nei qui magistratus prove magistratu —- eius h(ace) l(ege)] / (78) nihilum rogato. vvvvvvv de provocation[e vocation]eque danda.” Cf. Cic., Balb. 54; Strab. 4.1.12; Asc., Pis. 3C; App., B Civ. 2.26; Gai., Inst. 1.95-96. Liv. 42.1.7-12.




Gaius Gracchus criticizes such behavior among his contemporaries is well known.134 However, as Coşkun points out,135 it is impossible to determine whether a Latin magistrate instantly became Roman136 or retained the possibility to decline this ‘privilege’. Furthermore, it is not known whether he had to stay in Rome to enjoy his citizenship, whether he forfeited his previous citizenship, whether his entire family had the same rights, or whether he had the same privileges as those later granted to Seleucos of Rhosos. It was also possible to lose Roman citizenship.137 One of the more obvious ways was through banishment. Cicero says that an exile (trying to evade conviction) lost his citizenship by taking up residence outside Roman territory.138 No source indicates, for instance, that Scipio Africanus, who was exiled to Liternum, lost his Roman citizenship.139 In an attempt to understand why Latin colonists leaving Rome lost their citizenship, Cicero assumed that it was either a deliberate renunciation or a disguised exile. In fact, he claims that had the colonists agreed to serve their sentences in Rome they would have kept their citizenship.140 This claim is significant because the colonists he discusses were not criminals but Cicero wanted to prove that everyone who lost his citizenship must have renounced it willingly. What is more important is the fact that, just as in the example of Menander or Balbus, Cicero connects change of citizenship to residence (perhaps domicile) in another community.141 This assumes that Roman law allowed not only the possibility of integrating foreigners into the civitas but also the possibility of making such citizens peregrine by registering them in a new city and, obviously, erasing their names from the Roman records. A passage in the lex Ursonensis strengthens the above discussed ‘renunciation of civitas’ as well as the close link between personal status and property. It is a known fact that decurions were required to own a home of six hundred roof tiles inside the city. A new fragment adds that colonists who were not decurions should have a house of at least three hundred roof tiles, probably in colonia intra qua aratro circumductum est (cf. Sisani in this volume).142 New settlers would have two years after the founding of the colony to build

134 135 136

137

138

139



Gell., NA 10.3.3. Coşkun 2009, 137-42. Galsterer 1995, 83: he notes that the lex Acilia repetundarum and the Tarentum fragment do not indicate that magistrates automatically received civitas. However, Cicero (Caecin. 101) declares that it was impossible to take away civitas from a citizen who did not want to relinquish it and that Sulla’s law against Arretium violated established legislation. Cic., Dom. 29.77: “Sed cum hoc iuris a maioribus proditum sit, ut nemo civis Romanus aut libertatem aut civitatem possit amittere, nisi ipse auctor factus sit.” Cf. Galsterer 1995, 80. Cic., Caecin. 100: “Nam, cum ex nostro iure duarum civitatum nemo esse possit, tum amittitur haec civitas denique, cum is qui profugit receptus est in exsilium, hoc est in aliam civitatem.” (“For as, according to our law, no one can be a citizen of two cities, when the one who has fled is hosted as an exile, that mean received in another citizenship, he lose our citizenship”) (transl. M. Tarpin: the double meaning of the word civitas makes it difficult to follow exactly what Cicero means). There is absolutely no evidence on this subject. It is

140

141 142

not even known whether Scipio was buried in Roman territory. Cicero’s exile was very short and occurred during a period when there was no census. Cic., Caecin. 98: “Quaeri hoc solere me non praeterit – ut ex me ea quae tibi in mentem non veniunt audias – quem ad modum, si civitas adimi non possit, in colonias Latinas saepe nostri cives profecti sint. Aut sua voluntate aut legis multa profecti sunt; quam multam si sufferre voluissent, manere in civitate potuissent.” Kremer 2006, 65-66 assumes that the multa is the debt owed by proletarians destined to leave for the colonies who could have fallen under nexus. It sounds anachronistic. What does nexus mean in this case? For an example of threatened penalty for colonists in archaic Rome, see Plut., Coriol. 13.3. Cic., Balb. 11.28; Caecin. 100. AE 2006 645 = 1991 1020 a-b = 2004 744: §14: “Quicumque in col(onia) G(enetiva) I(ulia) decurio erit, is decurio in ea colon(ia), / intra qua aratro circumductum est, aedificium, quod / non sit minus tegular(um) DC, qui colonus neque decurio erit, / is aedificium, quod non sit minus tegularum CCC, habeto / in biennio proxumo, quo ea colon(ia) deducta erit.”


a house; if they failed to do so they would not be counted as colonists.143 Legally, a person who did not build a house was in the position of an incensus, a citizen who had not been registered in the census and was therefore liable to exclusion from the city or even sale into slavery.144 This is why a census had to be held two years after the founding of the colony. This rule could also explain a peculiarity regarding the second foundation of Aquileia in 169 B.C (cf. Sewell in this volume). An inscription found in 1995 indicates that the triumvir T. Annius made three selections of senators.145 The text is undated, which might be taken to mean that he held three different censuses, five years apart, returning to Rome after each mission. However, it is also possible that the triumvir had already enforced the above discussed principle known from the lex Ursonensis. In the year of the foundation, he could have designated an initial group of decurions (perhaps the equites, who received larger lots and who gained immediate access to the required census) among those who “gave their names”. In the following year he could have registered decurions and citizens who had already built their houses. Finally he could have repeated the process a third time. In this scenario, the whole procedure would have actually taken place in only two years.146 Subsequently, the city censors themselves could confirm or deny the integration of new citizens. The two years during which one of the triumvirs was present would have been a reasonable amount of time to establish the city. It took at least one year to ensure the self-sufficiency of the colony in terms of food.147 The exact procedure is never reported but a period of three years for the foundation seems reasonable and accords with what Livy suggests about the term triumvirs served.148 Note that, beginning in 169 B. C., T. Annius would have achieved his mission in 167 B.C. at the moment when the colonial censors had to deliver their registers to the Roman censors. The period of two years can furthermore be explained by considering property laws. Property was normally granted to colonists in ownership and therefore entailed a mancipatio, or something similar, which was adapted to the Latin colonies. Consequently, at the end of two years, during which a third party in theory could contest the property claims, a colonist would have been full owner of his plot. This rule is

143

144

145

Broadhead 2008, 455-56. Confusion with § 91, which obliges decurions and priests to be domiciled in the oppidum or within a limit of one thousand paces for five years after their election should be avoided. These were long-term functions and applicants had to make this commitment to maintain their position during their tenure. Cic., Caecin. 99: “Cum autem incensum vendit, hoc iudicat, cum ei qui in servitute iusta fuerunt censu liberentur, eum qui, cum liber esset, censeri nolverit, ipsum sibi libertatem abiudicavisse.” The punishment for refusal to submit a declaration to the census was a general rule, which also appears in the Oscan law of Bantia, l. 20-23 (Crawford 1996, vol. 1, 277). The Latin translation is as follows: “sed si quis in censum non venerit dolo malo ast eius vincitur ipse in comitio caedatur pro magistratu populo praesente sine dolo malo et veneat omnis familia et pecunia omnis quae eius fuerit quae incensa fuerit publica esto.” See references in Broadhead 2003, 133. AE 1986, 685: “T. Annius T. f. tri(um)vir / is hance aedem / faciundam dedit / dedicavitque legesq(ue) / composivit deditque / senatum ter coptavit.” Liv. 43.17.1. See Gordon and Reynolds 2003, 221; Bispham 2009, 155-56.

146

147

148

Excluding preparatory works carried out in Rome or in Northern Italy The very lacunose elogium of Brundisium has given rise to much discussion, because the inscription mentions facts that do not correspond to the foundation of Brundisium. (Barbula cos: 317, 311, 281, 230 B.C.). The text does not give us the name of the magistrate who primus senatum legit (in Brundisium?). However, the date (230 B.C.) seems too late for the first comitia of a colony founded in 247 or 244 B.C., Bispham 2009, 153-54 and suggests that it might have been some kind of informal consilium between the foundation and 230 B.C., which was a censorial year. Muccigrosso 2003 suggests that the first censor should be Appius Claudius (cens. 311 B.C.), instead of Cunctator (cos. 230 B.C.). E.g., the lex Aelia on the founding of two Latin colonies: Liv. 34.53.1-2. Ferrary 2003, 111. The foundation took place in 193 B.C. at Castrum Frentinum (Liv. 35.9.7-8) and Livy mentions it in connection with the conclusion of the census. It is therefore likely that the triumvirs then deposited the first records of the new city in Rome, according to the rule laid down in 204 B.C.




attested in the Twelve Tables.149 After two years, ownership of the property was unquestionable and the colonist could at that time be fully registered. Upon closer examination it seems that the text of the law of Tarentum is also consistent with that for Urso. The former demands only a fine, for the benefit of the city, from a decurio who still did not have a house (of one thousand five hundred roof tiles) or had acquired one by fraud.150 This difference can be explained by the fact that Urso was a colony that had just been founded, whereas Tarentum was a municipium in which the notables already owned large houses. Moreover, the colonists of Urso were in the process of changing their citizenship, whereas the Tarentines moved en masse from one status to another: the citizen lists were already established. It was absolutely out of the question to deny any Tarentines citizenship in the new municipium of Tarentum.151 It should be understood that colonists received their new citizenship only at the moment they recorded their houses and other property in the census records. This recording procedure also helps to explain the earlier discussed case of the Ferentinates. If the hypothesis is indeed right that they hoped to become Roman unconditionally at the moment of signing up for colonization, we can use the Urso case to understand why the Senate could not accede to their demand. During the time between the registration of the ‘volunteers’ and the two-year local census, the future colonists probably lived under the terms of the lex agraria: “quei in colonei numero scriptus est”.152 The text of the law of Urso does not say what would happen to those who did not register. Perhaps they would have had the option of resuming their previous status, although this may be unlikely.153 Since the text does not say that they would forfeit their lot, it may be inferred that they probably remained in the territory of the colony, perhaps as incolae.154 Two famous examples suggest that a person could be an incola of a city without being a citizen of another city. The first example is that of the Samnites who were incolae of Aesernia (cf. also Hermon in this volume);155 the second is that of the Salassi of Augusta Praetoria, who were incolae qui se primi in colon(iam) contulerunt.156 These are references to natives who did not act according to the procedure which enabled people to become a citizen when the colony was created.157 Another category (perhaps also incolae) might have been composed of non-citizen

149

150



Leg. XII Tab., 6, 3 (Crawford 1996 vol. 2, 658-59): “auctoritas fundi biennium <esto. ?ceterarum rerum? annus esto.>”. For an estate auctoritas <is to be> two years. <? For other things? it is to be one year.> (Crawford,1996). See Cic., Top. 4.23. Cic., Caecin. 19.54; Gai., Inst. 2.42; Gai., Inst. 2.52-53. Crawford 1996, vol. 1, 304, Col. I, l. 26-28: “[…] quei decurio municipi Tarentinei est erit queiue in municipio Tarenti[no in] / senatu sententiam deixerit, is in o[pp]ido Tarentei aut intra eius muni[cipi] / fineis aedificium quod non minu[s] (mille quingentis) tegularum tectum sit habeto [sine] / d(olo) m(alo). Quei eorum ita aedificium suom non habebit seiue quis eorum / aedificium emerit mancupioue acceperit quo hoic legi fraudem f[aciat], / is in annos singulos (sestertium) n(ummum) (quinque milia) municipio Tarentino dare damnas esto.” “Whoever is a member of the municipal Senate of Tarentum or whoever gives his vote in the Senate in the municipality of Tarentum shall own without malicious deception a dwelling roofed with at least 1,500 tiles within the town of Tarentum or within the territory of that municipality. If

151

152 153

154

155 156 157

any senator does not own such a house of his own, or if anyone of them purchases such a house or acquires possession thereof in such a way as to evade this law, he shall be liable to a penalty of 5,000 sesterces for each year to the municipality of Tarentum” (transl. Crawford). This is one of the points that illustrates the fundamental difference between a colony and a municipium. Cf. also Capogrossi Colognesi 2002, 4-5. Supra, note 56. If so, a Roman who had been sent against his will to a Latin colony, would only have not to build his house to return to Rome as a full citizen. This to me seems unbelievable. Galsterer 1988, 80, note 16 supposes that, in Irni, “it remains possible that many inhabitants, previously in a dependent status, did not benefit from the new rights, but became incolae.” This is the most likely solution. CIL I2 3201. ILS 6753. Gagliardi 2011, 65-69.


colonial migrants. Some legal texts decree that incolae had to be domiciled within the city, just as the colonists were, and share their duties. This would be a possible status for those colonial migrants who were unable to build their house in the stated time.158 These cases all illustrate how diverse the social, ethnic and legal statuses of the various people living in a Roman colony could be. It is possible to distinguish I) the citizens of the colony: Roman citizens, coming from Rome or from Roman colonies, as well as municipes. Some of those were the true colonists who had signed up for colonization and others were II) possessores on the colonial territory. There were also III) incolae of different statuses, among which were also natives. Finally, there were perhaps IV) incensi (“those who had initially given their names for the colony”) but who did not meet the requirements to qualify as a citizen (for this category there is no epigraphic evidence). Maybe this category was included amongst the incolae. Probably there were many more categories of which there are no remaining traces. Adding to this colorful panorama the effects of individual migration and of naturalization granted as a personal favor, it is obvious how important it was to keep accurate lists for each category of people contributing to the city. These lists had to record not only the people but also the pieces of land they owned or possessed, for obvious taxation purposes.159

Between personal and territorial status In terms of landed property, citizens of a Latin colony possessed their land under the laws of their city and could not be owners ex iure Quiritium.160 Conversely, in municipia, which apparently followed Roman institutions, property had the same status as that in Roman colonies.161 Since there are normally no new colonists in a municipium, it can be assumed that the land of municipia was assigned to veteres possessores by professio (when there was not already a local land register).162 The same procedure was followed in the (re)assignment of colonial land allotments to the previous occupants of these lands, allowed to remain on the property that had previously been theirs. Such a procedure is shown by the famous example of

158

159

Under the Empire, the incolae had the right to vote, all in the same curia, by drawing lots. Law of Malaca, CIL II 1964; ILS 6089: “[53]: r(ubrica) in qua curia incolae suffragia / ferant / quicumque in eo municipio comitia IIuiris / item aedilibus item quaestoribus rogan/dis habebit ex curiis sorte ducito unam / in qua incolae qui cives R(omani) Latinive cives / erunt suffragi ferant eisque in ea cu/ria suffragi lato esto.” Modestinus, Dig. 50.1.35. This very right is granted to the Latins under the Republic. For instance, the Roman lawyers point out that provincial land could not be sold by mancipatio, which remained the privilege of the state (and the Princeps under the Empire). Frontin., De contr. 36 L.: “Possidentur tamen a privatis, sed alia condicione: et veneunt, sed nec mancipatio eorum legitima potest esse. Possidere enim illis quasi fructus tollendi causa et prestandi tributi condicione concessum est”. Gai., Inst. 2.7; Gai., Inst. 2.31. Cf. Capogrossi Colognesi 2004, 10. For the main definitions, see Roselaar 2010, 121-36 (with discussion of

160

161

162

vectigalia on ager quaestorius: 122-24) and 136-44 for the category of “ager publicus belonging to communities”; Capogrossi Colognesi 2002, passim; Saumagne 1965, 80-85. At this time, one solution was offered by the ius Italicum, which assimilated the soil of a provincial colony to a piece of Italy, thereby allowing the possessor to become a dominus, and to sell his domain by mancipatio. See Chouquer and Favory 1992, 29. Capogrossi Colognesi 2002, 3. Ibid., 194, recalling that the Social War contributed to the general uniformity of land statuses in Italy. However, Gaius (Gai., Inst. 3.145) seems to believe that the normal situation in municipia was that of a perpetual conductio-locatio, which he admits could easily be confused with emptio. The veteres possessores might have been the natives to whom parts of land were given back (Gagliardi 2011, 65; id. 2014, 66-67), but could also have been colonists from a previous phase, or, after the Social War, municipes, which territory was used for a new colony.




FIG. 2. Minturnae, schematic drawing after a vignette of Hyginus, Vat. Palat. lat. 1564, f. 88r. fig. 89 Th., 150 L.

Minturnae (fig. 2).163 Certainly, the same rule was also applied to the old colonists when a decision was made to found a new colony on the same territory. This might also explain the occasional occurrence of what are called ‘double communities’.164 In this context, it should be remembered that it was rarely the case that the conquered were completely exterminated. The agrimensores clearly underline the necessity of taking into account the presence of peregrine communities in colonial territory. In some cases, their lands retained peregrine status and did not fall under the authority of the colony.165 What is more, in a somewhat difficult passage, Hyginus notes that a peregrine oppidum could still exist in the territory of a colony. In this instance he says the oppidum would retain its previous status. In fact, the authority of the colony was limited to what had been given to the new

163



Chouquer and Favory 1992, 53. The drawing shows the land of the colony: on the right is the division into regular squares for new settlers; on the left is the ager adsignatus per professiones to the veteres possessores who received the right to stay where they had previously been settled. Limitatio is made here by using natural landmarks and visible monuments (here a grave and a statue of Diana). Sic. Flacc., Cond. agr. 159.1420 L., 125 Th.: “aliquando vero in limitationibus si ager etiam ex viciniis territoriis sumptus non suffecisset, et auctor divisionis assignationisque quosdam cives coloniis dare velit et agros eis assignare, voluntatem suam edicit commentariis aut in formis extra limitationem, MONTE ILLO, PAGO ILLO, ILLI IVGERA TOT, aut ILLI AGRVM ILLVM, QVI FVIT ILLIVS; hoc ergo genus fuit assignationis sine divisione.” Clavel-Lévêque et al. 1993, 81-82.

164

165

Cf. Pelgrom in this volume, who, however, uses the term to describe a different scenario (i.e. the sharing of a territory by a colonial and a non-colonial community). On this see below. See now Gagliardi 2011, 68-72. Hyg., Cond. agr. 82 Th.; Hyg., Cond. agr. 119 L: “Alioqui<n>, cum ceteros possessores expelleret et pararet agros quos divideret, quos dominos in possessionibus suis remanere passus est, eorum condicionem mutasse non videtur: nam neque cives coloniae accedere iussit.” Hyg., Cond. agr. 83 Th.; 120 L.: “Illud vero observandum, quod semper auctores divisionum sanxerunt, uti quaecumque loca sacra, sepulchra, delubra, aquae publicae ac vicinales, fontes fossaeque publicae vicinalesque essent, item si qua conpascua, quamvis agri dividerentur, ex omnibus eiusdem condicionis essent cuius ante fuissent.”


Fig. 3. Orange, “Cadastre B,” slab III J (drawing by M. Tarpin).

community (cf. Pelgrom in this volume).166 This approach can be illustrated by Centuriation B of Arausio (Orange) in which the lands assigned or entrusted to the colony or given back to the Tricastini are carefully listed (fig. 3).167 Lorenzo Gagliardi nevertheless concludes that these Tricastini had been placed under the authority of the colony as incolae.168

166

Hyg., Cond. agr. 81 Th.; Hyg., Cond. agr. 118 L.: “Alioquin saepe et intra fines dictos et oppidum est aliquod; quod cum in sua condicione remaneat, <e>idem est in id ipsum ius, quoi ante fuit: ita illa interpretatione oppidum civesque coloniae pariter adsignaret. Sed nec fuisse<t> necesse in legibus ita complecti quos agros, quae loca quaeve aedificia, si universa regio, quae cancellata erat, coloniae iuris dictioni accederet: dixisset enim intra finem illum et flumen illud et viam illam iuris dictio cohercitioque esto coloniae illius. Ita excipitur id quod non adsignatum est vocaturque subsicivum. Ergo, ut saepius repetam, hoc ait: quos agros, quae loca, quaeve aedificia dedero adsignavero, in eis iuris dictio cohercitioque esto [colonorum] coloniae illius, quoius civibus adsignati erunt agri.” The act that formalized the return of the land and oppidum to the natives and the definition of the civic body of a peregrine city should have more or less taken a form known from the lex Antonia de Termessibus of 68 B.C., fixing a date on which the list of people and of their property would be closed. See CIL I2, 589; ILS 38; Crawford 1996, vol. 1, 319-40, col. I, l. 1-

167

168

7: “quei Thermeses Maiores Peisidae fuerunt, queique / eorum legibus Thermesium Maior<u>m Pisidarum / ante k(alendas) April(es), quae fuerunt L. Gellio Cn. Lentulo co(n)s(ulibus), / Thermeses Maiores Pisidae factei sunt, queique / ab ieis prognati sunt erunt, iei omnes / postereique eorum Thermeses Maiores Peisidae / leiberi amicei socieique populi Romani sunto.” (“Whoever have been Termessians and whoever by the laws of these Termessians have become Termessians before April I, when Lucius Gellius and Gnaeus Lentulus were consuls, and whoever have been and are born from them: all these and their descendants, being Termessian citizens, shall be free, friends, and allies of the Roman people.”) (transl. Crawford). Cf. Gagliardi 2014, 62-66. See http://www.archeogeographie.org/index.php?rub=arpentage/romain/orange/b196 and http://www.archeogeographie.org/index.php?rub=arpentage/romain/orange /cadastres. Gagliardi 2011, 66-69. But, unlike the imperial example of Pisidian Antioch cited by Gagliardi, the Tricastini had their own city, Augusta Tricastinorum (Saint-Paul-Trois Châteaux) near Arausio.




Conversely, Roman space might also be encountered in peregrine territory. Siculus Flaccus mentions, for example, montes Romani in Picenum. These hills became a Roman territory inside a conquered territory which, however, was not assigned.169 In the opinion of the author, the few pagi Romani attested by inscriptions should be interpreted in the same way. The two most obvious examples are on the Tabula alimentaria Ligurum Baebianorum (fig. 4). The tabula reveal a pagus Romanus located in finibus Beneventanorum and also a pagus Romanus in Ligustino, located in Beneventano.170 It can be suggested that the first mentioned pagus was a plot of Roman public land in the territory of the former Latin colony of Beneventum whereas the second would have been a parcel of Roman land in the territory allotted to the Ligurian community at the expense of the Hirpini, later integrated into the triumviral Roman colony of Beneventum. The mentioning of a res publica Baebianorum in the inscription suggests that a Ligurian oppidum was still there under Trajan. The pattern that can be abstracted from these examples reflects, once again, the complexity of colonial territorial history (fig. 4). These examples also elucidate the diverse and complex connections that existed between taxation, different jurisdictions and individuals (cf. Hermon in this volume). In this context it is interesting to consider Pomponius’ definition of territorium, by a false etymology, as an area subject to the power of a magistrate.171 It can be deduced from the combination of evidence that territories did not always constitute broad, contiguous lands (cf. the contributions of Pelgrom and Stek in this volume). The fact that a city, Capua is the best-known example but Arpinum can also be cited,172 benefited from vectigalia in remote areas makes it difficult to determine with any certainty what law would have been in force in these territories. Perhaps Frontinus’ statement is helpful in this context; he labels such enclaves praefecturae.173 These prefectures may have just been additional sources of income. One can conceive they were, for example, peregrine territories acquired by Rome for which the tribute was assigned directly to a colony rather than to the Roman aerarium. But the term praefectura, especially if it could be used for Republican Italy, might also indicate that jurisdiction over these distant lands was exercised by the colony because the occupants fell under the same jurisdiction as those of the colonial pertica, at least according to the principle set out above.174 Conversely, it is always possible that occupied land was given back to the former owners or possessores, excluding them from any assignment (agri excepti/concessi) and probably from taxes.175 Certainly some landlords obtained exemptions for their lands from the assignments as a personal favor176 but there were also some conquered peoples who were allowed to recover a portion of their original territory, as the examples

169

170

171

172



Sic. Flacc., Cond. agr. 21 (136-37 L.): “Alii ita remanserunt ut tamen p(opuli) R(omani) <terri>toria essent; ut est in Piceno, in regione Reatina, in quibus regionibus montes Romani appellantur. Nam sunt p(opuli) R(omani) <terri>toria quorum vectigal ad aerarium pertinet.” Hence the reconstruction of events I proposed more than a decade ago in a drawing (in the style as the Roman land surveyors did) seems in light of this evidence potentially invalid (Tarpin 2002, 453; interpretation of G. Chouquer:http://www.archeogeographie.org/index.php? rub=dossiers/etudes/fragment). Pompon., Dig. 16.239.8: Territorium est universitas agrorum intra fines cuiusque civitatis; quod magistratus eius loci intra eos fines terrendi, id est summovendi ius habent. Cic., Fam. 13.11.1; Dio Cass. 49.14.5; Vell. Pat. 2.81;

173

174

175 176

Suet., Aug. 46; Tac., Hist. 1.78.1; Rigsby 1976; Ducrey 1969. For other examples, see CIL XI 5291; Paci 199697, etc. Frontin., De contr. 49 L., 40 Th.: “Coloniae quoque loca quaedam habent adsignata in alienis finibus, quae loca solemus praefecturas appellare […] solent et privilegia quaedam habere beneficio principum, ut longe [et] semotis locis saltus quosdam reditus causa acceperint.” Cf. Capogrossi Colognesi 2002, 14. Supra note 171. For a new example in Macedonia, see Rizakis 2012. Chouquer and Favory 1992, 34-5. See Deniaux 1993; id. 1998. Agen. Urb., De contr. 81 L.: “Subsicivorum autem genera sunt duo […] aliut etiam integris centuriis intervenit”. And he adds obviously: “de quo maximae controversiae agitantur”.


FIG. 4. Beneventum: schematic layout of cities and pagi after the triumviral colony of 42 B.C. (drawing by M. Tarpin).

given above show.177 As a rule, these exceptions in the status of the soil represented complications, and Agennius Urbicus says precisely that de proprietate agatur, non de loco (p. 80 L.); thus from a technical perspective, the difficulty was not that of the legal definition of a space but of ascertaining property titles. Summing up, it is possible to find examples of Italian colonial territory assigned with full ownership ex iure Quiritium, ager publicus entrusted as a possessio either to colonists or to extra-colonial citizens or even to peregrines and land left under the terms of its previous status, which might have been a possessio of ager publicus, or even a peregrine property. In the last case, as seen above, jurisdiction was exercised by the original community. This is probably why an agrimensor had no choice but to mention a peregrine oppidum on the forma. The question of whether all these different types of territory were subject to any strict limitatio cannot be resolved by the important passage in what is known as the lex Mamilia Roscia Peducanea Alliena Fabia, quoted by the land surveyors. This text states: Anyone who, according to this law, founds a colony or establishes a municipium, prefecture, forum, (or) conciliabulum, shall ensure that boundaries and decumani are drawn and boundary stones are set up in the

177

This is a category of agri redditi (Chouquer and Favory 1992, 35-36); Sic. Flacc., Cond. agr.156 L.: “Nec tamen omnibus personis victis ablati sunt agri; nam quorumdam dignitas aut gratia aut victorem ducem

movit, ut ei<s> concedere agros suos”. Gagliardi 2011, 70, (based on Siculus Flaccus and Hyginus) notes that sometimes all the land could have been confiscated and that the city might have been restricted to within its walls.




territory that will be included in the boundaries of this colony, municipium, forum, conciliabulum, (or) prefecture.178

According to Luigi Capogrossi Colognesi, this clause would have been valid only under a specific law (ex hac lege) perhaps Caesarean.179 In short, it would be hazardous to generalize on the matter, especially because the gromatic drawings give precise exceptions to limitatio.

Conclusion: back to the Palatinus vignette The complex panorama sketched above must, as promised at the start of this paper, lead to a better comprehension of the famous gromatic vignette of the Palatinus 1564 (fig. 1). It will be clear by now that the vignette is a strongly simplified version of reality.180 However, at the same time, is succeeds in transmitting crucial information on this territory. The main point the composer of the vignette wanted to communicate was that the colony (colonia Augusta) was located in a predominantly centuriated area, organized in the vicinity of three major intersections. The landscape is furthermore shaped by mountains, the public status of which is specified, and by rivers. Rivers were particularly important because their course might determine a boundary or the presence of subsecivae. Besides the colony, on the border of the limitatio, the vignette indicates an oppidum Atelle, which should not be identified with the municipium of Atella itself but rather should be taken as an example of oppida that survived in colonial territory, perhaps in the manner as that of Caudium.181 On the left, a third town with its own system of major axes is designated as praefectura coloniae Augustae ex finibus Antemnatium. It is obviously a prefecture of the colony located at the center of the vignette and is situated on land taken from a neighboring people.182 The vignette thus successfully represents three distinct realities: the colony – a city (Roman or Latin) founded with a constitution based on its specific relationship with Rome; the old peregrine oppidum; and the prefecture, a remote community subject to the authority of the magistrates of the colony but without a proper constitution of its own. The forma also includes the names of the neighboring communities from whom land had been seized for assignment: the Ottimi, Bitivamenses, Hirrenses, and Venetiatenses. The word veterum (top left) is certainly incomplete (perhaps veterum possessorum, or veterum possessorum, concessum, this can also be seen on the Minturnae drawing). The reference on the right is unclear but the correction proposed by Gérard Chouquer and François Favory would perhaps enable us to read, “prefecture dependent on the oppidum Atella from the territory of the Hirrenses”,183 which would complete the range of possible situations. Two mountains are public areas, whereas a third marks a boundary. It should be assumed that such montes finitimi represented a common space between the two communities. Finally, the author of the vignette

178

179 180



Lex Mamilia Roscia Peducanea Alliena Fabia: Grom., 264 L.: “Qui hac lege coloniam deduxerit, municipium praefecturam forum conciliabulum constituerit, in eo agro, qui ager intra fines eius coloniae municipii fori conciliabuli praefecturae erit, limites decumanique ut fiant terminique statuantur curato”. Capogrossi Colognesi 2002, 211-12. I chose this version rather than that of Gudianus. My drawing incorporates the legends as they can be read, with uncertainties in some illegible names, perhaps attributable to copy errors. Cf. Chouquer and Favory 1992, 59-63; Capogrossi Colognesi 2002, 299.

181

182

183

CIL IX, 2165: “Colonia Iulia Concordia Aug(usta) Felix Beneventum devota maiestati Aug(ustorum) in territorio suo quod cingit etiam Caudinorum civitatem muro tenus”; confirmed by Lib. colon. 232 L. Cf. Gabba 1994, 86; Giampaolo 1991, 123-31. Antemnae was supposedly taken by Romulus. The choice of this name therefore recalls the antiquity of the procedure, and justifies our interpretation of it as a theoretical example. Chouquer and Favory 1992, 59: “praetensura / praefectura (?) ex finibus Hirrensium Atellenatibus adlat(a).”


indicates two private areas which are excluded from the centuriation. One is the piece of land named concessum Lucio Titio Lesple, a rich landowner with political connections, and the second is unnamed. A few final details that can be added concern the mons sacer. Sacred territory is mentioned by Siculus Flaccus referring to different types of shrines outside the centuriated area.184 This is confirmed to some extent by the Oscan inscription of Abellinum which describes a sacred space that fell under the joint control of Abellinum and Nola. The quaestor of Abellinum indeed signed the agreement, whereas Nola sent its meddix deketasiís.185 Worship there was open to representatives of both communities.186 The Palatinus vignette thus turns out to be a rather convincing graphic reflection of what the territory of a city with a long and dynamic history might have looked like in legal, social and ethnic terms. Moreover, it illustrates how various elements can intersect within the concept of territory. The agrimensores often insisted on the legal function of territory and on the extent of magistrates’ powers. But they also discuss the vectigalia cities could collect and illustrate that legal jurisdiction and taxation did not always precisely overlap. Furthermore, it is seen how individual privileges result in gaps in the centuriation on the vignette, an apt illustration that the distribution of land was not always as straightforward as, for example, Appian proposed.187 The represented complex landscape agrees particularly well with what is known from Octavian’s letter in which he talks about the honors accorded to Seleucos.188 The immunity offered to the navarch at the time he was made a citizen meant that his property was exempt from city taxes as well as from charges that he otherwise would have had to pay to Rome. The favors were granted to Seleucos as a personal reward so that any new acquisition he made would be exempt not only from city taxes but from those of the Empire. Once one understands the dynamics of Roman dealings with the different statuses of people and lands in colonial context, it is no longer surprising that numerous treatises De controversiis agrorum were published in Antiquity. Good lawyers and good land surveyors were essential to determining the legal and tax inventory of the dynamic Roman world.189

Acknowledgements This paper is the development of a talk presented in Ravenstein (Nijmegen) as part of a ESF workshop organized by Tesse D. Stek and Jeremia Pelgrom. I would like to thank them for the invitation and for their lectures presented in Grenoble on Roman colonial territorial organization, which contributed to this work. I must also thank John N. Dillon for reviewing and improving the English text.

184

185 186

Sic. Flacc., Cond. agr. 162-63 L.: “Collegia sacerdotum itemque virgines agros et territoria quaedam etiam determinata et quaedam aliquibus sacris dedicata, in eis etiam lucos, in quibusdam etiam aedes templaque. Quos agros quasve territoriorum formas aliquotiens comperimus extremis finibus conprehensas sine ulla mensurali linea, modum tamen inesse scriptum.” Hyg., Cond. agr., 83 Th.; 120 L. (supra, note 165). Aberson 2010, 409; Rix, ST, Cm 1, p. 114 sq. The agreement between Rome and Clusium included a similar clause (Liv. 8.14.2).

187

188 189

App., B Civ. 1.1.7. Senatore 2004, 86. The distinction between ager adsignatus and ager occupatus is certainly true in his analysis, but does not systematically cover the distinction between agri culti and agri inculti. Supra, pp. 178-79. See Capogrossi Colognesi 2002, 210: “[…] il tipo di relazioni esistenti tra regime giuridico e assetto gromatico del territorio romano […] è un problema quasi mai affrontato nel pur straordinario corpus del nostro sapere giuridico.”




References Aberson, M. 2010. “Les lois sacrées en Italie du VIe au Ier siècles av. J.-C., auteurs, formulations, affichages, applications,” in L. Lamoine, C. Berrendonner, M. Cébeillac-Gervasoni (edd.), La Praxis municipale dans l’Occident romain (Clermont-Ferrand) 401-19. Ando, C. 2008. “Aliens, ambassadors, and the integrity of the Empire. Part I. The conduct of war in the ancient world and early islamic history,” Law and History Review 26, 491-519. Bispham, E. 2009. From Asculum to Actium. The municipalization of Italy from the Social War to Augustus (Oxford). Broadhead, W. M. 2001. “Rome’s migration policy and the so-called ius migrandi,” Cahiers du Centre Gustave Glotz 12, 69-89. Broadhead, W. M. 2002. Internal migration and the transformation of Republican Italy. Doctoral thesis, Univ. of London, 2002. Broadhead, W. M. 2003. “The local élites of Italy and the crisis of migration in the IInd century BC,” in M. Cébeillac-Gervasoni and L. Lamoine (edd.), Les élites et leurs facettes. Les élites locales dans le monde hellénistique et romain (Rome) 131-48. Broadhead, W. M. 2004. “Rome and the mobility of the Latins: problems of control,” in C. Moatti (ed.), La mobilité des personnes en Mediterranée de l’Antiquité à l’époque moderne. Procédures de contrôle et documents d’identification (Rome) 315-35. Broadhead, W. M. 2008. “Migration and hegemony: fixity and mobility in second-century Italy,” in L. de Ligt and S. Northwood (edd.), People, land, and politics. Demographic developments and the transformation of Roman Italy, 300 BC-AD 14 (Leiden/Boston) 451-70. Brunt, P. 1982. “The legal issue in Cicero, Pro Balbo,” CQ, New Series 32-31, 136-47. Capogrossi Colognesi, L. 2002. Persistenza e innovazione nelle strutture territoriali dell’Italia romana. L’ambiguità di una interpretazione storiografica e dei suoi modelli (Naples). Capogrossi Colognesi, L. 2004. “Le statut des terres dans l’Italie républicaine. Un aspect de la romanisation des campagnes (IVe-Ier siècle avant J.-C.),” Histoire et Sociétés Rurales 22-22, 9-28.



Chastagnol, A. 1995. La Gaule romaine et le droit latin (Lyon). Chouquer, G. 2008. “Arpentage, cadastre et fiscalité foncière, de l’Antiquité à l’époque moderne,” Études rurales 181-81, 203-36. Chouquer, G. and Favory, F. 1992. Les arpenteurs romains. Théorie et pratique (Paris). Clavel-Lévêque, M., et al. (edd.) 1993. Siculus Flaccus: Les conditions des terres (Naples). Coşkun, A. 2009. Bürgerrechtsentzug oder Fremdenausweisung? Studien zu den Rechten von Latinern und weiteren Fremden sowie zum Bürgerrechtswechsel in der Römischen Republik (5. Bis frühes 1 Jh v.Chr.) (Stuttgart). Coşkun, A., forthcoming. “Zum Rechtsstatus der spätrepublikanischen Kolonie Comum und ein zweifelhafter Fall von Bürgerrechtsanmaßung im Jahr 51 v. Chr.” Crawford, M. H. 1996. Roman statutes, 2 vols (London). Deniaux, E. 1993. Clientèles et pouvoir à l’époque de Cicéron (Rome) 1993. Deniaux, E. 1998. “Recherches sur les propriétés foncières des amis de Cicéron en Afrique,” L’Africa romana 12 (Olbia, Sassari) 151-60. Ducrey, P. 1969. “Trois nouvelles inscriptions crétoises,” BCH 93, 846-52. Ferrary, J.-L. 2003. “La législation romaine dans les livres 21 à 45 de Tite-Live,” in T. Hantos (ed.), Laurea Internationalis. Festschrift für Jochen Bleicken zum 75 Geburtstag (Stuttgart) 107-42. Gabba, E. 1994. Italia romana (Como). Gagliardi, L. 2011. “Brevi note intorno ai rapporti giuridici tra romani e indigeni all’interno delle colonie romane,” in A. Maffi and L. Gagliardi (edd.), I diritti degli altri in Grecia e a Roma (Sankt Augustin) 64-77. Gagliardi, L. 2014. “Approche juridique des relations entre Romains et indigènes. Le cas des colonies romaines,” in E. Gojosso, D. Kremer and A. Vergne (edd.), Les colonies. Approches juridiques et institutionnelles de la colonisation. De la Rome antique à nos jours (Poitiers) 59-76. Galsterer, H. 1988. “Municipium Flavium Irnitanum: a latin town in Spain,” JRS 78, 78-90. Galsterer, H. 1995. “La trasformazione delle antiche


colonie latine e il nuovo ius Latii,” in A. Calbi and G. Susini (edd.), Pro poplo Arimenese (Faenza) 79-94. Gascou, J. 1999. “Hadrien et le droit latin,” ZPE 127, 294-300. Giampaolo, D. 1991. “Benevento” in La romanisation du Samnium aux IIème et Ier siècles av. J.-C. (Naples) 123-31. Gordon, R. and J. Reynolds 2003. “Roman Inscriptions 1995-2000,” JRS, 212-94. Grandazzi, A. 1996. “Identification d’une déesse: Ferentina et la ligue latine archaïque,” CRAI 1401, 273-94. Humbert, M. 1978. Municipium et civitas sine suffragio. L’organisation de la conquête jusqu’à la guerre sociale (Rome). Hin, S. 2008. “Counting Romans,” in L. de Ligt and S. Northwood (edd.), People, land, and politics. Demographic developments and the transformation of Roman Italy 300 BC-AD 14 (Leiden/Boston) 187-238. Kremer, D. 2006. Ius latinum: le concept de droit latin sous la République et l’Empire (Paris). Laffi, U. 1995. “Sull’esegesi di alcuni passi di Livio relativi ai rapporti tra Roma e gli alleati latini et italici nel primo quarto del II sec. A.C.,” in A. Calbi and G. Susini (edd.), Pro poplo Arimenese (Faenza) 43-77. Lo Cascio, E. 2008. “Roman census figures in the second century BC and the property qualification of the fifth class,” in L. de Ligt and S. Northwood (edd.), People, land, and politics. Demographic developments and the transformation of Roman Italy 300 BC-AD 14 (Leiden/Boston) 239-256. McDonald, A. H. 1944. “Rome and the Italian confederation (200-186 BC),” JRS 34, 11-33. Martin, P.-M. 2001. “La tradition sur l’intégration des peuples vaincus aux origines de Rome et son utilisation politique,” in G. Urso (ed.), Integrazione mescolanza rifiuto - Incontri di popoli, lingue e culture in Europa dall’Antichità all’Umanesimo (Rome) 65-88. Moatti, C. 1993. Archives et partage de la terre (Rome).

Mouritsen, H. 2007. “The civitas sine suffragio: Ancient concepts and modern ideology” Historia 56. 2, 141-158. Muccigrosso, J. D. 2003. “The Brindisi ‘Elogium’ and the rejected ‘Lectio Senatus’ of Appius Claudius Caecus,” Historia 52-4, 496-501. Paci, G. 1996-97. “Terre di Pisaurum nella valle del Cesano,” Picus 16-17, 115-48. Piper, D. J. 1987. “Latins and the Roman citizenship in Roman colonies: Livy 34, 42, 5-6, Revisited,” Historia 36-1, 38-50. Raggi, A. 2004. “The epigraphic dossier of Seleucus of Rhosus: a revised edition,” ZPE 147, 123-38. Rigsby, K. J. 1976. “Cnossus and Capua,” TAPhA 106, 313-30. Rizakis, A. D. 2012. “Une praefectura dans le territoire colonial de Philippes les nouvelles données,” in S. Demougin and J. Scheid (edd.), Colons et colonies dans le monde romain (Rome) 87-105. Roselaar, S. T. 2010. Public Land in the Roman Republic: a social and economic history of Ager Publicus in Italy, 396-89 BC (Oxford). Salmon, E. T. 1933. “The Last Latin Colony,” CQ 27-1, 30-35. Salmon, E. T. 1936. “Roman Colonisation from the Second Punic War to the Gracchi,” JRS 26, 47-67. Sanchez, P. 2007. “La clause d’exclusion sur l’octroi de la citoyenneté romaine dans les traités entre Rome et ses alliés (Cicéron, pro Balbo 32),” Athenaeum 95-1, 215-70. Saumagne, C. 1965. “Les Domanialités publiques et leur cadastration au premier siècle de l’empire romain,” Journal Des Savants 1, 73-116. Senatore, F. 2004. “Il lessico delle distribuzione agrarie in Appiano,” in A. Storchi Marino (ed.), Economia, amministrazione e fiscalità nel mondo romano. Ricerche lessicali (Bari) 85-96. Smith, R. E. 1954. “Latins and the Roman citizenship in Roman colonies: Livy, 34, 42, 5-6,” JRS 44, 18-20. Stevenson, G. H. 1919. “Cn. Pompeius Strabo and the franchise question,” JRS 9, 95-101. Tarpin, M. 2002. Vici et pagi dans l’Occident romain (Rome).




Colonisation romaine et ‘espaces ripariens’ dans les Civitates Campaniae de Sylla aux triumvirs Ella Hermon

Résumé Cette communication a pour objectif de situer mes travaux sur le thème en question en mettant en lumière diverses formes de la colonisation romaine durant la République, perçues comme étant des alternatives au modèle institutionnel et urbain proposé par Edward Togo Salmon (v. Pelgrom et Stek dans ce volume). La communication vise également à ouvrir quelques perspectives d’histoire environnementale attribuant à la colonisation républicaine le rôle de ferment de la construction d’espaces ‘ripariens’. Je soulignerai ainsi le rapport intrinsèque existant depuis les Gracques entre la législation agraire et la colonisation, en subordonnant la seconde à la première forme de distribution de terres. La Guerre Sociale (91-88 av. J.-C.) a renversé ce rapport pour ériger la colonisation en génératrice de normes, soit par la reconstitution des lois agraires - modèles à partir de mesures de nature diverse, soit pour organiser la gestion des ressources naturelles en fonction de la maîtrise des ressources en eau. Pour sa part, Salmon identifie les normes d’un modèle urbain et institutionnel des deux formes urbaines de colonisation – romaine et latine – en vigueur depuis la dissolution de la Ligue latine en 338 av. J.-C. sans prendre en compte la forme parallèle de colonisation rurale qui peut être identifiée en Italie au IIIe s. av. J.-C. dans l’intervalle entre M. Curius Dentatus et Flaminius. Celle-ci n’a pas bénéficié d’un degré semblable de conceptualisation. De plus, les normes et les procédures de la colonisation romaine et latine se mettent en place lentement et sans la rigueur absolue d’un modèle d’application. La période des Gracques nous a semblé être une étape importante dans l’évolution de la colonisation républicaine pour avoir créée une ambigüité initiale – de agris dandis adsignandis et de coloniis deducendis – dont les lois agraires sont désormais investies. Cette dualité aurait donné naissance simultanément à une colonisation urbaine et rurale en Italie et à travers l’Empire. Néanmoins, la Guerre Sociale a opéré une synthèse entre la colonisation urbaine et rurale en Italie et a mis en évidence les limites du modèle urbain et institutionnel avancé par Salmon. En effet, dès Sylla aux Triumvirs, la colonisation romaine à répétition sur les mêmes sites en Italie a fait éclater les cadres de la Cité classique exprimés par le binôme urbs-ager (sur urbs v. Sisani dans ce volume). Cela a été réalisé grâce à trois phénomènes conjugués: la création des municipes de type nouveau en Italie; la législation agraire et les distributions de terres aux différentes catégories sociales; la colonisation militaire avec des formes de fondation alternatives au modèle institutionnalisé proposé par Salmon (v. Tarpin dans ce volume). À la lumière des recherches récentes sur les dimensions environnementales de ces questions, cette communication se propose de suggérer quelques éléments d’un nouveau modèle de colonisation romaine mis en place à la fin de la République. Un aperçu rapide du fonctionnement du trinôme urbs-iter-ager dans les notices du Liber coloniarum I nous permettra d’expliquer la situation coloniale dans les Civitates Campaniae caractérisée à cette époque par des variations climatiques et des inondations en l’Italie centreméridionale. Je tenterai ensuite une double explication du vocabulaire tripartite de la cité – urbs-iter-ager – en tenant compte de la mention insolite de la servitude d’iter dont les dimensions sont souvent sans proportion avec celles obligatoires et connues pour la limitatio et le système de voirie. Ainsi la raison d’être de cette mention dans le Liber serait soit la séparation des communautés distinctes par des voies vicinales, soit la nécessité de protéger les zones ‘ripariennes’ inondables. Les deux éventualités ne sont pas




incompatibles vu la tendance de l’avancement de l’habitat vers des zones à risque environnemental. La constitution d’espaces ‘ripariens’ nous apparaît ainsi se réaliser au fur et à mesure avec des travaux de bonification et de cadastration des territoires adjacents grâce aux interventions agraires effectuées à partir de Sylla. Nous nous inspirons également des représentations graphiques des perticae coloniales des manuscrits médiévaux du Corpus Agrimensorum Romanorum qui suggèrent l’existence de divers modèles d’espaces ‘ripariens’ dans le cadre de la colonisation romaine.

Introduction: colonisation et législation agraire républicaine Des lois agraires, une ambiguïté fondamentale depuis les Gracques. Depuis 133 av. J.-C. et suite à la loi agraire du tribun de la plèbe, Tibérius Gracchus,1 les lois agraires de modum agrorum se transforment en lois de agris dandis adsignandis en inaugurant ainsi à travers l’Italie une vague d’assignations viritanes sur l’ager publicus. Fruit de conquêtes, l’ager publicus avait gardé un statut plutôt théorique grâce aux modalités décrites en détail par Appien et Plutarque. Dix ans plus tard, avec le tribun de la plèbe Caius Gracchus, la loi agraire devient loi-programme2 avec la proposition de blocs de colonies en Italie et à travers l’Empire,3 en complément des distributions viritanes à la plèbe dans des communautés italiennes et le long des voies.4 Du coup, la législation agraire assume une dualité initiale de agris dandis adsignandis et de coloniis deducendis. Cette dualité de la législation agraire crée une ambiguïté fondamentale de la colonisation: alors que la colonisation urbaine trouve tout un arsenal institutionnel qui accompagne la charte de fondation, la lex colonica et la cérémonie de la deductio, la colonisation rurale est plus expéditive et elle est connue le plus souvent grâce aux commissions décemvirales et des traces sur le terrain. Si la colonisation urbaine est à l’origine du modèle institutionnel développé par Salmon qui alterne chronologiquement des blocs de colonies romaines et latines munies de procédures et de normes précises sur le nombre de colons, les dimension des lots de terres etc., la colonisation rurale n’a pas bénéficié d’un modèle semblable de conceptualisation.5 Cependant, en utilisant le système parallèle de la civitas sine suffragio la construction des voies ainsi que des distributions viritanes de terres, une colonisation rurale avec le développement précoce des conciliabula semble s’étendre dans l’intervalle entre Dentatus6 et Flaminius7 de la Sabine à l’ager Gallicus et Picenus, en offrant ainsi des formes de colonisation alternatives à la deductio et au modèle urbain en Italie (v. Vermeulen dans ce volume). En effet, des citoyens romains et des vétérans sont installés sur le territoire de communautés locales de statuts divers sans deductio coloniale mais par leur regroupement en conventus dans les civitates sine suffragio et les préfectures de iure dicundo.8 Les grandes voies consulaires ont vu le jour au IIIe s. av. J.-C. et cette colonisation viritane a pu également se réaliser par

1 2 3 4 5



Hermon 1976a; ead. 1997a. Ead. 1982. Ead. 1972; ead. 1975; ead. 1976b. Ead. 1982. Salmon 1967; id. 1969. Cette thèse se situe dans la tradition des recherches antérieures sur la colonisation romaine et dans la tendance généralisée à cette époque d’ignorer les populations rurales par rapport aux bourgeoisies citadines dans les sociétés impériales (Benier et Hopital 1983, 430). Ce sont des retombées du courant épistémologique qui a donné naissance aux thèses de l’impérialisme romain dont

6 7 8

la colonisation est considérée un apanage. À l’ère de la décolonisation et de la démystification des thèses sur l’impérialisme romain, il convient désormais de considérer cette colonisation comme un produit propre de la société romaine et de son évolution, lorsque la conquête n’est pas forcement arrimée aux destructions massives des structures et des populations indigènes en Italie et à travers l’Empire. Hermon 1997b; ead. 1998a. Ead. 1989. Hermon 1989; ead. 1997b; ead. 1998a; ead. 2001.


l’installation de communautés du type de viasiei vicanei comme ce fut probablement le cas de la via Curia9 et de la via Flaminia (v. Bradley dans ce volume).10 Par ailleurs, l’impact de l’imperium des généraux aux fins de la colonisation militaire fut précoce à Réate en Sabine. Ces possibilités nous ont semblé des solutions plausibles pour envisager une application réelle des actes, considérés légendaires, de Marcus Curius Dentatus à la fin des guerres samnites et, un demi-siècle plus tard, celle de la lex Flaminia de agris dandis adsignandis, considérée longtemps sans effet. Il a été ainsi possible d’intégrer ce mouvement colonisateur romain en Italie au même titre que la colonisation romaine et latine.11 Les deux formes reconnues de colonisation, romaine et latine, auraient été instituées suite à la dissolution de la Ligue latine en 338 av. J.-C. Elles se distinguent davantage par les profils que par des normes mises de l’avant par Salmon et dont l’application mathématique a été remise en question.12 La colonisation latine ne semble pas avoir modifié le droit latin des Priscae Latinae Coloniae qui favorisait une colonisation mixte et, dans un premier temps, la colonisation romaine se dessine comme un siège de préfecture sur le territoire des tribus rustiques.13 Nous avons plutôt identifié un profil économique du point de vue de la propriété des terres: le développement de la propriété moyenne par la colonisation latine; la complémentarité des terres communautaires (ager publicus) par des lots privés pour la colonisation romaine. Le profil politique à la fois intégrateur et assimilateur des populations locales accompagne la colonisation latine, tandis qu’une fonction d’enclave institutionnelle d’appoint au système de la civitas sine suffragio peut être attribuée à la colonisation romaine d’avant la guerre sociale en Italie.14 Ce rôle d’enclave institutionnelle pouvant servir

9

10

11

Le trace de la via Curia permet d’envisager l’existence d’assignations viritanes entre Réate et Interamna Nahars en Sabine, ce qui suppose que les indigènes, auxquels on avait préalablement octroyé la civitas sine suffragio, ont accompli les travaux d’aménagement de la voie (Hermon 2001, 188 sqq.). À titre d’exemple, une centuriation précoce à Saepio dans la vallée de Cesano semble être liée davantage à l’aménagement des trajets locaux sur la via Flaminia qu’au voisinage de la colonie latine d’Ariminium dans le Picenum (Hermon 2001, 258 sqq.). Il est aussi intéressant de mentionner que le trajet de cette voie exclue aussi bien la colonie de Sena Gallica que l’habitat de Suessa, le dernier tirant son origine plutôt des assignations viritanes de 232 av. J.-C., Dall’Aglio 2010, 35. Les cippi du lucus Pisauriensi qui sont considérés contemporains de la Lex Flaminia de agris dandis adsignandis de 232 av. J.-C. et de la formation d’un conciliabulum. Il semble également que les centres ruraux d’Ostra et Suessa dans l’ager Gallicus se détachent assez tôt du territoire de la colonie romaine de Sena Gallica (283 av. J.-C.) (Hermon 2001, 220 sqq., 258-61). En effet, des recherches récentes confirment la naissance comme préfecture et déjà avec un aspect monumental de la future cité de Suessa au lendemain des assignations viritanes de la lex Flaminia. Les traces de l’habitat datées de la pre-

12

13

14

mière moitié du IIIe s. av. J.-C., récemment mises à jour, couvrent le parcours Nevola-Cessano en englobant la cuve Sassoferrato et la côte (Dall’Aglio 2010, 35; De Maria 2010, 22). Ainsi, entre autres, la programmation du groupe d’une dizaine de colonies maritimes jusqu’en 241 av. J.-C. est remise en question, alors qu’une véritable programmation des colonies romaines daterait des années 19794 av. J.-C.; la question de la nomination des commissions agraires qui se pose seulement en 296 av. J.-C. avec la fondation d’un premier couple de colonies par paires, Minturnae et Sinuessa, (Hermon 2001, 210-7 et bibliographie). Pour la continuité du droit latin des priscae coloniae latinae (Hermon 1998b, 164-73) et contre la conviction de Salmon (1969, nn. 63 et 67) sur un nouveau droit latin après 338 av. J.-C. qui repose sur une présumé lex colonica de Cales fondée en 334 av. J.-C. Quant à Antium, réputée être la première colonie romaine en 338 av. J.-C., il faut tenir compte de la confusion des sources qui décrivent une telle fondation en 467 av. J.-C. plutôt qu’en 338 av. J.-C. et dont l’identité distincte du municipium de 317 av. J.-C. est imprécise (Hermon 2001, 201 sqq.). À Terracina et à Minturnae, l’habitat indigène s’est développé à côté des colonies fondées en 296 av. J.-C. en jouissant du statut de civitas sine suffragio (Humbert 1978, 189). Hermon 2001, 202.




souvent de préfecture de iure dicundo expliquerait à notre sens pourquoi les procédures reconnues de la colonisation se mettent en place successivement à partir de la colonisation romaine.15 Néanmoins, la colonisation urbaine aussi bien que celle rurale se sont épanouies au sein des diverses formes de communautés distinctes qui se sont partagées les territoires. Dans les faits et dans un laps de temps variable, les fonctions économiques du territoire, représentées par des communautés rurales, fora et conciliabula, fusionnent avec les fonctions juridiques-administratives (praefectura) et politiques et urbaines (municipium)16 à côté ou dans la pertica des colonies romaines ou latines. D’autre part, la Guerre Sociale, qui fait de l’Italie le théâtre d’affrontements et le lieu d’implantation de colonies de vétérans dans des communautés préexistantes, embrouille cette dualité initiale de la colonisation tant par l’utilisation des procédures et la nature des communautés que par le statut des personnes qui partageaient désormais la citoyenneté romaine. Après la Guerre Sociale, avec la création de municipes de type nouveau, l’établissement des communautés de vétérans peut se réaliser par les leges datae et leur installation fut confiée aux magistrats municipaux. Dans le conflit politique opposant les populares et les optimates, les lois agraires sont brandies comme des programmes de propagande en opérant plutôt une synthèse de la dualité initiale de la législation agraire et en creusant davantage l’ambiguïté de la colonisation romaine en Italie dès Sylla aux Triumvirs. Cet état de fait révèle néanmoins la nécessité de légaliser cette synthèse des deux formes de colonisation qui n’ont pas joui du même degré de modélisation et qui ont évolué dans le sens de la création de communautés distinctes se partageant le même territoire. L’octroi de la citoyenneté romaine à toute l’Italie après la Guerre Sociale ne fait qu’accélérer le processus de fusion de ces communautés en érigeant la colonie romaine comme l’aboutissant de ce processus. Les limites du modèle urbain et institutionnel de Salmon apparaissent ainsi non tant par l’application mathématique de normes que par l’impossibilité d’y intégrer cette synthèse de colonisation urbaine et rurale qui se met en place après la Guerre Sociale par les effets conjoints des guerres civiles et de la municipalisation de l’Italie au Ier s. av. J.-C. Il ne s’agirait donc pas d’une simple œuvre de rationalisation des procédures de fondation de colonies à cette époque comme le pense Daniel Gargola,17 car cette œuvre ne se manifestait pas par des projets de rationalisation à l’échelle de l’Italie et par rapport à l’envergure de l’entreprise de chacun des imperatores, à savoir, la limitatio des terres redevenues publiques suite aux confiscations, la fondation de colonies avec deductio et l’assignation aux vétérans, la réalisation des cadastres et, enfin, leur enregistrement, éventuellement sur des formae des colonies.18 Les assignations coloniales aux vétérans, sans abandonner complètement la fondation de colonies avec deductio, s’alignent davantage depuis Sylla sur le modèle alternatif des assignations viritanes connues par la colonisation rurale répandue en Italie

15

16

17

18



La première lex colonica – Lex Atinia – date de 197 av. J.-C. et donne une nouvelle fonction de la loi en effaçant le rôle du sénatus-consulte en matière de deductio coloniale (Hermon 2001, 214 sqq.). Voir à propos de la genèse de l’habitat de la cité de Suessa, processus accompli dans le laps de temps entre la lex Flaminia de agris dandis dividendis et le Ier s. av. J.-C. (Giorgi 2010, 57). Nous apprenons qu’un tel projet avec la nomination de 200 finitores a été conçu, selon les dires de Cicéron par Rullus (Cic., Leg. agr. 2.13-32; Gargola 1995, 18689) mais il reste un projet en large mesure démagogique. Les formae, connues depuis le IIe s. av. J.-C., sont complémentaires aux opérations de centuriation (ex. la

forma de l’ager Campanus, établie un siècle auparavant, fut détruite par Sylla, Moatti 1993, 84-89). Elles donnent les mêmes indications que la lex colonica. Les formae se présentent cependant comme un document juridique et administratif définitif avec des indications de valeur géographique et paysagère (fleuves, montagnes aqueducs, cadastres) (id., 45.61.69). En général et surtout à la fin de la République, on ne peut pas présumer la constitution des formae après chaque division agraire, ni une planification des grilles cadastrales uniformes, ce qui aurait exigé une période de préparation plus longue que celle concédée par les colonisations répétées sur les mêmes sites dans la tourmente des guerres civiles (voir à ce propos Muzzioli 2001, 7-20).


centre-méridionale depuis le IIIe s. av. J.-C. Un nouveau fait urbain, nous semble-t-il, voit le jour au Ier s. av. J.-C., et il doit tenir compte des formes coloniales et viritanes des distributions de terres, des colonisations hâtives, répétées sur le même site et souvent éphémères, ce qui engendre une complication pour la légalité des opérations et l’existence de communautés distinctes qui se partagent le même territoire. En réalité, cette ‘rationalisation’ par l’utilisation sélective des normes vide de sens la procédure elle-même de la deductio en mettant en cause le modèle institutionnel de colonisation romaine avancé par Salmon. En fait, le dernier siècle républicain est caractérisé par la coïncidence des deux formes politiques en Italie: la colonisation militaire et la municipalisation et une série de lois et de mesures à connotation agraire plutôt que définies comme des lois agraires. La colonisation romaine le Ier s. av. J.-C.: un lieu commun des faits politiques et administratifs simultanés  Il convient ainsi de ne pas dissocier le processus de municipalisation de l’Italie mis en place à la fin de la Guerre Sociale de cette colonisation militaire.19 Depuis Sylla, la création de nouveaux municipes par le synœcisme des agglomérations rurales (vici, fora, conciliabula) peut souvent s’accompagner de l’installation de communautés de vétérans dans l’une de ces communautés rurales devenues ainsi des chefs-lieux. Ce processus est bien décrit dans les textes gromatiques. Siculus Flaccus en évoque ainsi la pertica:20 id est omnium territorium, coloniae eius in qua colonis deducti sunt. Lors de la dictature de César et en jugeant d’après les documents officiels de l’époque, le processus de municipalisation est considéré comme étant achevé, car le terme de municipium remplace celui de la colonie dans la hiérarchie des communautés administratives connue depuis la Guerre Sociale.21 Dans la pratique, la municipalisation se poursuit jusqu’à Auguste. Les nouvelles communautés citadines de vétérans s’intègrent ainsi, avec la municipalisation, dans un processus de transformation de colonies en municipes22 et de municipes, avec le synœcisme des communautés rurales, en sièges de préfecture de iure dicundo. La législation agraire pouvait-elle servir toujours de cadre pour cette synthèse de la dualité initiale instaurée depuis les Gracques? Lois et mesures à connotation de législation agraire Peu nombreuses sont les lois agraires qui peuvent assurer durant cette période le suivi de la procédure institutionnelle de la lex agraria. Ce qui caractérise la période c’est notamment le manque de programmation et la nécessité de reconstituer des politiques agraires et de colonisation à partir d’une série de lois et de mesures mises sur pied par Sylla, César et les Triumvirs, processus qui trouva sa confirmation dans la création de loismodèles des arpenteurs qui signalent en fait les cycles agraires républicains: lex Gracchana, lex Cornelia, lex Iulia, lex triumvirale. À titre d’exemple, la lex Cornelia agraria du Liber coloniarum I comporte plusieurs phases

19

20

Humbert 1978; sur les rapports entre ces deux phénomènes: Laffi 1973, 37-53 et maintenant dans Laffi 2007; Coarelli 1992, 21-30; et sur les effets de la municipalisation dans le Samnium: D’Henry 1992, 12-19. Sic. Flacc., Th. 128 (trad. Clavel-Leveque et al. 1993, 293): “Mais voici ce qu’on a trouvé: plusieurs municipes se sont vu donner des frontières dans des conditions suivantes. Comme des peuples avaient été expulsés et que des colonies avaient été déduites dans une seule cité parmi d’autres, on a enlevé à beaucoup, nous l’avons rappelé plus haut et souvent, des territoires, et les terres de plusieurs municipes furent divisées, ce qui fut compris dans une seule limitation; c’est devenu la pertica complète, c’est-à-dire englobant tous les territoires de la co-

21

22

lonie dans laquelle des colons ont été déduits.” Tarpin 2001, 72-80. Cela se traduit par une phase historique contemporaine à César où le duumvirat s’impose sur le quattuorvirat dans les constitutions des municipes (Hermon 2008a). L’exemple d’Aesernia dans le Samnium interne est révélateur de ce processus: colonie latine au IIIe s. av. J.-C., elle a intégré une communauté locale des Samnites libres (incolae). Municipe après la Guerre Sociale, elle aurait subi des confiscations et des distributions de terres à l’époque de Sylla. Devenue colonie, elle se reconvertira rapidement en municipe (D’Henry 1992, 13-15); sur les incolae d’Aesernia, (Hermon 2007, 29; v. Tarpin dans ce volume).




législatives qui concèdent à Sylla, dictator legibus scribundis, le droit de légiférer sur des leges datae municipales et de déléguer des pouvoirs pour l’assignation de terres en intégrant ainsi dans un même processus la colonisation militaire et l’organisation municipale. Par conséquent, elle justifie sa définition de loi-modèle des arpenteurs par le besoin de classement de diverses données concrètes en rapport avec l’intervention agraire syllanienne.23 En effet, le manque d’implication d’autres instances législatives et institutionnelles dans ce procès rend désuet l’aspect institutionnel de la deductio et l’efficacité de la lex colonica pour cette colonisation militaire.24 Paradoxalement, il convient donc de concéder aux lois-modèles des Libri Coloniarum la fonction de référence pour tenter de reconstituer les mesures agraires attribuées aux grands protagonistes de la fin de la République et de discerner ainsi les formes de colonisation qui peuvent leur être vraisemblablement attribuées. Si rationalisation y en a, elle se fait sur le terrain et non pas dans l’esprit des imperatores. Les Libri sont en fait la source la plus complète de cette colonisation par la place octroyée à chacun des protagonistes des guerres civiles et des auteurs des cycles agraires qu’il convient encore d’envisager globalement,25 et cela malgré la nécessité d’une confrontation constante avec des recherches sur le terrain des sites spécifiques. En tout cas, quelle que soit la crédibilité de ces informations, ce document d’archives gromatiques témoigne de l’importance de la législation agraire comme seul cadre normatif de la colonisation. Il est établi en fonction des traces sur le terrain alors qu’il est difficile de le confirmer par les procédures institutionnelles.

Les Libri coloniarum: une source à reconsidérer Les Libri coloniarum, source technique d’interprétation difficile, offrent ainsi la dimension diachronique de la colonisation républicaine. Cette liste sélective des colonies romaines en Italie porte la mémoire des interventions agraires depuis les Gracques suivant le langage juridico-technique des arpenteurs.26 Les difficultés d’interprétation de la terminologie des Libri, identifiées depuis longtemps,27 incitent à la prudence quant aux fondations coloniales lorsque les données ne sont pas corroborées par des informations provenant d’autres sources. Ces notices signalent en effet des distributions de terres28 attribuées aux grands intervenants républicains et sanctionnées par une législation agraire. Cependant, les Libri coloniarum ont enregistré à l’usage des arpenteurs et d’une façon sélective les interventions agraires.29 Néanmoins, la prise en compte

23

24 25

26

27



À l’exemple de la Lex Sempronia et de la Iulia, Hygin l’Arpenteur, Constitutio Limitum, Clavel-Lévêque et al. 1996, 14-15; Hermon 2008a. Ead. 2006a; ead. 2006a. Les Gracques: Hermon 1976a; ead. 1982; Peyras 2006, 47-63; Lex Cornelia agraria: Hermon 2006a, 31-45; Gonzales 2006, 18-19 dresse des listes des assignations et les colonies attribuées aux divers intervenants et infra nn. 25, 30, sur l’importance de ces repères chronologiques. Éditions récentes: Campbell 2000; Del Lungo 2004; Besançon 2009 se sont ajoutées à l’édition canonique de Lachmann et al. 1848. Les notices sont hétérogènes quant au contenu et la syntaxe, et les opinions restent partagées s’il s’agit de l’œuvre de plusieurs compilateurs ou d’un projet unique qui est tout de même le résultat de traditions différentes collationnées dans les listes des colonies. Pour une œuvre de compilation homogène: Grelle

28

29

1992, 78-80; pour la bibliographie: Gonzales 2006, 16; Hermon 2008b. Il s’agit d’opérations agrimensorales reconnues sur le terrain et enregistrées dans les divers documents écrits connus durant la République (ex. tabulae censoria: Cic., Leg. agr. 1.2.4; monumenta vetera, litterae, et senatus consulta (Cic., Leg. agr. 2.31; Cic., Leg. agr. 2.88); libri beneficiorum ex.: les terres du temple de Diane à Tifata par Sylla, Moatti 1993, 56 et 69) pour être inscrites sur des formae qui ont généré les listes de colonies. D’autres opérations de cadastration peuvent être imputées au processus de municipalisation qui suit la Guerre Sociale (Gabba 1989). Sur la pratique agrimensorale de la préparation des deux plans (formae) des terres centuriées: l’une pour tabularium, l’autre pour les archives locales, et quidquid aliud ad instrumentum mensorum pertinebit, non solum colonia sed et tabularium caesaris manu conditoris subscriptum habere debebit (fig. 135, 135a).


de ce témoignage reste indispensable pour toute recherche sur la colonisation romaine républicaine et l’installation matérielle des communautés de vétérans. Le problème de l’arrimage de ce moment majeur de réorganisation administrative de l’Italie par la municipalisation, les données des Libri sur les interventions agraires et la colonisation militaire ainsi que des traces de cadastration sur le terrain, est toujours délicat. C’est ainsi que des cadastres reconnaissables sur le terrain des municipes ne sont pas toujours signalés par les Libri;30 ces municipes peuvent bénéficier d’ailleurs d’un statut de chefs-lieux des communautés rurales31 et des colonies peuvent se transformer en municipes.32 Il s’agit cependant de processus conjoints aboutissant à l’installation hâtive des vétérans avec la création de communautés distinctes sur les mêmes sites qui ne s’accompagnent pas toujours du modèle institutionnel de la colonisation romaine par la deductio. Il nous semble également, malgré les limites inhérentes du modèle institutionnel de Salmon pour expliquer les formes coloniales de cette époque, qu’il ne faudrait pas mésestimer importance de ce modèle comme repère pour la reconstitution historique en manque chronique d’information. Un tel repère est sans doute le caractère programmatique de “la loi” alors que des incertitudes subsistent sur le nombre et les modalités concrètes de la fondation des colonies.33 Les actes institutionnels comme la deductio et les statuts des cités ont des connotations précises dans les sources épigraphiques ou littéraires, et leur utilisation par les arpenteurs reste à discerner dans le cadre d’un mouvement de colonisation au Ier s. av. J.-C. réputé massif et multiforme. Les chiffres avancés par les sources littéraires sont imposants,34 ce qui justifie la recherche des traces de cette colonisation dans l’amalgame des données disparates et décalées chronologiquement contenues dans les notices des Libri Coloniarum. En l’absence de tels repères les Gromatici sentent le besoin de reconstituer les lois-modèles en les investissant de la synthèse de la dualité initiale entre colonisation urbaine et rurale connue depuis les lois agraires des Gracques. Y-a-t- il une nouvelle vision de l’espace colonial? Faute de pouvoir partager chronologiquement ces informations, je me suis demandée si l’agencement du lexique dans la structure elle-même des notices ne pouvait pas fournir des indices sur l’évolution du vocabulaire de la cité à l’instar de documents épigraphiques officiels de cette époque. En fait, ces notices suivent généralement la structure binaire (urbs-ager) du concept de la cité et du modèle classique de la colonisation romaine qui s’exprime par le statut de la cité et la morphologie de son ager. Cependant, le vocabulaire de la cité pour une centaine de notices des Libri avec une répartition régionale assez représentative suggère l’éclatement de ce modèle dans ses rapports d’inclusion (intra muros) de l’espace civique et d’exclusion (extra muros) des agglomérations secondaires sur l’ager, lorsque une servitude de passage iter populo debetur/non debetur s’interpose entre le vocabulaire de la cité et la morphologie de son ager.35

30

31

32

Hygin l’Arpenteur, Constitutio limitum, Clavel-Lévêque et al. 1996, 144. Il s’agit d’une pratique importante depuis la construction du Tabularium à Rome par Sylla (Moatti 1993, 70; Hermon 2006b). À titre d’exemple, Cubulteria sur la vallée du Moyen Volturne entre la Campanie et le Samnium serait un municipe après la Guerre Sociale et atteste des traces de cadastration sur le terrain sans bénéficier d’une notice dans le Liber (Cera 2004, 27 et bibliographie). Dans la mesure où Cadatia du Liber désigne la Caiatia samnite, il s’agirait d’un oppidum chef lieu d’une communauté rurale, tout comme Calatia adjugé par Sylla à l’ager Campanus (Renda 2004, 421 sqq.). Supra, note 14 et infra, note 37.

33 34

35

Hermon 2001, 217. Entre 80,000 à 120,000 hommes pour Sylla et lors de la lex Iulia agraria de 59 av. J.-C., 40,000 hommes, vétérans et prolétaires (Brunt 1971, 305, 312-15). Quatre situations ont été distinguées pour les notices qui contiennent la servitude d’iter des deux Libri avec leur répartition régionale: 97 notices dont 33 avec la servitude d’iter populo debetur et 64 non debetur; 49 notices considérées comme homogènes et qui sont exclusives aux Civitates Campaniae contiennent en plus de la servitude d’iter un lexique de la fondation pour chacune de ces formes: 20 avec la servitude d’iter debetur et 29 avec la servitude d’iter non debetur (tab. I) (Hermon 2006c; ead. 2008a).




TAB. 1. La servitude de passage de l’iter dans les Libri coloniarum (statistiques générales).

J’ai ainsi identifié dans les notices des Libri une structure tripartite homogène (urbs-iter-ager). Sa mention est systématique pour un nombre de notices des cités des régions suivantes: Apulia, Calabria, Valeria, Ve région et Septempeda (Liber I), Picenum, Samnium, Apulia (Liber II). De plus, les Civitates Campaniae du Liber I contiennent les notices avec la servitude d’iter, doublées d’un vocabulaire de la “fortification” qui apparaît comme une caractéristique de la lex Cornelia agraria. Ce type de recherche statistique n’est pas nouveau, car plusieurs tentatives de surmonter ainsi les difficultés factuelles dans l’interprétation des notices des Libri ont produit des indices sur la répartition géographique des interventions agraires et sur la morphologie agraire, mais nous y recherchons la cohérence interne du vocabulaire de la cité sans envisager séparément les différentes catégories d’information des notices elles-mêmes.36 Cet éclatement de la structure classique du vocabulaire de la cité est suggéré par l’interposition d’une servitude de l’iter entre urbs et ager (tab. 1).

Les Civitates Campaniae et les Libri Coloniarum entre Sylla et les triumvirs37 Le fait urbain entre colonisation et municipalisation Cette région, située entre la Campanie et le Samnium actuel, est la plus touchée par la colonisation romaine à répétition réalisée dans la période entre les Gracques et la fin de la République en passant par Sylla, César et les Triumvirs (fig. 1). Depuis Sylla, les sources littéraires font état d’impératifs politiques, de destructions

36



À titre d’exemple, la structure des notices est utilisée par (Compatangelo-Soussignan 1991, 143-46) pour une analyse factorielle de 24 centres des Civitates Campaniae afin de dégager des explications d’ordre historique sur la répartition géographique des six rubriques identifiées.

37

Les divisions administratives du Liber coloniarum I du temps de Constantin ont compris dans les Civitates Campaniae la Campanie, le Samnium et des parties du Latium. Le Samnium fut organisé comme province indépendante entre 336-46 av. J.-C. (Gonzales 2006, 15).


massives, de confiscations et de prolifération de colonies militaires pour les milliers de vétérans des guerres civiles, dont peu ont pu être localisées.38 Tout comme l’installation de vétérans par Sylla, réputée massive, l’intervention de César lui-même a laissé peu de traces sur le paysage italien, compte tenu de la difficulté de l’identifier par rapport à l’œuvre triumvirale qui en a pris la relève sur un laps de temps plus long. Une nette discordance démographique entre le nombre respectif de colons établis en Italie qui fut avancé, d’une part, par les sources littéraires, et les quelques mentions fortuites des colonies, d’autre part, laisse le chercheur toujours perplexe39 et l’incite à explorer des solutions alternatives. Pour les Civitates Campaniae, le vocabulaire de la cité se dédouble avec un vocabulaire de la “fortification”. Malgré les réserves sur la valeur historique des notices des Civitates Campaniae dans le premier Liber,40 elles apparaissent néanmoins comme l’archétype de l’évolution du concept de la cité, par l’éclatement du binôme urbs-ager avec une FIG. 1. La Campanie (Barrington Atlas 2000, 44). cohérence qui leur est propre en désignant d’emblée la particularité des Civitates Campaniae comme laboratoire d’étude.41 Un premier échantillon de 10 notices du Liber coloniarum I sur la colonisation de Sylla, confirmée par d’autres sources, nous a révélé une grande cohérence dans leur structure. Bien que la typologie de la fortification ait joui de traductions littérales, les nuances entre ces différentes formes de la fondation/fortification sont subtiles lorsqu’on examine l’agencement du vocabulaire de la notice. L’utilisation de la typologie est la suivante: de muro ducta/um se rattache aux assignations, la seconde est munitum s’applique exclusivement aux terres laissées à l’occupation sans mentionner aucune morphologie de l’ager (Tab. 3).42 L’examen de cet échantillon m’a conduite aux constatations suivantes: I) le dédoublement du vocabulaire de la cité lorsque l’“acte fondateur” est accompagné de “l’acte de la fondation”, soit celui de la fortification avec l’installation des communautés de vétérans;  II) l’adéquation de “l’acte de la fondation/fortification” avec les opérations qui se font sur l’ager (muro ductum: assignatio; est munitum/circum ductum: occupatio);

38

39 40

41

Sur les listes des colonies certaines, Brunt 1971, 30011; Gabba 1973, 172 sq.; et pour l’époque triumvirale, Keppie 1983, 16. Supra n. 34. Malgré la transmission de la tradition manuscrite qui crée des disparités dans les notices de diverses régions, notamment pour les Civitaes Campaniae (Grelle 1992, 67-85), les servitudes d’iter sont signalées également pour d’autres régions (Tab. 1). En revanche, seules les Civitates Campaniae présentent les deux formes. Toutes les éditions récentes, supra, n. 17, reproduisent

42

intégralement pour cette région les notices des Civitates Campaniae de la première édition de Lachman de 1848. Cette dernière typologie de l’occupatio pourrait bien s’inscrire dans la situation où les colons syllaniens ont obtenu paludes et silvas: App., B Civ. 1.100, Cic., Leg. agr. 2.71; Brunt 1971, 311, n. 4; la typologie de circum ducta/um désignerait un champ militaire retranché, des vraies castra des vétérans (Hermon 2006a, 31-41). Sur sa traduction littérale équivalente à la deductio, ead. sous presse, 58 et bibliographie.




TAB. 2 Lex Cornelia agraria dans le Liber coloniarum I (Hermon 2006a, 45).

III) l’articulation de la servitude d’iter avec le vocabulaire de la fondation/fortification et les opérations sur l’ager (iter populo debetur: muro ductum/assignatio; iter populo non debetur: est munitum/ circum ductum/ occupatio). La lex Cornelia agraria de 59 av. J.-C. Le trinôme urbs (res publica) – iter (populo debetur/non debetur) – ager suggère également quelques repères pour la colonisation de César43 en vertu de la lex Iulia agraria de 59 dans  le  Samnium44 et en continuité avec l’intervention césarienne sur l’ager Campanus (tab. 2).

43



La province initiale du consul César en 59 av. J.-C. comprenait la maîtrise de selvae callesque, Suet., Iul. 19.2 est tout à fait adaptée à la topographie du Samnium. Bovianum oppidum (Lib. colon. I L.231.8 et II L. 233.14) et Aufidena (Lib. colon. II 259.17) seraient des municipes ex novo avec une constitution duumvirale (Laffi 1973, 48, n. 55; id. 47, n. 52). Dans le Samnium interne la colonie Aesernia (Lib. colon. I.23.14) aurait vite récupéré son statut municipal à l’époque d’Auguste, voir également, supra, n. 22 et Hermon 2010c, 9; Bovianum subit un processus inverse par la transformation du municipe césarien en colonie à l’époque

44

triumvirale (Bovianum Vetus), De Benedettis 2004, 28. Nous limitons ici l’enquête sur la lex Iulia agraria de 59 aux Civitates Campaniae du Samnium interne. Sur cette loi Crawford 1989, 179-90; et sur l’importance de l’intervention de César sur l’ager Campanus, Oliviero 2002, 51-56. Sur l’intervention césarienne à Capoue et à Calatia ainsi que sur l’importance de cette loi pour la municipalisation de l’Italie Hermon 2008a et supra. Sur l’aménagement du territoire de Capoue: Franciosi 2002, 19-23 et sur l’aménagement de la via Appia avec des murs de soutien à l’occasion de la déduction de la colonie césarienne, Quilici Gigli 2002, 98.


TAB. 3. Les opérations agraires entre Sylla et César (Hermon 2008a).




Malgré la même structure tripartite, ces notices ne présentent pas la même homogénéité par: I) le dédoublement du vocabulaire de la cité avec une typologie de la fortification45 alors qu’aucune typologie de est munitum / circum ductum n’est associée à une occupation militaire du site par l’installation des unités homogènes de vétérans; II) par le rattachement, dans le cas d’Aesernia, de la terminologie de muro ductum au statut municipal (oppidum) alors que la deductio désigne le statut colonial;46 III) par l’imposition d’une servitude d’iter debetur identique sans rapport avec le statut de la cité ou les opérations sur l’ager (tab. 3). Ces constations renforcent l’hypothèse que la structure identifiée comme homogène du nouveau vocabulaire de la cité, dédoublée par le vocabulaire de la ‘fortification/fondation’, est une caractéristique de la lex Cornelia agraria. Attardons-nous à cette servitude d’iter. La légalité de ces nouvelles res publicae sur l’ager des cités préexistantes semble être régularisée par des servitudes de passage. Cette servitude est connue depuis les XII Tables. Elle s’applique aux voies publiques et aux limites des terres centuriées qui ont le même statut de voie publique et remplissent les mêmes fonctions. Leurs dimensions sont connues et dispensent normalement de toute mention de la servitude d’iter dans les deux cas de figure qui apparaissent dans les Libri.47 Résoudre le mystère de leur mention dans les Libri coloniarum a ainsi fait couler beaucoup d’encre,48 car il faut imaginer des situations exceptionnelles qui justifient l’imposition de la servitude aussi bien que son exemption. En effet, les dimensions inhabituelles de la servitude d’iter dans la structure homogène des notices du Liber coloniarum I pour l’ager Campanus, qui n’ont pas d’équivalent dans le système cadastral ou de voirie, m’ont incitée à envisager une explication circonstanciée aux guerres civiles lorsque la séparation des communautés rivales se partageant le même territoire pouvait se matérialiser par une servitude substantielle de l’iter sur les voies vicinales.49 Cette éventualité est d’autant plus vraisemblable si les territoires ayant subi la limitatio dans la tourmente des guerres civiles sont restés dans la situation juridique de la possessio par occupation, n’ayant pas encore été inscrits dans une forma coloniale. Frontin est clair sur la nécessité d’un droit de passage sur un territoire de ce statut.50 D’une

45

46



Seule Aesernia dans le Liber coloniarum II (Civitates regionis Samnii) présente la typologie de muro ductum associée à son statut municipal d’oppidum, L 260, 7, qui serait postérieur à celui de colonie mentionné dans le Liber I, La 233.14 (Civitates Campaniae). Aufidena, Aesernia et Bovianum, situées sur les voies de passage et de transhumance entre la Campanie et le Samnium, s’intègreraient directement dans le processus de municipalisation. La terminologie lege Iulia sine colonis deduxerunt des municipes de Bovianum et Aesernia mérite un examen conjoint avec les autres notices des Libri qui portent cette mention. Nous remarquons cependant qu’elle remplace dans les notices attribuées à César celle de la fondation/fortification ayant probablement la même fonction de dédoublement du vocabulaire de la cité dans un processus intense de municipalisation alors que la deductio suggère le statut de colonie. Seul l’examen de l’histoire de ces communautés pourrait

47

48

49

50

donner quelques éléments de réponse. De Nardis 2002, 117, sur les dimensions des limites dans les divers textes gromatiques, qui ne devraient pas dépasser 40/30 pieds pour le decumanus maximus, 30/20 pieds pour le cardo maximus, 15 /12 pieds pour les axes intermédiaires et 8 pieds pour les subruncivi. En dernier, De Nardis 2002; Hermon 2006a, voir également Annexe 3, “iter populo debetur ped(um) (tot) et iter populo non debetur” dans les “libri coloniarum”, Brunet et al. 2008, 57-61. Hermon 2006 par une mise en contexte d’une intuition de Voight; dans les cités de l’ager Campanus: des dimensions de 120 ped., 100 ped., et 80 ped., de la servitude d’iter auraient grevé des voies vicinales comme des zones tampons entre communautés rivales. Voir à ce propos l’analyse de Peyras 2006, 62; Behrends et al. 1998, 69, 34-35.


façon plus générale et en tenant compte d’un contexte climatique relativement plus humide à cette époque,51 il est légitime de se demander si les dimensions démesurées de l’iter debetur ne pouvaient pas correspondre, pour certains sites, à une mesure de protection face aux effets néfastes des débordements des fleuves et de leur affluents ou de la proximité de la mer. Ainsi, le rapprochement fait par Mauro De Nardis entre cette servitude d’iter et les subsecivi et le modus exceptus le long du lit des fleuves peut paraître tout simplement comme un modus flumini.52 La servitude d’iter debetur/non debetur apparaît ainsi comme une mesure d’exception dans les circonstances où l’inscription des nouvelles distributions de terres n’ont pas encore subi l’ensemble des opérations cadastrales, notamment le transfert des droits de propriété. Il s’agirait dans ce cas d’une sorte de légitimité pour les assignations de terres accompagnées de la servitude d’iter sur un territoire appartenant au domaine public, mais resté en principe et en fonction de ses dimensions ouvert à l’occupatio. En d’autres termes, il s’agit d’envisager si l’installation hâtive et à brève échéance des communautés mixtes sur le même territoire ne répondait-elle également, à part les impératifs des distributions hâtives de terres depuis le Ier av. J.-C., à la nécessité d’une maîtrise des bassins versants de grandes artères fluviales avec l’occupation de terres à risque par la progression de l’habitat vers les plaines fluviales et les points de passage névralgiques. Les formae des colonies et les espaces ripariens La représentation graphique des formae coloniales des vignettes des Gromatici désigne d’emblée la colonisation romaine comme un effet structurant des “espaces ripariens” par l’intégration de communautés de statuts divers dans un écosystème incluant la mer et des montagnes avec une grande diversité des ressources en eau et pouvant offrir de meilleurs conditions pour la gestion des ressources naturelles régularisée par des cadastres (cf. Tarpin dans ce volume).53 À cet égard, la représentation graphique de la forma de Minturnae (Tarpin dans ce volume, fig. 2) vraisemblablement colonie césarienne,54 est très suggestive de la vision écosystémique d’un espace riparien. C’est ainsi que le mons Vescinus, un élément paysager de la vignette a été identifié au pagus civium Romanorum du même nom signalé par une inscription récemment découverte. Cette inscription a permis de situer le territoire de Vescia, la cité aurunque disparue, à St. Lorenzo, lequel territoire serait l’un des rares qui, encore ager publicus, ait pu être convoité par Rullus en 63 ap. J.-C.55 Les caractéristiques du terrain ont ainsi permis de confirmer la véridicité de cette représentation graphique qui est un modèle “d’espace riparien” colonial ayant pu être fonctionnel dès César tout en étant construit par

51

52

53

54

Sur cette phase climatique plus humide au Ier siècle av. J.-C., Allinne 2008; Bencivenga compte 10 crues catastrophiques du Tibre au Ier siecle av. J.-C., et 9 au siècle suivant, cit., Leveau 2008, 138, sur les inondations du Tibre et le travail d’aménagement des ripae, Muzzioli 2009, 392-93. Suggestion de De Nardis 2002, 120-22. Sur le modus flumini voir Hermon 2009. Dilke 1988, 199. Il faut cependant convenir que les vignettes n’ont pas les mêmes statuts et logique que les formae, mais leur parenté est réelle et peut dériver des maquettes que les arpenteurs ont rédigé avant la constitution de la forma, Moatti 1993, 43. Minturnas. Muro ducta colonia, deducta a Gaio Caesare, ier populo non debetur, ager eius pro parte in iugeribus est adsignatus: ceterum in absoluto est relictum, L 255, 12-14. L’épisode de Marius, poursuivi par les si-

55

carii de Sylla, suggère fortement que des confiscations aient suivi et que des vétérans syllaniens y aient été installés. Cette intervention s’exprimerait dans la notice du Liber par l’acte fondateur représenté par la terminologie de la fortification, muro ducta, Hermon 2006a, p. 35. La notice du Liber I présente une structure parfaitement homogène avec le dédoublement des actes fondateurs liés cette fois-ci à la terminologie de la fortification, et l’acte de fondation se rapportant à la deductio, vraisemblablement de César. Nous avons suivi Keppie 1983, 141 sq., pour l’identification de César comme deductor de la colonie, voir également, Arthur 1991, 41-44 et 57. Si la notice n’est pas précise pour la chronologie, elle suit néanmoins une logique précise dans le dédoublement des actes de la fondation. AE 1989, 150; Coarelli 1989, 32 sq.; sur ager Vescinus, Cic., Leg. agr. 2.66.




étapes avec la diversification de la gestion des ressources en eau. La vignette figurant sur un manuscrit d’Hygin l’Arpenteur du VIe s. ap. J.-C. du traité Constitutio limitum, ainsi que le passage qui l’explique (142Th fig. 89), confirment les données de la notice du Liber. Néanmoins, la conjugaison de ces deux séries d’informations nous laisse entendre que l’incorporation du territoire de ce pagus civium Romanorum, laissé à la libre occupation et qui s’était développé indépendamment de la colonie de Minturnae, a été faite lors de la constitution de la pertica de la colonie sur les deux versants du Liris. La constitution de sa forma serait intervenue probablement à un autre moment que les deux actes fondateurs républicains: muro ducta colonia par Sylla et deductio par César avec des assignations aux vétérans au-delà du Liris. La vignette montre ainsi un cadastre uniforme qui aurait pu être constitué par les avancées successives vers la mer et des assignations le long de la rive droite du Liris, mais également à partir des anciennes assignations sur l’autre versant du fleuve. Il semble ainsi que le territoire du pagus en deçà du Liris ait été assigné lors de la constitution de la forma per professionem en le laissant au mode arcifinal: citra Lirem postea adsignatum per professiones ueterum possessorum, ubi iam opportunarum finium commutatione relictus primae adsignationis terminis more aecifinio possidetur (souligné par l’auteur).56 La notice du Liber (tab. 1.1 et n. 54) reproduit en abrégé les mêmes informations sans précisions d’ordre géographique, mais avec des repères chronologiques pour les marqueurs du terrain. Les deux témoignages aient pu puiser l’information d’une même source: la forma de la colonie de Minturnae. Il s’agirait d’un règlement définitif tel que exigé par une forma coloniale qui montre clairement une vision d’un espace intégré des ressources naturelles dans le cadre de la colonie, ce qui s’apparente à certains égards à la notion écosystémique moderne. Il aurait tranché également les cas de changement des limites en faveur des premières divisions de terres laissées également au mode arcifinal. Cette colonisation militaire, aurait-elle contribué en fin de compte à la construction des riparia, que nous avons défini comme des milieux caractérisés par différents éléments paysagers en interaction avec une diversité des corps d’eau eau, dont les dimensions écosystémiques s’établissent en fonction de perceptions et de représentations sociales?57 Le concept de riparia, par sa vision écosystémique n’est pas identique aux autres concepts d’espace – le paysage et les frontières – et les vignettes des manuscrits des Gromatici représentent la colonisation romaine comme un élément structurant. Je me réfère ici aux travaux du dernier colloque que j’ai organisé en 2009 à l’Université Laval dans le cadre de ma chaire de recherche du Canada en interactions sociétéenvironnement naturel dans l’Empire romain (2003-2009) où j’ai présenté l’évolution sémantique du concept à l’époque romaine, à partir de l’agronome Caton, de la vision politique de Cicéron, du Corpus agrimensorum Romanorum et des chartes alto-médiévales. Il s’agit d’une approche globale des rives des fleuves depuis les pieds de montagnes aux rivages de la mer, des lacs et des marais avec la vision écosystémique des différentes communautés biotiques du milieu comme ressource naturelle à exploiter (ex. riparia hirundo, rivalis alecula). Cette reconstitution à travers 8 siècles d’histoire a permis de constater l’affirmation de la configuration spatiale des riparia dès le Ier s. ap. J.-C. avec Varron et l’évolution d’un complexe notionnel durant l’Antiquité tardive et le Haut-Moyen Âge qui a établit la dynamique des interactions société-environnement propre à ce type d’écosystème. Les différents usages économiques lui confèrent un statut à la fois politique, fiscal et juridique; les dimensions socio-religieuses ajoutent une perception propre de cet espace. En fait, nous pouvons distinguer les trois dimensions de cet espace – connu, construit, perçu –, tandis que l’étendue de la dynamique société-

56



Encore, comme en Campanie, sur le territoire de Minturnes: la nouvelle assignation, au-delà du Liris, est contenue par des limites; en deçà du Liris, on a assigné par la suite d’après les déclarations des anciens possesseurs, là où après un changement des fines, on a laissé celles de la première assignation et on a commencé à posséder désormais à la manière arcifinale, trad. Clavel-Lévêque et al. 1996, 51.

57

Paysage culturel et espace socio-économique façonné par les interactions société-environnement qui permettent, entre autres, d’évaluer le poids relatif du facteur climatique dans son évolution, Hermon 2010a, 8; ead. 2010b, 338.


environnement naturel tient de sa perception comme un écosystème propre. Le VIe s. ap. J.-C., ce moment-clé de la fixation des traditions, entre autres la codification du droit et l’apparition des vignettes sur les manuscrits du Corpus Agrimensorum Romanorum a été identifié comme le point de convergence qui réunit les tendances antérieures des représentations sociales de cet espace comme un écosystème dont celle examinée pour Minturnae.

Conclusion: Colonisation militaire au Ier siècle a.v. J.-C. et la construction d’espaces ‘ripariens’: une réponse sociétale à une phase climatique plus humide? La vignette de Minturnae, datant du VIe s. J.-C., est ainsi la preuve de cette perception car elle reproduit la synthèse d’une évolution historique d’un espace riparien qui se construit progressivement durant la République depuis les montagnes jusqu’à la mer pour des cités littorales et qui est connu par les formae coloniales, plus tardives. L’échelle spatiale des riparia est représentée à partir de la configuration spatiale connue en fonction du système naturel qui identifie la côte, la mer, les cours d’eau, des lacs ainsi que les divers éléments morphologiques de cette zone transitoire entre la terre et l’eau à courte, moyenne et longue distance selon l’échelle d’observation. En effet, les représentations graphiques des manuscrits médiévaux des traités d’agrimensure ont pour but de situer la colonie au cœur de la construction du système riparien avec les éléments du paysage, mais sa perception sociale peut décomposer cette unité écosystémique en fonction des critères politiques comme les frontières entre les diverses agglomérations, mais également naturels comme les variations et changements climatiques. En revanche, et compte tenu du contexte climatique plus humide et une phase érosive plus importante qui nous a concédé une nouvelle interprétation à la servitude d’iter debetur/non debetur, il est tentant de supposer que l’Italie centre-méridionale ait pu absorber, et au détriment des provinces, la plus grande vague de colonisation militaire du Ier s. av. J.-C. à la faveur de cette phase climatique plus humide.58 En effet, on a constaté à cette époque une phase alluvionnaire importante dans une zone volcanique et sismique entre la Campanie et le Samnium méridional, du littoral aux zones collinaires intérieures, abritant les torrents affluents du fleuve Volturne en lien étroit avec le réseau de voirie – les ramifications de la Via Appia et de la Via Latina – et le bassin versant du Volturne qui assure la communication.59 Addendum: J’ai repris récemment60 les hypothèses développées dans cet article à propos de la formule juridico technique d’iter populo debetur/non debetur en l’envisageant comme droit de passage et non pas comme servitude d’iter. Compte tenu de variations climatiques caractérisées par des inondations abondantes et la réalité historique d’un mouvement massif de colonisation militaire au Ier av. J.-C. en Italie centre-méridionale, j’ai proposé une explication convergente aux deux cas de figure antagonistes: iter populo debetur correspondant aux bords du lit d’inondation et iter poplulo non debetur concordant avec les rives du lit

58

59

Les principales modifications de la plaine du Volturne et le long du littoral ont été établies à partir de recherches géomorphologiques sur ces sites menées par F. Ortolani et S. Pagliuca. Franco Ortolani m’a fait l’amitié de me communiquer une brève synthèse de leurs travaux sur ces questions, voir Ortolani et al. 2002, 55-100. Dans la région d’Allifae, Casilinum, et Urbana, Caiazza 2002, 17 sqq.; id. 1997, 67 sqq. Cet auteur mentionne des vestiges des ponts antiques à l’appui de son allé-

60

gation sur l’augmentation du niveau des sols avec l’éventuel changement du lit du cours d’eau, phénomènes attribués à cette époque. Cette recherche a également permis à cet auteur de localiser à Torre degli Schiavi, connue par des documents d’archives, la colonie Urbana déduite à Capoue par Sylla et dont les traces ont complètement disparues. Sur des recherches récentes, voir également Quilici et Quilici Gigli 2010 et bibliographie. Hermon 2013.




ordinaire des cours d’eau. Elle est compatible avec la condition publique des rives des fleuves publics et l’occupation progressive des bords de l’eau. Cette interprétation peut être confrontée avec l'étude de cas en Campanie et le Samnium méridional.

Bibliographie Allinne, C. 2008. “L’évolution du climat a l’époque Romaine en Méditerranée occidentale. Aperçu historiographique et nouvelles approches,” in E. Hermon (éd.), Vers une gestion intégrée de l’eau dans l’Empire romain (Rome) 79-97. Arthur, P. 1991. Romans in Northern Campania. Settlement and land-use around the Massico and the Garigliano bassin (Rome). Behrends, O. et al. 1998. Œuvre Gromatique, Frontin. Corpus Agrimensorum Romanorum IV (Luxembourg). Benier, R. et R. Hopital 1983. “Les communautés rurales dans l’empire romain,” in J. Bodin (éd.), Les communautés rurales (Paris) 430-50. Brunet, C. et al. 2008. Libri Coloniarum. Corpus Agrimensorum Romanorum VII (Besançon). Brunt, P. 1971. Italian Manpower, 225 B.C.-A.D. 14 (2nd edn., Oxford). Caiazza, D. 2002. “Modificazione geomambientali nella piana del Volturne dal I sec A.C. e la crescita dei suoli ad Allifae, Casilinum et Urbana,” in D. Caiazza (éd.), I torrenti assassini del Matese meridionale (Alife) 17-32. Caiazza, D. 1997. “Il territorio tra Matese e Volturne. Note di topografia storica,” in D. Caiazza (éd.), Il territorio tra Matese e Volturne (Naples) 67104. Campbell, J. B. 2000. The writings of the Roman land surveyors: introduction, text, translation and commentary (Londres). Cera, G. 2004. “Il territorio di Cubulteria,” in L. Quilici et S. Quilici Gigli (éds.), Carta archeologica e ricerche in Campania, vol. 1: Comuni di Alvignano, Baia e Latina, Caiazzo, Castel Campagnano, Castel di Sasso, Dragoni, Piana di Monte Verna, Ruviano (Rome) 21-238. Clavel-Leveque, M. et al. 1993. Les conditions des terres, Siculus Flaccus. Corpus Agrimensorum Romanorum I (Naples).



Clavel-Leveque, M. et al. 1996. L’établissement des limites, Hygin L’Arpenteur. Corpus Agrimensorum Romanorum IV (Naples). Coarelli, F. 1989. Minturnae (Rome) 29-33. Coarelli, F. 1992. “Colonizzazione e municipalizzazione; tempi e modi,” DialArch 10, 21-30. Compatangelo-Soussignan, R. 1991. “I catasti e strutture agrarie regionali del Sannio,” in La Romanisation du Samnium aux IIe et Ier siècles av. J. C.: actes du colloque organisé par le Centre Jean Bérard (Naples) 139-47. Crawford, M. H. 1989. “Lex Iulia agraria,” Athenaeum 77, 179-90. D’Henry, G. 1992. “La romanizzazione del Sannio nel II e I secolo A.C.,” in La Romanisation du Samnium aux IIe et Ier siècles av. J. C.: actes du colloque organisé par le Centre Jean Bérard (Naples) 1-19. Dall’Aglio, P. L. 2010. “La media valle del Cesano tra vecchie ricerche e nuove prospettive,” in E. Giorgi et G. Lepore (éd.), Archeologia nella valle del Cesano da Suasa a Santa Maria in Portuno (Bologne) 33-38. De Benedettis, G. 2004. “Bovianum, Aesernia, Monte Vaiarno: considerazioni sull’evoluzione dell’insediamento del Sannio Pentro,” in H. Jones (éd.), Samnium. Settlement and cultural change (Providence, RI) 23-33. De Maria, S. 2010. “Suasa, la città e la sua storia. Venti anni di sacvi e ricerche,” in E. Giorgi et G. Lepore (éd.), Archeologia nella valle del Cesano da Suasa a Santa Maria in Portuno (Bologne) 19-32. Del Lungo, S. 2004. La pratica agrimensoria nella tarda antichità e nell’alto medioevo (Spoleto). Dilke, O. 1988. “Roman contribution to cartography,” in M. Sordi (éd.) Geografia e storiografia nel mondo classico (Milan) 194-201.


De Nardis, M. 2002. “Viabilità campana e scritti gromatici,” in G. Franciosi (éd.), Ager Campanus (Naples) 115-23. Franciosi, G. 2002. “I due misteri dell’Ager Campanus” in G. Franciosi (éd.), Ager Campanus (Naples) 19-23. Gabba, E. 1973. Esercito e società nella tarda repubblica romana (Florence). Gabba, E. 1989. “Sui sistemi catastali romani in Italia,” Athenaeum 67, 566-70. Gargola, D. J. 1995. Lands, laws & gods (Chapel Hill). Giorgi, E. 2010. “La città e il territorio: riflessioni sull’origine dell’abitato” E. in Giorgi et G. Lepore (éd.), Archeologia nella valle del Cesano da Suasa a Santa Maria in Portuno (Bologne) 55-61. Gonzales, A. 2006. “Autour d’un palimseste de l’histoire gromatique: les Libri Coloniarum” in A. Gonales et J.-Y. Guillaumin (éds.), Autour des Libri Coloniarum. Colonisation et colonies dans le monde romain (Besançon) 13-22. Grelle, F. 1992. “Strutture e genesi dei Libri Coloniarum,” in O. Behrens et L. Capogrossi Colognesi (éds.), Die römische Feldmesskunst: interdisziplinäre Beiträge zu ihrer Bedeutung für die Zivilizationsgeschichte Roms (Gœttingue) 67-85. Hermon, E. 1972. La loi agraire de Saturninus de 100 av. J.-C. et la colonisation latine de la Narbonnaise, IVRA 23, 67-102. Hermon, E. 1975. “La possible application de la deuxième loi agraire de Saturninus dans la Gaule Narbonnaise,” IVRA 26, 79-98. Hermon, E. 1976a. “La loi agraire de Tiberius Gracchus,” Ktema 1, 179-86. Hermon, E. 1976b. “La date de la fondation de la colonie Narbo Martius en Gaule Narbonnaise,” Revue historique de Droit français et étranger 54, 229-39. Hermon, E. 1982. “Le programme agraire des Gracques,” Athenaeum 70, 258-72. Hermon, E. 1989. “La Lex Flaminia de agro gallico dividundo modèle de romanisation au IIIe siècle av. J.-C.,” in M.-M. Mactoux et É. Geny (éds.), Mélanges Pierre Lévêque (Paris) 274-84. Hermon, E. 1997a. “Les lois Liciniae Sextiae reconsidérées,” Ktèma 19, 119-42.

Hermon, E. 1997b. “M. Curius Dentatus et les ventes quaestoriennes au IIIe siècle av. J.-C.,” Scripta Classica Israelica 16 (2), 32-42. Hermon, E. 1998a. “M. Curius Dentatus - conquête et transhumance en Sabine au IIIe siècle,” CahÉtAnc 34, 55-64. Hermon, E. 1998b. “Les Priscae latinae coloniae et la politique colonisatrice de Rome,” AJAH 14, 143-79. Hermon, E. 2001. Habiter et partager les terres avant les Gracques (Rome). Hermon, E. 2006a. “La lex Cornelia agraria dans le Liber Coloniarum I,” in A. Gonzales et J.-Y. Guillaumin (éds.), Autour des Libri Coloniarum. Colonisation et colonies dans le monde romain (Besançon) 31-43. Hermon, E. 2006b. “L’intervento e la legislazione agraria di Silla in Italia nella tradizione agraria e nella memoria degli archivi,” Index 34, 439-53. Hermon, E. 2006c. “Intra et extramuros: le concept de la cité et de son ager. Pour une relecture des Libri Coloniarum,” Caesarodunum XL, 409-29. Hermon, E. 2007. “Des communautés distinctes sur le même territoire: quelle fut la réalité des incolae?,” in R. Compatangelo-Soussignan et G. Schwentzel (éds.), Étrangers dans la cité romaine, “Habiter une autre patrie”: des incolae de la République aux peuples fédérés du Bas-Empire (Rennes) 25-42. Hermon, E. 2008a. “Colonisation et municipalisation entre Sylla et César dans les Libri Coloniarum,” Ius Antiquum 2-22, 58-75. Hermon, E. 2008b. “Les savoirs traditionnels et la perception de l’eau comme patrimoine naturel et culturel dans Le Corpus Agrimensorum Romanorum,” in E. Hermon (éd.), L’eau comme patrimoine (Québec) 441-60. Hermon, E. 2009. “Le bord de l’eau: un environnement à risque dans les villes et les campagnes de l’Occident romain,” in R. Bedon (éd.), Vicinitas aquae. La vie au bord de l’eau. En Gaule romaine et dans les régions voisines, Caesarodunum (Limoges) 13-29. Hermon, E. 2010a. “Riparia: pour la définition d’un concept,” in E. Hermon (éd.), Riparia dans l’empire romain – pour la définition du concept (Oxford) 3-12. Hermon, E. 2010b. “Sémantique, droit et pratiques




agrimensorales pour la représentation spatiale des riparia,” in E. Hermon (éd.), Riparia dans l’empire romain – pour la définition du concept (Oxford) 231-44. Hermon, E. 2010c. “Conclusion,” in E. Hermon (éd.), Riparia dans l’empire romain – pour la définition du concept (Oxford) 329-38. Hermon, E. 2013. “Iter populo debetur/ non debetur et la gestion d’un environnement à risque,” in Clavel-Lévêque, M., F. Ouachour and I. Pimouguet-Pédarros, Hommes, cultures et paysages de l’Antiquité à la période moderne. Mélanges offertes à Jean Payras (Rennes) 189-206. Humbert, M. 1978. Municipium et civitas sine suffraggio. L’organisation de la conquête jusqu’à la guerre sociale (Rome). Keppie, L. 1983. Colonisation and veteran setlement in Italy, 47-14 B.C. (Rome). Lachmann et al. 1848. Gromatici Veteres, ex recensione Caroli Lachmann, I, Berlioni, 1848 (la); Corpus agrimensorum Romanorum recensuit, C. Thulin, I, 1, Opuscula agrimensorun veterum, Leipzig, 1913 (th). Laffi, U. 1973. “Sull’organizzazione amministrativa dell’Italia dopo la guerra sociale,” in Akten des VI internationalen Kongress für grieschiche und lateinische Epigraphik (Munich) 37-53. Laffi, U. 2007. Colonie e municipi nello stato romano. Storia e letteratura. Raccolta di studi e testi (Rome). Leveau, P. 2008. “Les inondations du Tibre a Rome: politiques publiques et variations climatiques a l’époque romaine,” in E. Hermon (éd.), Vers une gestion intégrée de l’eau dans l’Empire romain (Rome) 137-46. Moatti, C. 1993. Archives et partage de la terre dans le monde romain, IIe siècle av. - Ier siècle ap. J.-C. (Rome). Muzzioli, M. P. 2001. “Sui tempi di insediamenti dei



coloni nel territorio,” ATTA 10, 7-20. Muzzioli, M. P. 2009. “Le piene del Tevere e la sistemazione delle ‘ripae’ a Roma. Il contributo della Pianta di Via Anicia,” in E. Hermon (éd.), Société et climats dans l’Empire romain (Naples) 388-405. Oliviero, G. M. 2002. “La legislazione agraria di Cesare,” in G. Franciosi (éd.), Ager Campanus (Naples) 51-56. Ortolani, F. et al. 2002. “I dissesti catastrofici del Matese merdionale nel quadro delle modificazioni geoambientali avvenute durante il periodo storico in Campania,” in D. Caiazza (éd.), I torrenti assassini del Matese meridionale (Alife) 55-100. Quilici Gigli, S. 2002. “Sulle vie che ricalcano gli antichi assi centuriati,” G. in Franciosi (éd.), Ager Campanus (Naples) 95-114. Quilici, L. et S. Quilici Gigli 2010 (éds.). Carta archeologica e ricerche in Campania, vol. 4: Comuni di Amorosi, Faicchio, Puglianello, San Salvatore Telesino, Telese Terme (Rome). Peyras, J. 2006. “Les libri coloniarum et l’œuvre gracchienne,” in A. Gonzales et J.-Y. Guillaumin (éds.), Autour des Libri coloniarum: colonisation et colonies dans le monde romain (Besançon) 47-63. Renda, G. 2004. “Il territorio di Caiatia,” in L. Quilici et S. Quilici Gigli (éds.), Carta archeologica e ricerche in Campania, vol. 1: Comuni di Alvignano, Baia e Latina, Caiazzo, Castel Campagnano, Castel di Sasso, Dragoni, Piana di Monte Verna, Ruviano (Rome) 239-426. Salmon, E. T. 1969. Roman colonization under the Republic (Londres). Salmon, E. T. 1967. Samnium and the Samnites (Cambridge). Tarpin, M. 2001. Vici et pagi dans l’Occident romain (Rome).


Early colonization in the Pontine region (Central Italy) Peter Attema, Tymon de Haas and Marleen Termeer

Introduction This paper presents a study of changes in the urban and rural settlement of the Pontine region in southern Lazio from the Archaic (6th c. B.C.) to the Mid-Republican period (circa 200 B.C.). Its aim is to increase understanding of the colonization of this key area in the context of the political, economic and territorial expansion of Rome during the 5th and 4th c. B.C.1 Eight colonies were reportedly founded in the Pontine region in this period: Circeii, Cora, Pometia, Norba, Antium, Satricum, Setia and Tarracina (fig. 1), and instances of viritane land distributions are recorded as well. We follow leading ancient historians in presuming that these recorded foundations and distributions should be taken as a reflection of expansionist undertakings of some sort.2 In this paper, we discuss these two forms of colonization, highlighting the various historical, geographical and organizational conditions under which the colonization of the Pontine region occurred. So far, the early colonization of the Pontine region has attracted only scant attention from historians. Nevertheless, rudimentary scenarios about its nature have been formulated. A ‘statist’ scenario, which postulates radical interventions in the colonized areas, planned and organized by the Roman political body, has been suggested for the Regal period.3 For the Early Republic, the importance of elite individuals and their gentes in colonial foundations is broadly stressed.4 The question here is how their activities relate to Rome. While it might be possible to view the gentes as a political power with direct links to Rome, early colonial initiatives in Etruria and Latium Vetus could also have been the enterprises of ‘warlords’ and their followers, acting on their own initiative (cf. Terrenato in this volume).5 So far, these scenarios have not been actively applied to describe and explain the changes in the nature of colonization over time, nor have they been fully examined in conjunction with the archaeological evidence available in any attempt to assess the impact of colonization. Furthermore, archaeological studies indicate that new forms of organization of urban and rural settlement are not always congruent with the recorded

1

2 3 4

Considering the debate on the nature of early colonization, we have refrained from using terms such as ‘Roman’ or ‘Latin’ colonization. For previous archaeological field studies dedicated to the colonization of the Pontine region, see Attema 1993; Attema and Van Leusen 2004; De Haas 2011. See for a concise discussion Termeer 2010, 44-45. Coarelli 1990. See below for further discussion. A recent comprehensive work is Chiabà 2011. This study of the so-called priscae Latinae coloniae includes the Pontine colonies with the exception of

5

Tarracina. For the gentes view also Càssola 1988, 17; Torelli 1999, 18; Coarelli 1990, 152; Bandelli 1995; Chiabà 2006; Bradley 2006. Torelli 1999 is a translation of “Aspetti ideologici della colonizzazione romana più antica,” in DialArch 6 (1988), 65-72; references in this article will be to the 1999 translation. The contrast is not absolute of course cf. Bradley 2006, 168-69. However, Bradley does seem to play down the political power of Rome to a larger extent than do Coarelli 1990 or Torelli 1999.




foundation dates of the various Pontine colonies, and that the transformation of settlement and landscape patterns was a much more gradual process. Therefore this paper presents an evaluation of the above-mentioned scenarios of colonization in the light of archaeological data. To this end, we shall first discuss a study by Filippo Coarelli which analyses different periods of colonization in the Pontine region, problematizing the way in which archaeological evidence is adduced to support the historical narrative.6 Having done this, we present archaeological case studies of three colonial landscapes from recent field surveys of the Pontine Region Project (PRP), which document local and regional changes in rural and urban settlement from the 6th to the 3rd c. B.C. By approaching the colonies from an archaeological perspective, looking specifically at their urban and rural development over the longer term, we hope to contribute to the vexata quaestio of early colonization.

Colonization in the Pontine region: historical and archaeological backgrounds According to the literary sources, the first colonies in the Pontine region were founded in the Late Regal period. This is stated unequivocally by Livy and Dionysius of Halicarnassus for the colonies of Signia (outside the Pontine region) and Circeii, but the evidence for Pometia and Cora is more circumstantial.7 The reliability of these records is of course problematic, but the significance of the passages is a matter which deserves investigation rather than being rejected out of hand.8 Coarelli argues that these literary references are actually supported by archaeological evidence. He believes that the vast network of subterranean canals or cuniculi which drain the valley floors in the area between Velitrae, Cora and Satricum are incontrovertible evidence of the Archaic colonization of these areas, which must have had an important agricultural component.9 The motive behind this large-scale land reclamation would have been to safeguard the grain supply for Rome.10 Therefore, he regards colonization in this period as a ‘statist’ phenomenon. However, his arguments are not fully convincing. One objection is that the dating of these drainage works is still being debated, as some consider them to be part of a late 4th c. B.C. Roman reclamation effort.11 Moreover, the archaeological evidence indicates that, prior to the putative Tarquinian colonization, major settlements had already developed in various landscape units in the Pontine region: in the tuff hills

6

7



Despite the fact that colonization is not the central theme of Coarelli 1990, it is the only paper which postulates ideas about various phases of colonization in the Pontine region between the Late Regal and the Mid-Republican period. For Signia and Circeii: Liv. 1.56.3; Dion. Hal., Ant. Rom. 4.63. The existence of a colony at Cora and Pometia in the late 6th c. B.C. is implied in Liv. 2.16.7-8, although a foundation date after the expulsion of the king cannot be ruled out. As an illustration of this problem compare Cornell 1995a, 303, who lists these four colonies under Tarquinius Superbus with Forsythe 2005, 191 who lists Cora and Pometia under the year 501 B.C. (note that this is highly implausible as this is the year in which they allegedly defected from Rome). More generally, for Cora the existence of a colony in the Regal or Early Republican period has been cast into doubt by Palombi 2000, 99 and id. 2003, 202-5. See

8

9 10

11

also Torelli 1999, 16 who includes Tarracina as one of the colonies “between the last decades of the monarchic age and the beginning of the Republic”. Neither Cornell 1995a nor Forsythe 2005 deals explicitly with the reliability of the sources for the Regal period, but both consider the information on colonial foundations in the Early Republic as reliable: this information is probably derived from annalistic sources such as the Annales Maximi, which are believed to go back to the beginning of the Republic (but not before) (Cornell 1995a, 15; Forsythe 2005, 191). See also Cifarelli 1993 for a positive view on the historicity of the Regal colony at Signia, based on the presence of archaeological material of this period. Coarelli 1990, 142-48. Ibid., 143, 147-48; De Haas 2011, Chapter 12 with references to the sources. Quilici Gigli 1983. See also Attema 1993, 72-76.


FIG. 1. The Pontine region with sites mentioned in the text and the following landscape zones: - The foothills of the limestone range of the Monti Lepini, where the colonies of Norba and Setia were founded. Precolonial settlements in this area include Caracupa/Valvisciolo and Cora. To the south, in the foothills of the Monti Ausoni, is the Volscian town and Roman colony of Tarracina, overlooking the Tyrrhenian coast - The limestone promontory of Mount Circeo with the colony of Circeii overlooking the Tyrrhenian coast - The lower Pontine plain, a marshy low-lying area. It was subject to rural colonization by Rome during the MidRepublican period and the tribus Oufentina was founded in it. The area is traversed by the Via Appia - The coastal landscape of ancient and recent dunes with the colony of Antium, a site already settled in the Iron Age and Archaic period - The Astura River Valley with the colony of Satricum, a site already settled in the Iron Age and Archaic period - The tuff hills with the pre-colonial settlement of Caprifico di Cisterna, perhaps to be identified with the colony of Pometia




(Caprifico di Cisterna), the foothills of the Monti Lepini (Cora and Caracupa/Valvisciolo), the Astura Valley (Satricum) and the coastal zone (Antium) (fig. 1).12 As were the proto-urban centres along the River Tiber, the Archaic communities in the Pontine region were very probably autonomous polities.13 If the cuniculi are of Archaic date, it is quite possible that their construction was realized by the Latin peoples living in the proto-urban settlements in this area.14 In the early 5th c. B.C., the period for which the ancient sources report the foundation of new colonies at Norba (492 B.C.) and slightly later at Antium (467 B.C.), the proto-urban settlements in the Pontine region underwent considerable changes: some contracted (Satricum), others were abandoned (Caracupa/Valvisciolo, Caprifico di Cisterna).15 Changes in rural occupation were less radical and, despite a drop in the number of settlements, most rural sites show continuity from the Archaic into the Mid-Republican period.16 Coarelli’s scenario for this second wave of colonization at the beginning of the Republic (which included the re-colonization of Cora, Pometia and Circeii and the foundation of nearby Velitrae) combines ‘statist’ and ‘gentes/warlord’ elements. On the one hand, he believes that the late Archaic antefixes found at Circeii and Norba which might have been produced at Rome are evidence of the involvement of Rome in these colonies.17 On the other hand, he suggests that this colonization was bound up with the activities of the Valerii clan. Archaeological support for this point of view he finds in the famous Lapis Satricanus inscription, which mentions a certain Poplios Valesios (generally identified as Publius Valerius Publicola) and his followers.18 Clan leaders like this Poplios Valesios presumably enjoyed wide-ranging autonomy, and could be responsible for the foundation of colonies. While Coarelli thus suggests a ‘private’ warlord scenario, there are several problems of interpretation. First, he does not explain how the use of ‘Roman’-style temple decorations fits in such a narrative. Second, the use of these temple decorations as a sign of Roman cultural and/or political influence is problematic from both a typological and a chronological view, as one of us has recently pointed out.19 Finally, we should

12

13

14

15

16

17



For the Pontine region, see Attema 2005, 121-22; Attema and De Haas 2012. Rural settlement around these centres intensified in the 7th and 6th c. B.C. In contrast, no large Archaic settlements are known in the Monti Ausoni, the lower Pontine plain itself or in the Mount Circeo area. From an archaeological perspective, these areas seem to have been marginal in the process of proto-urbanization which took place in the north-western part of the region. Cornell 1995b, 123; Attema 1993, 213-26; Attema, Burgers and Van Leusen 2010, 135-46. For its interpretation as an Archaic achievement not necessarily instigated by Rome, Attema 1993, 223-24. On Satricum: see below; on Caracupa Valvisciolo: Attema 1993 and Attema and De Haas 2012; on Caprifico di Cisterna: Melis and Quilici Gigli 1972; Attema 1993; Quilici 2004; Attema and De Haas 2012. The drop in site numbers can partly – but certainly not fully – be explained by methodological problems in recognizing ceramics of this period; see also below. Coarelli 1990, 153; id. 1992, 22. See for a positive assessment of the matching of archaeological evidence with the reality of these early colonies Torelli 1999, 18: “And we do not lack archaeological evidence (such as architectural terracottas and some diagnostic pot-

18

19

tery) to support the idea of Roman presence in many of the early sites of Latium, like Norba, Satricum, and Signia, before the Volscian invasion”. Coarelli 1990, 151-52. More generally, Torelli 1999, 16 states that “the discovery of the Lapis Satricanus, apart from the controversial literal meaning of the text, has shown that in this very early colonization a principle role was played by warlike groups – the suodales of the Satricum inscription – under the command of leading figures – Poplios Valesios in the Lapis Satricanus – of tyrannical type, be these Caelius or Aulus Vibennae, Servius Tullius, both the Tarquinii, the Priscus or the Superbus, or Valerius Publicola”. Termeer 2010, 49, focusing on antefixes representing Juno Sospita. In short, only one of four types of these antefixes can be related to Rome. Specimens of this type are found in colonies and non-colonies alike, which problematizes their use as evidence of the active involvement of Rome in the colonies. A comparable discussion on terracottas as indicators of Roman influence and power concerns the earlier so-called Veii-Roma-Velletri temple decoration systems dated to the reign of Tarquinius Superbus, which have been found at Caprifico di Cisterna (e.g. Lulof 2011, 23-24).


note that while it does attest to the importance of the individual Poplios Valesios the Lapis Satricanus is not found at a colony of this period. It can therefore not be taken as a reflection of the involvement of warlords in Roman colonization programs.20 After a gap of about 75 years, the foundation of colonies at Circeii (393 B.C.), Satricum (385 B.C.) and Setia (383 B.C.) marked a third phase of colonization which also involved a large viritane land distribution program in the ager Pomptinus.21 The Pontine area was subsequently incorporated into the ager Romanus, affirmed by the foundation of the tribus Pomptina in 358 B.C.22 and the tribus Oufentina in 318 B.C. By this time, the ager Romanus also included the territories of the citizen colonies of Antium and Tarracina, founded in 338 and 329 B.C. respectively. Other parts of the region were indirectly drawn into the Roman sphere of influence, as they were part of the territories of the Latin colonies of Cora, Norba, Setia and Circeii. The period is characterized by major improvements in the infrastructure in the region, most importantly the construction of the Via Appia (312 B.C.) which opened up the lower Pontine plain to commerce and transport. Coarelli argues that this formalized colonization involved a ‘radical remodeling’ of the conquered areas, implying a Roman, state-organized policy.23

Field survey data and early colonization Coarelli’s narrative is based primarily on the historical sources and relates archaeological data directly to the histoire événementielle. As we have tried to demonstrate above, this approach is problematic: the sparse archaeological evidence he refers to does not allow such direct links to be made. Furthermore, it mixes elements of statist and ‘gentes/warlord’ colonization scenarios without properly considering the (presumed) impact of these scenarios on the archaeological record. Landscape archaeology offers a different approach to the early colonization of the Pontine region, as it looks at the subject on the timescale of the conjoncture (cf. the contributions of Bellini et al. and Marchi in this volume).24 From the outset, the Pontine Region Project (PRP) has studied early colonization primarily on the basis of field survey data, in an attempt to detect changes in settlement patterns which could relate to historically known colonial undertakings.25 In other words, it has sought to investigate the ‘geographical dimension’ of colonization. We believe that this is an important addition to the assessment of the different colonization scenarios, as these imply different scales of impact on the landscape. The assumption is that ‘statist’ colonization would have had a more profound, extensive impact on rural settlement patterns than ‘warlord’ colonization. In the following three case studies, we shall therefore combine a discussion of landscape archaeological data with an evaluation of the archaeology of the colonies. In the conclusion we shall assess to what extent the data supports any of the colonization

20

21 22

Coarelli 1990, 149, following Stibbe 1987, assumes that Satricum is to be identified with the colony of Pometia. However, the identification of the colony of Pometia is disputed: the suggestion of Melis and Quilici Gigli 1972, 241-47 that it should be identified with the site of Caprifico di Cisterna is still followed by Quilici 2004 and Chiabà 2006, 97, note 35. Either way, Pometia was a colony of the previous Late Regal phase. Therefore, although the Lapis Satricanus indicates the activities of warlords in the Early Republican period, it cannot be connected with a colonization event known from the literary sources. Cornell 1995a, 319; Coarelli 1990, 141. This tribe was named after the site of Pometia, on

23

24

25

whose territory it was presumably established. Ibid., 141-42 stating that “Si tratta di un processo di radicale rimodellamento del territorio, che si risolve in una profonda romanizzazione e nella totale destrutturazione dell’insediamento più antico.” While he does not explicitly refer to archaeological evidence, we will argue below that it is indeed in this phase that the impact of colonization becomes clear in the archaeological record. On landscape archaeology and the Annales paradigm Bintliff 1991; Knapp 1992. For the Pontine region, see Attema 1993, 18-21, 212; De Haas 2011, 5-14, 303. Attema 1993; Attema and Van Leusen 2004 see for a review also De Haas forthcoming.




scenarios in different phases. Before doing so, some methodological remarks on the interpretation of field survey data are in order. Not all landscape archaeological studies are equally suited to examine the possible effects of early colonization on rural settlement. For example, high-quality topographic inventories are available for the territories of Circeii, Tarracina, Cora and Satricum, but these are biased towards architectural remains and pottery scatters with well-known ceramic classes of the (Late) Republican and Imperial periods.26 As there are not yet any well-defined pottery typo-chronologies for the 6th to the 4th c. B.C. and many wares continued to be used in the 3rd c. B.C., pottery fragments diagnostic of this period are scarce in surface contexts. Consequently, topographic surveys have left the Archaic and Early Republican colonial landscapes undetected.27 Thanks to the application of intensive field survey strategies and detailed pottery studies based on fabric analyses, the research carried out by the PRP over the last 25 years has detected many smaller pottery scatters of the Archaic and Early Republican periods, or, as we shall refer to the period 480-350 B.C., the Post-Archaic period.28 However, the functional interpretation and chronology of these scatters remain problematic: they are often represented only by small amounts of pottery on multi-period sites, which means that we can only tentatively classify sites of the colonial periods into types.29 Furthermore, they often have broad date ranges, which only allow an Archaic (6th c. B.C.), a Post-Archaic (5th - mid-4th c. B.C.) and a Mid-Republican phase (350-200 B.C.) to be discerned.30 Therefore the rural settlement data we use in the following case studies allow us to track changes in the context of the early 5th c. B.C. colonization, but only to a lesser extent for the early 4th c. B.C. phase of colonization.31 Finally, in order to deal with the differing amounts of dating evidence on individual sites, we also distinguish ‘certainly’ and ‘possibly’ occupied sites for each of these periods.32

Three case studies The three case studies we present focus on four urban colonies and their territories: Norba (Case 1), Antium and Satricum (Case 2) and Setia (Case 3), and one case of colonization through viritane land distributions around the Via Appia in the lower Pontine plain (Case 3). Each case study commences with an overview of the most relevant information provided by the ancient sources.33

26

27

28

29



Lugli 1926; id. 1928; Brandizzi Vittucci 1968; Piccarreta 1977; Attema 2000. The same can be said about a series of small-scale topographic inventories (Zaccheo and Pasquali 1972; Cancellieri 1987; Bruckner 1995). Although Piccarreta 1977, 18 states in his introduction that 5th-4th c. B.C. artefacts occur on many rural sites, he has not recorded their presence at individual sites. For the term Post-Archaic, see Attema 1993, 25. For case studies and discussions of field survey methodology, see Attema and Van Leusen 2004, Attema, De Haas and Tol 2011; De Haas 2011. For a study of 6th to 4th c. B.C. pottery from stratified contexts at Satricum in relation to surface pottery, see Attema et al. 2001-2; Attema and Van Oortmerssen 1997-98. The only distinction we can observe for the Archaic period is between sites with tiles and those without.

30

31

32

33

Scatter size in some cases provides an additional criterion of differentiation. The typical Post-Archaic fabrics (sandy-white to yellow fabrics, often described as impasto chiaro sabbioso) occur over a longer time span, and their association with Black Gloss wares implies that these were still used in the Mid-Republican period. The data presented in the case studies below are largely derived from De Haas 2011 (cf. Attema et al. forthcoming). In our discussion of the colonies, similar issues of archaeological visibility do of course play a role, as not all of these are equally wellknown archaeologically (cf. Termeer 2010, 47-49). As argued in De Haas 2011, the settlement trends, including both certain and possible sites, should be regarded as more reliable, especially for the Mid-Republican period. Attema 1991; id. 1993, Chapter 7.


Case 1: Norba Livy records that Norba was founded in 492 B.C. as a stronghold in the Pontine region.34 No other mention of Norba is made in the sources for the first 150 years after its foundation. Only in the mid-4th c. B.C. does Norba appear again, as a loyal ally of Rome: it is the victim of attacks by the Privernates first in 341 B.C., and again in 330 B.C., when its territory was plundered.35 In the Social War, in 81 B.C., its inhabitants set fire to their houses to prevent the town from being taken by Sullan troops.36 After this, Norba disappears from the historical record. Although the archaeological evidence points to frequentation of Norba before its recorded FIG. 2. Site distributions around Norba in the Archaic, Postfoundation,37 the primary settlement in the area in the Archaic and Mid-Republican period (white: certain sites; Archaic period was the proto-urban site of Caracupa grey: possible sites). Valvisciolo.38 A number of fortified sites has been identified in its surroundings, and PRP research at the site of Contrada Casali suggests that some of these formed a class of secondary centres, probably villages. From the intensive field surveys, we now know that many rural sites existed in the area as well (fig. 2).39 On many such sites scattered along the foothills of the Monti Lepini, Archaic tiles have been found, indicating the presence of permanently inhabited farms. These occur side by side with pottery scatters without evidence of tiled roofs; the remains of ephemeral structures probably representing poor farms or non-settlement sites. In the plain away from the Lepine foothills, the extent of rural infill is more difficult to assess because of the dearth of surveys.

34

35 36

Liv. 2.34. A possible pre-existing town is implied by Dionysius of Halicarnassus, who lists Norba as a member of the Latin League (Dion. Hal., Ant. Rom. 5.61, reading Νωρßανῶν for Mωρεανῶν). Ogilvie 1965, 280 suggests that the inclusion of Norba in the list is an anachronism. Liv. 7.42; 8.1; 8.19-20; 27.10. App., B Civ. 1.94.

37

38

39

For pre-colonial remains, see Savignoni and Mengarelli 1901; De Spagnolis 2003. For a discussion see also De Haas 2011, 241-42. Savignoni and Mengarelli 1903; Mengarelli and Paribeni 1909; Quilici and Quilici Gigli 1987; iid. 1988; Attema 1993. See also Attema and De Haas 2012. Van Leusen et al. 2003-4; De Haas 2011, 244-45.




FIG. 3. Site trends for the Lepine margins around Norba, the coastal landscape between Satricum and Antium and the lower Pontine plain.

At the end of the Archaic period, major changes in settlement occurred. Although some defended sites show continuity into the Post-Archaic period,40 the main sites of Caracupa Valvisciolo and Contrada Casali were either abandoned or much reduced41 (fig. 2). Around this time (when its foundation is also historically attested), Norba developed into an important centre, as three sanctuaries were built at the site.42 Other Post-Archaic remains have been found in the lower strata of the excavations of the town walls and in the depression between the two acropoleis.43 A Norban origin has been suggested for an inscription referring to an aedil, tentatively dated to the 5th c. B.C.44 If it indeed refers to a magistrate from Norba, it would imply that the site already had an official administrative organization in the Post-Archaic period. Although a general reduction in rural settlement occurred in the area around Norba during the PostArchaic period, many sites do show continuity from the Archaic period. On virtually all rural sites from this period, tiles occur and they probably represent farms. Consequently, although it seems that the newly founded colony of Norba usurped the functions of Caracupa Valvisciolo as center of the area, the impact of colonization on the rural settlement pattern was limited. Our data suggests that there was no demographic increase in the rural population at the time of colonization in this area (fig. 3). In the Mid-Republican period, Norba developed into a settlement with urban features including defensive

40 41

42

ď&#x153;˛ď&#x153;ąď&#x153;¸

Quilici Gigli 2004; De Haas 2011, 246-47. At Caracupa Valvisciolo, traces of Post-Archaic settlement have only been found on the defended slopes; the lower settlement was abandoned and the votive deposit was closed. On the settlement, see Attema 1993, 168-78; for the Post-Archaic traces on the defensive terraces Attema 1993, 162-63. On the votive deposit, see Mengarelli and Paribeni 1909, 252 and Mangani 2004. For Contrada Casali, see Attema 1991 and id. 1993, Chapter 7. On the sanctuary of the minor acropolis, see Savignoni and Mengarelli 1901, 538-39 and Quilici Gigli 1996,

43

44

289-90). On the sanctuary of Juno Lucina, see Savignoni and Mengarelli 1903, 248; Rescigno 2003; Bellelli 2003. On the sanctuary of the major acropolis see Savignoni and Mengarelli 1901, 530-34; Quilici Gigli 1996, 289. Savignoni and Mengarelli 1901, 547-48; see also Bellelli 2003, 397 with references. CIL I2 361 = CIL VI 357. Quilici Gigli 1996, 290-95; Carfora and Nonnis 2009, 183-85. It should be noted that both the date of this inscription and its provenance have not been established beyond doubt.


walls and a street grid.45 The three sanctuaries were actively being used, as is shown by the abundant presence of Mid-Republican votives. Minor settlements in the area continued to function as well. Rural settlement in general expanded, generating many new foundations and the continuation of Post-Archaic sites along the foothills.46 There were changes in site typology in this area, as polygonal masonry platforms and/or agricultural terraces were constructed at many sites.47 Modest farms certainly continued to exist alongside these more monumental sites. Case 2: Antium and Satricum In our second case, we discuss Antium and Satricum, as our field surveys have covered parts of the hinterland of both colonies.48 Antium and Satricum are generally considered towns of Latin origin, but the historical sources for both record a period of Volscian control in the 5th and 4th c. B.C., followed by the foundation of a colony. Satricum was conquered by Volscian troops led by the exiled Roman commander Coriolanus in 491 B.C., and might have continued to be under Volscian influence afterwards (cf. Bradley in this volume).49 Antium is repeatedly mentioned in the context of struggles between the Romans and the Volsci in the first half of the 5th c. B.C. The Romans established a colony in 467 B.C., but just three years after its inhabitants showed themselves to be less than co-operative towards Rome,50 and thereafter Antium again sided with the Volsci. Historical references to Antium in the second half of the 5th c. B.C. are rare, which could imply that the situation in the area had temporarily stabilized.51 With Roman raids of the countryside around Antium in 406 B.C., instability returned, marked by uprisings in Velitrae and Satricum.52 Subsequently, a colony was founded at Satricum in 385 B.C.53 Livy records that 2,000 colonists were sent out, each of whom was granted a plot of 2.5 iugera.54 However, the foundation of the colony did not provide Rome with a stable base in this contested area: in 383 B.C. the Volsci conquered

45

46

47

48

49

The polygonal masonry walls date to the mid- or late 4th c. B. C., with major reconstructions in the first half of the 3rd c. B.C. (Quilici and Quilici Gigli 2001, 24344; Quilici Gigli 2003, 23-25). The street grid presumably shows two phases: a first one with a number of diagonal axes across the site and a second, orthogonal grid pertaining to the Late Republican period (Schmiedt and Castagnoli 1957, 147; Quilici Gigli, Ferrante and Caputo 2003, 316). The access roads with their polygonal masonry revetment walls probably date to the 3rd c. B.C. (Quilici Gigli 1988, 229; Quilici 1991, Quilici and Tognon 2001; Quilici Gigli 2003, 25-26). Because of the limited chronological resolution of much Republican pottery, only a low number of rural sites can be defined as certainly Mid-Republican. However, non-diagnostic Black Gloss shards and socalled tegole sabbiate, tiles of Republican date, occur at many sites. Many of these are likely to have been occupied in the Mid-Republican period. For a discussion of their chronology, distribution and function see De Haas, Tol and Attema 2012. Attema et al. 2007-8; Attema, De Haas and Tol 200910 and iid. 2011. The historical development of Antium and Satricum was closely linked. Liv. 2.39. After the Coriolanus episode, Satricum is not

50

51

52

53 54

mentioned in historical sources for almost a century. The foundation of the colony is reported in Liv. 3.1 and Dion. Hal., Ant. Rom. 9.1-2. Both mention the inclusion in the colony of non-Romans; Volsci in Livy and Latins and Hernici in Dionysius. See also Bradley 2006, 167-68 on Antium. For the changing relation with Rome, see Liv. 3.4 and 5, at the end of which Livy states that 1,000 soldiers from Antium were sent back because they came to aid Rome too late in its battles against the Aequi. The period of few references to both Antium and Satricum might reflect a situation in which political and military conditions in the coastal area had become more stable. Rome at this time focused its military campaigns against the Aequi in the area north of the Alban Hills (Van Royen 1992, 452-53). For comments on the continued Volscian influence in Satricum, see ibid., 451. Liv. 4.59. In his description of the struggles with the Volsci and Aequi in 408-7 B.C., Livy also mentions Antium as heading the forces and as the place where these were beaten (Liv. 4.56-57). However, in this context he also states that the Romans laid waste to a nearby camp, which was situated alongside the Fucine Lake. It is therefore possible that for Antium we should actually read Antinum. Diod. Sic. 14.102; Liv. 6.6-8. Liv. 6.16; Cornell 1995a, 303.




Satricum, and Roman attempts to regain control in 381 B.C. apparently failed.55 The defeat of combined forces of Latins and Volsci by Rome in 377 B.C. led to the surrender of Antium to Rome, and the destruction of Satricum by its disappointed Latin allies, who spared only the temple of Mater Matuta.56 In 347 B.C., Antium and Rome were locked in competition over Satricum: Antium, still under Volscian control, sent out a colony to rebuild Satricum, but Roman forces took the town and burnt it to the ground in 343 B.C.57 After the Latin War, Antium became a citizen or maritime colony in 338 B.C.58 Later, it developed into an important harbour town and imperial residence, but Satricum disappeared FIG. 4. Site distributions between Satricum and Antium in from the historical record after the destruction of the the Archaic, Post-Archaic and Mid-Republican period temple of Mater Matuta in 207 B.C.59 (white: certain sites; grey: possible sites). In both Antium and in Satricum, remains go back to the Iron Age.60 The archaeological evidence shows that in the Archaic period, Satricum was a proto-urban settlement with several sanctuaries (most notably the temple dedicated to Mater Matuta on the acropolis), courtyard houses and a road network in the lower town.61

55

56 57

58



However, the course of events is somewhat confused in this episode; see also Gnade 2002, 146 (note 541). Liv. 6.22 and 32-33. Liv. 7.27; for the second time the temple of Mater Matuta was spared. However, the repetition of the destruction of the town and the sparing of the temple might be a duplication (Oakley 1997, 456-57). Liv. 8.13-14, stating that the Antiates themselves could also enrol. According to Bandelli 1995, 173, Antium became a civitas sine suffragio, to which a colonia maritima was added, and therefore only the colonists received full citizenship. Cornell 1995a, 349-50, however, suggests that all inhabitants received full citizenship.

59 60

61

Liv. 28.11. For Anzio see Alessandri 2009, 104-10, for Satricum see Alessandri 2007, 95-98. For a general overview, Gnade 2007; Attema and De Haas 2012. Literature on the Mater Matuta sanctuary is extensive; see, most recently, Knoop and Lulof 2007. A second sanctuary is located in the south-western part of the settlement: Ginge 1996, 85-94. For the excavations on the acropolis: Maaskant Kleibrink 1987; ead. 1992b; Gnade 2003. On the courtyard houses: Stobbe 2007; Van ‘t Lindenhout 2010. On the roads in the lower settlement area: Gnade 2002, Chapter 2; ead. 2006 and ead. 2007, 51-58.


The remains of agger/vallum type defences and the topography indicate that in this period the settlement extended over an area of 40 ha.62 There is evidence of local production of pottery, tiles and metal.63 The status of Antium in the Archaic period is less clear-cut, as the published pre-colonial remains are almost exclusively of Iron Age date, including agger/vallum type defences and necropoleis.64 Presumably, the settlement covered the area known as the Vignacce plateau, an area of some 22 ha.65 In the Archaic period, the coastal landscape between Satricum and Antium was exploited from isolated rural settlements which can be classified into three types on the basis of the surface pottery assemblages (fig. 4).66 Those on which only Archaic pottery occurs probably represent simple hut dwellings or non-settlements, while those with Archaic tiles represent more durable structures with tiled roofs, interpreted as farmsteads. Furthermore, on the basis of its size and the presence of tiles, one site east of Satricum has been interpreted as a hamlet.67 Sites with Archaic tiles are more abundant in the Astura Valley south of Satricum than in the wellinvestigated area northeast of Antium, and Satricum presumably supplied such rural sites via the River Astura.68 In the Post-Archaic period, the settlement system underwent major changes. Although solid data about the extent and character of the settlement at Antium in this period lacks, continuity of occupation is suggested by a 5th c. B.C. opus quadratum reinforcement wall in the agger.69 In contrast, there is plenty of evidence of the Post-Archaic settlement of Satricum. Fifth c. B.C. cemeteries occur in various parts of the former settlement area and the tombs and their inventories indicate the presence of Volscian people.70 The location of the cemeteries suggests that the settlement contracted in the 5th c. B.C. and lost most of its urban characteristics. At this time, it probably consisted of a number of small, isolated nuclei. Economic conditions possibly improved in the later 5th c. B.C., as the tombs of this period contain larger numbers of gifts.71 Pottery and metalwork continued to be produced at the site, although on a smaller scale than in the Archaic period.72 There is little evidence for

62

63

64

65

66

67

Gnade 1999. Remains of the agger were already noted during the late 19th and early 20th century excavations, and Castagnoli hypothesized an agger of some 5 to 6 m high on which a palisade would have stood; he also observed a curvilinear wall in opus quadratum which might represent a retaining wall (Castagnoli 1963, 505-18). Quilici Gigli has recently suggested that the settlement would have covered a significantly smaller area (Quilici Gigli 2004, 256-60). Nijboer 1998, 83-89; 244-47. Based on fabric studies of material from the Agro Pontino Survey, Mater 2005, Chapter 3 concludes that the pottery produced at Satricum was not distributed over large areas. The construction of the agger is dated between the 9th and the early 7th c. B.C. (Egidi and Guidi 2009). On the necropoleis, see Gierow 1961; De Meis 1984. Both a votive context and the necropoleis reportedly show continuity from the Iron Age into the Republican period, but as far as we know there is no solid, published evidence for this. Attema, De Haas and Tol 2011, Chapter 8; Attema et al. 2007-8; Attema, De Haas and Tol 2009-10. For an example of an Archaic farmstead, see Site 15108 discussed in Attema, De Haas and Tol 2009-10. For the hamlet, see Attema et al. 2007-8, 432-33 and fig. 14.

68

69

70

71 72

Nijboer 1998, 88-89 states that the Archaic pottery workshop at Satricum served a local market, especially in Satricum itself. However, rural pottery workshops are unknown in the area and no pottery production at Antium has so far been attested. In addition, both the votive context on Vignacce Hill and the necropolis west of Antium were supposedly continuously used, but so far no 5th c. B.C. graves or votives have been published (Brandizzi Vittucci 2000, 81-83). The exact date and the historical context of this wall are disputed (see Chiarucci 1989; Brandizzi Vittucci 2000; Egidi and Guidi 2009, 360). The main concentration of 5th c. B.C. graves can be found in what is known as the southwest necropolis: Gnade 1992; ead. 2002. Other graves were found on the acropolis: Maaskant Kleibrink 1992a; Ginge 1996, 136-37; Gnade 2003. Recent excavations have also brought to light graves in the lower town area: ead. 2007; ead. 2009. Ead. 2007, 65-66. Attema et al. 1992; Nijboer et al. 1995; for the products of the pottery workshop, see also Nijboer 1998, 89-91. For the metalwork: ibid., 264.

ď&#x153;˛ď&#x153;˛ď&#x153;ą


continued settlement in the 4th c. B.C., after the historically attested foundation of a colony in 385 B.C.73 In the countryside between Satricum and Antium, rural settlement contracted, although this contraction is less marked in the Astura Valley south of Satricum, where some new sites emerge. Most rural sites probably represent farmsteads with tiled roofs, but there are also larger nucleated settlements. The evidence from the towns and the countryside suggests a decline in population, settlement organization and economic conditions (fig. 3). As in the case of Norba, there is no indication of an influx of colonists or of a (re-)organization of the territory as part of the reported colonization events.74 In the Mid-Republican period, Antium gained importance. Although settlement evidence is scarce, several cultic contexts and two necropoleis from this period are known.75 By contrast, at Satricum only few remains dating to the 4th and 3rd c. B.C. century have been found.76 The southwest necropolis was abandoned in the early 4th c. B.C., and only few later tombs have been reported so far.77 Nevertheless, religious activity was obviously little affected, and probably continued to sustain the production of metal and ceramics.78 In the same period, the countryside grew more densely inhabited: the data indicate a sharp rise in the number of rural sites (figs. 3 and 4).79 Areas which had been abandoned in the Post-Archaic period were re-occupied and in areas north of Antium and south of Satricum settlement expanded. The survey data provide limited evidence of functional and/or typological differentiation between sites, but the majority should be regarded as modest farms with tiled roofs. The nucleated settlements which had existed in the previous period disappeared, perhaps an indication that people left these villages to live on isolated farms. However, a new nucleated settlement might have arisen in the coastal strip.80

73

74

75

76



Recent excavations in the Poggio dei Cavallari area have uncovered additional 4th c. B.C. tombs, but their precise chronology remains unresolved as yet (Marijke Gnade’s personal comment). It is perhaps not surprising that the existence and character of the colony founded at Satricum in 385 B.C. remains elusive, as the sources state that it was destroyed a few years after its establishment. Mid-Republican votive material has been found both on the Vignacce Plateau (Chiarucci 1989, 35 and 95; Jaia 2004, 256; Jaia et al. 2007) and in the lower area to the south ( Jaia 2002; id. 2004; Brandizzi Vittucci 2000, 49; Bouma 1996 vol. 3, 14; Jaia 2004, 256). Necropoleis are situated southwest and west of the Vignacce Plateau (Alessandri 2009, 104-5). Settlement strata of this period are almost entirely absent, although these could have been destroyed by ploughing and levelling. The agger had certainly fallen into disuse in the 3rd c. B.C. (Gnade 1999). The only in situ settlement remains are on the acropolis, and probably represent a rural courtyard building (Louwaard 2007, 75). In addition, artefacts of the 4th and 3rd c. B.C. have been reported in surface contexts in the Poggio dei Cavallari and Fornace areas (Gnade 2007, 52). Hence, some settlement clusters might have remained, perhaps representing isolated rural establishments or small settlement clusters.

77

78

79

80

See above note 73. Some Hellenistic tombs are mentioned in the diaries of the 19th century excavations: see Waarsenburg 1995, 123, n. 380; 146, n. 436. The sanctuary on the acropolis shows continuity, as votive deposits II and III contain Mid-Republican dedications. Votive deposits are also known from other parts of the site (Bouma 1996; Heldring 2007; Ginge 1996, 79-80). For continued pottery and metal production: Attema et al. 1992; Nijboer et al. 1995; Nijboer 1998, 129; Bouma 2001. As noted above, the limited chronological resolution of much Republican pottery only allows the identification of rural sites as possibly Mid-Republican. However, many of these sites with non-diagnostic Black Gloss shards and tegole sabbiate are likely to have been occupied in the Mid-Republican period. This is confirmed by the occurrence of Mid-Republican Black Gloss fragments, Graeco-Italic and Punic amphorae and collar rim jars. This settlement consists of a cluster of some 30 sites mapped by Piccarreta 1977. As they are located on a military base, they have not been investigated during Groningen Institute of Archaeology (GIA) field surveys.


Case 3: Setia and the colonization of the lower Pontine plain Our third case discusses Setia and the lower Pontine plain to its south. Setia itself is referred to in the sources for the first time on the occasion of its foundation as a colony on the Volscian frontier in 383 B.C.81 A second contingent of colonists was sent to Setia as reinforcements only three years later.82 During the Latin War, Setia sided with other Latin towns against Rome, and it retained its Latin status after 338 B.C.83 Although it is doubtful whether the viritane distributions to plebeians of 383 B.C. and the establishment of the tribus pomptina in 358 B.C. concerned the area under discussion, it was certainly involved in subsequent distributions in the later 4th c. B.C.84 The tribus Oufentina, created in 318 B.C., takes its name from the main river crossing the lower plain, the Oufens.85 That this area was under direct Roman control by this time is also evident from the construction of the Via Appia, commissioned by Appius Claudius in 312 B.C., which as a consular road traversed ager publicus. The Decennovium Canal was probably excavated contemporaneously, and together they can be considered the first securely dated investment by the Roman State in the reclamation of the plain.86 At Setia archaeological evidence of the Archaic period is completely absent.87 However, in the plain below Setia a Late Archaic sanctuary at the site of Tratturo Caniò is attested to by architectural terracottas, bronze kouros statuettes and Archaic pottery.88 This sanctuary arose at a location which had already been frequented over a long period.89 Field surveys show that the plain east of Tratturo Caniò was certainly frequented during the Archaic period, as Archaic tiles are found on a number of sites, which probably represent farmsteads (fig. 5).90 In contrast, an absence of such tiles suggests that the lower lying area farther to the south, part of a former lagoon, had not yet been permanently settled, and was frequented only on a seasonal basis. Post-Archaic settlement remains have not been found at Setia, but some assume that its polygonal masonry walls were erected at the time of the establishment of the colony in 383 B.C., or at the latest by the time of the Latin War.91 Although it has been suggested that there was a gap in frequentation between the first quarter of the 5th and the late 4th c. B.C. at the sanctuary at Tratturo Caniò,92 a study of the surface remains of a ploughed-out votive deposit at the site suggests that it was continuously frequented from the Archaic into the Mid-Republican period. 93 From the 4th c. B.C., rural occupation expanded into the area between Setia and Tratturo Caniò. Most rural sites in this area include typical Post-Archaic and MidRepublican tile fabrics and most of them probably form permanently settled farms (fig. 5).94

81

82

83

84

85

86

87

Vell. Pat. 1.14.2. Dion. Hal., Ant. Rom. 5.61 refers to Setia as a member of the Latin League, but this could be an anachronism (Ogilvie 1965, 280). Liv. 6.30. The sources do not mention the size of these colonist contingents, nor do they report on allotments. One of the leaders of the Latin revolt, the praetor Lucius Annius Setinus, was a native of Setia (Liv. 8.3-5). More likely, the allotments of 383 B. C. and the tribus pomptina should be placed in the tuff hills to the north (De Haas 2011: Chapters 8 and 12). Liv. 9.20. It is possible that slightly earlier allotments were made in the area, as such allotments are reported in the territory of nearby Privernum in 341 B.C. and, perhaps, 329 B.C. (Liv. 8.11; Ross Taylor 1960, 56; Pelgrom 2008, 322). See for the likelihood of contemporary construction Humm 1996; for a concise discussion see Pelgrom 2012, 101-3. Archaic finds have been recovered from a hill top to its west (De Haas 2011, 217-18).

88 89

90

91

92

93

94

Attema 2001; Bruckner 2003; Cassieri 2004. Artefacts of the Bronze and Iron Ages have been found at the site (Bruckner 2003, 94-95: Quilici Gigli 2004, 247-52; Van Leusen 2010; Feiken et al. 2012). Attema and Van Leusen 2004, 177; cf. Attema, De Haas and Tol 2014. Presence of Archaic tiles was recorded in transects but on account of methodological restraints cannot be ascribed to individual sites. For the first option, see Zaccheo and Pasquali 1972; for the second, see Bruckner 2000, 103-4. Bruckner 2003, 78-79; see also Cassieri 2004. They presume the sanctuary was abandoned as a consequence of Volscian occupation. Attema 2001, 74-78. Recent excavations at Tratturo Caniò have confirmed the presence of 5th and 4th c. B.C. century coarse wares (Van Leusen 2010). See also Attema and Van Leusen 2004, 177-78; Feiken et al. 2012.




This rural expansion in the 4th c. B.C. clearly distinguishes Setia from the two preceding case studies (fig. 3). Peter Attema and Martijn van Leusen connect this expansion to the foundation of Setia as a colony, although they leave open whether it implied an influx of colonists or instead a gradual expansion of local inhabitants into the plain.95 We note that the impact of Setia was limited to the northern part of the plain, as in the area to the south where tiles are absent, sites still represent seasonal establishments rather than permanent settlements. In all likelihood the environmental conditions in this area were still adverse to arable farming. FIG. 5. Site distributions in the lower Pontine plain in the This situation changed in the Mid-Republican Archaic, Post-Archaic and Mid-Republican period. Methodperiod. Settlement evidence from Setia itself is still ological restraints have meant that the presence of Archaic absent, but cultic contexts are known from various and Post-Archaic tiles has not been recorded for individual locations outside the polygonal masonry enceinte and sites (white: certain sites; grey: possible sites). large numbers of votives also attest to the continued frequentation of the sanctuary at Tratturo Caniò.96 Moreover, rural site numbers show a sharp increase in the Mid-Republican period.97 In the area between Setia and Tratturo Caniò, the rural settlement pattern stays as dense as in the Post-Archaic period, but farther south a major expansion in rural settlement now took place (fig. 5). Many rural sites were established along the Via Appia, but probably also at a distance of up to 3 km from the road.98 These sites contain tiles and most of them

95

96

97



Attema and Van Leusen 2004, 179-80 and 190-1. That there was a local population is suggested by the existence of the sanctuary of Tratturo Canió. Furthermore, Archaic rural sites in the nearby foothills are also known. Zaccheo and Pasquali 1972, 96-97; Bruckner 2003, 8384. A possible additional cult place was identified during recent PRP surveys (De Haas 2011, 217-19). For comments on the high proportion of ‘possible’ sites, see the previous case studies. The common oc-

98

currence of 3rd century Black Gloss pottery in this study area supports the view of a sharp increase in site numbers in the Mid-Republican period (De Haas 2011). A note of caution should be sounded here, as few studies have so far targeted areas away from the Via Appia. The evidence from recent PRP surveys (De Haas 2011, Chapter 4), however, point to a widespread occurrence of Mid-Republican sites.


can be interpreted as farms rather than as temporary establishments.99 Judging from the regular occurrence of 3rd c. B.C. Black Gloss, the inhabitants of these farms had access to relatively expensive tableware which might have been produced locally.100 This increase in rural site numbers in the lower plain is probably connected to infrastructural improvements, including the construction of the Via Appia and the Decennovium Canal, which provided the plain with a new transport route and better drainage conditions. New centres such as Forum Appi and Ad Medias also grew up along the Via Appia.101 We would argue that these developments were the result of direct interventions by Rome rather than a local effort by the colony of Setia: as a consular road, the Via Appia was established on Roman soil, and the sites of Forum Appi and Ad Medias were probably founded in conjunction with it. Furthermore, in the 2nd c. B.C. a Roman consul was involved in repairs to a temple at the sanctuary of Tratturo Caniò, suggesting that at this time the site was situated on ager Romanus.102 The archaeological evidence supports the idea that the lower plain was colonized in the latter part of the 4th c. B.C. The construction of the Via Appia and concomitant drainage works must have been a prerequisite for its regular exploitation.103 Moreover, there is cartographic evidence for the existence of a centuriation grid which covers more than 200 km2 of the lower plain. Although any connection between this grid and the foundation of the tribus Oufentina is speculative, the settlement evidence clearly indicates large-scale reclamations and the widespread development of arable farming.104

Discussion In the introduction to this paper we referred to the various scenarios proposed in the literature to describe the nature, scale and driving forces behind colonization during the Early and Mid-Republican period. Such scenarios range from statist (Rome) to private enterprises (gentes, warlords, not necessarily of Roman background). We drew attention to Coarelli’s study which proposes that state-led Roman interventions in the Pontine region had commenced as early as the Late Archaic period and argues for the intervention of powerful gentes in the colonization events of the 5th c. B.C. During the Mid-Republican period the colonization of the Pontine region would in his view have become institutionalized. As the archaeological evidence adduced to support these scenarios is as a rule anecdotal and multiinterpretable, in this paper we have tried to broaden the archaeological basis of the debate by bringing in a landscape archaeological approach to get a better grip on the scale of the colonial enterprises and their impact on settlement patterns. The archaeological observations described in our case studies indicate that the settlement organization of proto-urban polities of the Archaic period destabilized at the end of the 6th c. B.C. or the beginning of the 5th c. B.C.105 This corroborates the ancient sources, which describe the first

99 100

101

102

De Haas 2011, Chapter 4. This is suggested by the presence of a palmette dye at the site of Tratturo Caniò (Cassieri 2004, 174-76). At Forum Appi, Mid-Republican coarse wares and Black Gloss pottery have been found as well as votives dated to the 3rd and 2nd c. B.C. (Bruckner 1995). At Ad Medias, similar finds have been reported, including a series of 3rd c. B.C. dedications on small sheets of bronze (Cancellieri 1975; Solin 1999; Maassen 2006). For recent work on these sites, see now Tol et al. 2014. The consul’s name is Postumius Albinus. He is to be identified with either Spurius Postumius Albinus, consul in 110 B.C., or Aulus Postumius Albinus, consul in

103

104 105

99 B.C. (Bruckner 2003; Cassieri 2004, 177). Environmental evidence also suggests that land improvements were made in the Republican period (De Haas 2011, 213-15; Walsh, Attema and De Haas 2014). For an extensive discussion, see De Haas 2011, Chapter 8. As we have not dealt with any of the Late Regal colonies in our case studies, the impact of this phase of colonization cannot be assessed in detail. However, we have noted that Roman colonial interventions aimed at reclaiming land in the tuff hills during the Regal period cannot be substantiated by archaeological evidence, as such initiatives might also have emanated from Latin polities.




half of the 5th c. B.C. as a period of unrelenting warfare involving the destruction and new foundation of settlements. At the same time we note a considerable differentiation in the changes in urban and rural settlement patterns in various parts of the Pontine landscape, which indicates that the effects of colonization in the Early and Mid-Republican period were not uniform across the landscape (fig. 3). I) On the Lepine scarp, there is a probable causal connection between the abandonment of Caracupa Valvisciolo and the developments on the plateau of Norba, where the erection of temples is most conspicuous and might be related to the early years of the colony. However, neither the abandonment of Caracupa Valvisciolo nor the foundation of Norba seems to have had a lasting impact on the rural settlement pattern; despite a decline in site density, rural settlement shows a higher degree of continuity compared to the coastal landscape. The foundation of Norba might have been responsible for maintaining stability in the rural landscape.106 II) In the coastal area around Satricum and Antium, the scale of changes brought about in the Archaic settlement pattern was considerable. The configuration of the settlement of Satricum changed in the 5th c. B.C. and this might have been related to the Volscian presence at this site. A reconstruction of changes in the settlement of Antium during the period of Volscian occupation is not possible, as too little is known of the site in this period. Rural settlement between Satricum and Antium shows signs of decline in both population and economic conditions in the Post-Archaic period, and this allows us to deduce that warfare did indeed have a considerable impact here. However, the foundation of colonies at these sites does not seem to have affected the longer-term rural developments in the area, nor did it leave any clear archaeological traces at either Antium or Satricum. III) In the lower Pontine plain, the sanctuary at Tratturo Caniò was probably continuously in use throughout the Archaic and Post-Archaic period. In the Archaic and Post-Archaic periods, rural settlement remained limited to the area east of Tratturo Caniò, while only in the 4th c. B.C. rural settlement numbers increased and a definite expansion into formerly marginal areas took place. This development might have been initially connected to the foundation of a colony at Setia and, later, to direct colonization from Rome. When considering these observations, first of all it is remarkable that the 5th c. B.C. phase of colonization affected only those parts of the landscape which were already settled in the Archaic period. This implies that, in this period, colonization did not entail the reclamation of new land. Secondly, the differentiation in changes argues against the implementation of a ‘master plan’ of colonization pertaining to the Pontine region as a whole. Instead it supports a scenario of a series of events unconnected in time and space. The case of Norba seems to represent a larger-scale planned intervention with the establishment of three sanctuaries, but the impact on the regional rural landscape remained limited. We would suggest that such changes in the religious sphere do not sit easily with a warlord scenario. On the other hand, the lack of recognizable impact of colonization at Antium and, in the 4th c. B.C., Satricum, suggests that such planned large-scale interventions were not necessarily the rule. The archaeological reality at these sites is perhaps more easily explicable by a warlord scenario. Larger scale planned interventions such as the foundation of a new settlement might therefore have occurred alongside small-scale undertakings. The archaeological data for the later 4th c. B.C. indicate structural changes in the Pontine region. At this

106



The site of Caprifico di Cisterna in the tuff hills was, just as Caracupa Valvisciolo, abandoned at the end of the Archaic period. The limited survey of the landscape around Caprifico so far unfortunately does not allow a quantification of the change in the rural land-

scape, though some significant changes in the rural settlement pattern were noted: following the abandonment of Caprifico di Cisterna a dispersed pattern developed in the Post-Archaic period (Attema 1993, 208-9).


time, the construction of the Via Appia and the concomitant drainage works opened up previously marginal areas in the lower Pontine plain to arable farming. This area was clearly involved in a type of colonization which had a far more profound impact on the rural landscape than did the founding of colonies in the 5th c. B.C. The scarcity of Archaic materials in the lower Pontine plain and the very ephemeral Post-Archaic settlement remains in this area indicates that we are now dealing with reclamation of formerly unfarmed land, and it is plausible to relate this reclamation to a late 4th c. B.C. Roman colonial initiative. This conclusion ties in well with Coarelli’s suggestion that there was a radical remodelling of the territory during the Mid-Republican period, albeit limited to previously uncultivated land. The traditionally farmed lands in the territories of the former Archaic settlements developed more organically, despite temporary setbacks during the Post-Archaic period.

Conclusion In this paper we have tried to bring together historical scenarios and archaeological evidence and to broaden the historical/archaeological debate on early colonization by adding a landscape perspective. The archaeological evidence indicates that in the early 5th c. B.C., there was no question of a co-ordinated attempt by Rome to colonize the Pontine region and we are dealing with interventions unrelated in space and time. In the 4th c. B.C., colonization did involve the exploitation of new territory, which is most apparent in the lower plain near Setia where a rural landscape was developed from scratch. We find the putative link to the construction of the Via Appia convincing evidence that this colonization was initiated by Rome. The results obtained from our case studies show that the landscape archaeology approach offers valuable new insights into early colonization. First, it gives a clear picture of the variety in the geographical contexts in which colonization took place and its effects on the landscape. This has helped convince us that ‘models of colonization’ are not very helpful in thinking about colonization in this period, as each case seems to be unique, in archaeological terms at least. A second important aspect of our approach is that it allows the possibility to investigate the longer term consequences of colonization without necessarily making direct links between archaeological data and historical data on colonization events. While these consequences were often limited, we have argued that the effects of colonization on the rural landscape grew more profound and were more lasting in the 4th c. B.C. Having argued that a general colonization model does not do justice to the complexities of early colonization in the Pontine region, we suggest that an important step in future research should be the study of individual colonies in relation to their territories. In the Pontine region, there is an important gap in our knowledge about the early phases of Circeii and Tarracina and their rural territories.107 New fieldwork in the territories around these two towns seems warranted. Moreover, we have noted that the only case in which the impact of colonization on a colonial site is visible is Norba, and there it has an obviously religious connotation. Therefore it would seem worthwhile to pay more attention to the role of sanctuaries in early colonization, asking questions about the effect of colonization on Archaic cult places and the role of new cult places in the settlement configuration of the Post-Archaic and Mid-Republican period (see on this Stek in this volume, with previous bibliography).108 Finally, one fundamental field of study to be pursued in more depth is the diachronic and spatial

107

We have no archaeological insights at all into the early colonial phase of Circeii, while the role of the sanctuary at Monticchio in the Late Archaic and Early Republican period remains incomprehensible without a settlement context of some kind. See Torelli 1999, 22 for a discussion of the colony of Circeii as a military stronghold. The role of Archaic Tarracina/Anxur and its development during the Republican period is

108

equally obscure. So far, no intensive field surveys have been carried out in the territories of these settlements and Giuseppe Lugli’s inventories (Lugli 1926; id. 1928) only report Late Republican and Imperial remains. The role of sanctuaries in shaping the colonial environment in colonies in Latium and Central Italy is treated in the Ph.D research of the third author.




modeling of the extent of colonial territories and the ager Romanus on the basis of archaeological, environmental and historical data.109 This will greatly improve our understanding of Roman colonial geography in all its aspects: military, socio-economic and ideological.

Acknowledgements The first author thanks the organizers Tesse Stek and Jeremia Pelgrom for their invitation to participate in the workshop. The present article is based on the paper held at the workshop by the first author, titled The Pontine region, laboratory of Roman colonization? but incorporates current research by the second and third authors on the topic.

References Alessandri, L. 2007. L’occupazione costiera protostorica del Lazio centromeridionale (Oxford). Alessandri, L. 2009. Il Lazio centromeridionale nelle età del Bronzo e del Ferro (Ph.D thesis, Rijksuniversiteit Groningen). Attema, P. A. J. 1991. “The Contrada Casali, an intensive survey of a new Archaic hilltop settlement in the Monti Lepini, South Lazio,” Meded 50, 7-62. Attema, P. A. J. 1993. An archaeological survey in the Pontine region (Groningen). Attema, P. A. J. 2000. “Landscape archaeology and Livy: warfare, colonial expansion and town and country in Central Italy of the 7th to 4th c. BC,” BABesch 75, 115-26. Attema, P. A. J. 2001. “Ritual, economy and early roman colonisation in Lazio: colonial conjectures on a late Archaic sanctuary in the ager of Setia,” Caeculus 4, 69-80. Attema, P. A. J. 2005. “Early urbanization between 800 and 600 BC in the Pontine Region (South Lazio), the Salento Isthmus (Apulia), and the Sibaritide (northern Calabria),” in R. Osborne and B. Cunliffe (edd.), Mediterranean Urbanization 800-600 BC (Oxford) 113-42.

109



See Lanjouw 2011 for such a territorial study on Norba.

Attema, P. A. J. and T. De Haas 2012. “Intensive onsite artefact survey and proto-urbanization, case studies from Central and South Italy,” in F. Vermeulen et al. (edd.), Urban landscape survey in Italy and the Mediterranean (Oxford) 1-12. Attema, P. A. J. and M. Van Leusen 2004. “The early Roman colonization of South Lazio: a survey of three landscapes,” in P. A. J. Attema (ed.), Centralization, early urbanization and colonization in first millennium B.C. Italy and Greece. Part 1: Italy (Leuven) 157-95. Attema, P. A. J. and G. Van Oortmerssen 1997-98. “Ceramics of the first millennium BC from a survey at Lanuvium in the Alban hills, central Italy: method, aims and first results of regional fabric classification,” Palaeohistoria 39-40, 413- 39. Attema, P. A. J., G.-J. Burgers and M. Van Leusen 2010. Regional pathways to complexity: settlement and land-use dynamics in early Italy from the Bronze Age to the Republican period (Amsterdam). Attema, P. A. J., T. De Haas and G. Tol 2009-10. “The Astura and Nettuno surveys of the Pontine Region Project (2003-2005): 2nd and final report,” Palaeohistoria 51-52, 169-327. Attema, P. A. J., T. De Haas and G. Tol 2011.


Between Satricum and Antium. Settlement dynamics in a coastal landscape in Latium Vetus (Leuven). Attema, P. A. J. et al. 1992. “Il sito di Borgo le Ferriere ‘Satricum’ nei secoli V e IV a.C.,” in S. Quilici Gigli (ed.), I Volsci (Rome) 75-86. Attema, P. A. J. et al. 2001-2. “Pottery classifications: ceramics from Satricum and Lazio, Italy, 900-300 BC,” Palaeohistoria 43-44, 321-96. Attema, P. A. J. et al. 2007-8. “The Astura and Nettuno surveys of the Pontine Region Project (2003-2005): 1st interim report,” Palaeohistoria 49-50, 415-516. Attema, P. et al. forthcoming. In search of the Archaic countryside. Different scenarios for the ruralisation of Satricum and Crustumerium,” in P. Lulof and C. Smith (edd.), The age of Tarquinius Superbus. Ancient history, Archaeology and Methodology. Bandelli, G. 1995. “Colonie e municipi dall’età monarchica alle guerre sannitiche,” Eutopia 4 2, 143-97. Bellelli, V. 2003. “Appunti sul gorgoneion di Norba,” in L. Quilici and S. Quilici Gigli (edd.), Santuari e luoghi di culto nell’Italia antica (Rome) 385-98. Bintliff, J. (ed.) 1991. The Annales School and Archaeology (Leicester). Bouma, J. W. 1996. Religio votiva: the archaeology of Latial votive religion: the 5th-3rd c. BC votive deposit south west of the main temple at “Satricum” Borgo Le Ferriere, 3 vols (Groningen). Bouma, J. W. 2001. “Bouma, J. 2001. “Understanding a local economy: a 5th-3rd c. BC votive deposit at Satricum, Borgo le Ferriere (Italy),” Caeculus 4, 57-68. Bradley, G. 2006. “Colonization and identiy in Republican Italy,” in G. Bradley and J.-P. Wilson (edd.), Greek and Roman colonization: origins, ideologies and interactions (Swansea) 161-87. Brandizzi Vittucci, P. 1968. Cora (Rome). Brandizzi Vittucci, P. 2000. Antium: Anzio e Nettuno in epoca romana (Rome). Bruckner, E. C. 1995. “Forum Appi,” in Tra Lazio e Campania: ricerche di storia e di topografia antica (Naples) 189-221.

Bruckner, E. C. 2000. “Le fortificazioni di Setia,” in L. Quilici and S. Quilici Gigli (edd.), Fortificazioni antiche in Italia: età repubblicana (Rome) 103-26. Bruckner, E. C. 2003. “Considerazioni sui culti e luoghi di culto a Setia e nel suo territorio in età repubblicana ed imperiale,” in L. Quilici and S. Quilici Gigli (edd.), Santuari e luoghi di culto nell’Italia antica (Rome) 75-98. Cancellieri, M. 1975. “Un sepolcro romano a Mesa,” BLazioMerid 8 2, 5-20. Cancellieri, M. 1987. “La media e bassa valle dell’Amaseno, la via Appia e Terracina: materiali per una carta archeologica,” BLazioMerid 12, 41-104. Carfora, P. and D. Nonnis 2009. “Il bacino con dedica a Diana dall’acropoli maggiore di Norba: un riesame,” in L. Quilici and S. Quilici Gigli (edd.), Atlante tematico di topografia antica (Rome) 183-92. Cassieri, N. 2004. “Il deposito votivo di Tratturo Caniò a Sezze,” in G. Papi (ed.), Religio Santuari ed ex voto nel lazio meridionale (Terracina) 162-81. Càssola, F. 1988. “Aspetti sociali e politici della colonizzazione,” DialArch 6 (2), 5-17. Chiabà, M. 2006. “Da Σιγνούριον Σιγλιουρία (508 a.C.) a Velitrae (494 a.C.): note sulla colonizzazione del Lazio fra la caduta della monarchia e la sottoscrizione del foedus Cassianum,” in M. Faraguna and V. Vedaldi Iasbez (edd.), Dynasthai didaskein: studi in onore di Filippo Càssola per il suo ottantesimo compleanno (Trieste) 91-110. Chiabà, M. 2011. Roma e le priscae Latinae coloniae: ricerche sulla colonizzazione del Lazio dalla costituzione della repubblica alla guerra latina (Trieste). Chiarucci, P. 1989. Anzio archeologica (Rome). Cifarelli, F. M. 1993. “Primi dati per la storia della colonia di età regia a Segni,” RTopAnt 3, 157-62. Coarelli, F. 1990. “Roma, i Volsci e il lazio antico,” in AA.VV., Crise et transformation des sociétés archaïques de l’Italie antique au Ve siècle av. J.C.: actes de la table ronde organisée par l’École française de Rome et l’Unité de recherches étrusco-italiques associée au CNRS (UA 1132): Rome 19-21 novembre 1987 (Rome) 135-54.




Coarelli, F. 1992. “Colonizzazione e municipalizzazione. Tempi e modi,” DialArch 10, 21-30. Cornell, T. J. 1995a. The beginnings of Rome. Italy and Rome from the Bronze Age to the Punic Wars (c. 1000 - 264 BC) (London). Cornell, T. 1995b. “Warfare and urbanization in Roman Italy,” in T. Cornell and K. Lomas (edd.), Urban society in Roman Italy (London) 121-34. De Haas, T. 2011. Fields, farms and colonists: intensive field survey and early Roman colonization in the Pontine region, central Italy (Ph.D thesis, Rijksuniversiteit Groningen). De Haas, T., G. Tol and P. A. J. Attema 2012. “Polygonal masonry platform sites in the Lepine mountains (Pontine Region, Lazio, Italy),” Palaeohistoria 53-54, 195-281. De Meis, A. 1984. “Nuovo materiale della necropoli protostorica di Anzio,” BullCom 87, 237. De’ Spagnolis, M. 2003. “Una fibula protostorica da Norba,” in J. Rasmus Brandt, X. Dupré Raventós and G. Ghini (edd.), Lazio e Sabina, vol. 1: Primo Incontro di studi sul Lazio e la Sabina: atti del Convegno : Roma 28-30 gennaio 2002 (Rome) 65-66. Egidi, R. and A. Guidi 2009. “Anzio: saggi di scavo sul Vallo Volsco,” in G. Ghini (ed.), Lazio e Sabina, vol. 5: Quinto incontro di studi sul Lazio e la Sabina: atti del Convegno, Roma 3-5 dicembre 2007 (Rome) 355-61. Feiken, H. et al. 2012. “Reconstructing a Bronze Age hidden landscape: geoarchaeological research at Tratturo Caniò (Italy, 2009),” Palaeohistoria 53-54, 109-59. Forsythe, G. 2005. A critical history of early Rome: from prehistory to the first Punic War (Berkeley). Gierow, P. 1961. “La necropoli laziale di Anzio,” BPI 69-70, 243-57. Ginge, B. 1996. Northwest necropolis, southwest sanctuary and acropolis: excavations 19071910 (Amsterdam). Gnade, M. 1992. The southwest necropolis of Satricum: excavations 1981 - 1986 (Amsterdam). Gnade, M. 1999. “La ricerca sull’agger di Satricum,” TerraVolsci 2, 31-50.



Gnade, M. 2002. Satricum in the Post-Archaic period: a case study of the interpretation of archaeological remains as indicators of ethnocultural identity (Leuven). Gnade, M. 2003. “Satricum: la prosecuzione delle ricerche,” in J. Rasmus Brandt, X. Dupré Raventos and G. Ghini (edd.), Lazio e Sabina, vol. 1: Primo Incontro di studi sul Lazio e la Sabina: atti del Convegno: Roma 28-30 gennaio 2002 (Rome) 213-20. Gnade, M. 2006. “La ventottesima campagna di ricerca a Satricum dell’Università di Amsterdam nel 2004,” in G. Ghini (ed.), Lazio e Sabina, vol. 3: Terzo incontro di studi sul Lazio e la Sabina: atti del convegno: Roma 18-20 novembre 2004 (Rome) 255-60. Gnade, M. (ed.) 2007. Satricum: trenta anni di scavi olandesi (Amsterdam). Gnade, M. 2009. “La ricerca a Satricum dell’Università di Amsterdam nel 2007,” in G. Ghini (ed.), Lazio e Sabina, vol. 5: Quinto incontro di studi sul Lazio e la Sabina: atti del Convegno, Roma 3-5 dicembre 2007 (Rome) 363-68. Heldring, B. 2007. “Il deposito votivo III: una cisterna prima, un deposito votivo dopo,” in M. Gnade (ed.), Satricum: trenta anni di scavi olandesi (Amsterdam) 78-81. Humm, M. 1996. “Appius Claudius Caecus et la construction de la Via Appia,” MÉFRA 108 2, 693746. Jaia, A. M. 2002. “Antium,” StEtr 65-68, 490-93. Jaia, A. M. 2004. “I luoghi di culto del territorio di Anzio,” in G. Ghini (ed.), Lazio e Sabina, vol. 2: Secondo incontro di studi sul Lazio e la Sabina: atti del convegno: Roma 7-8 maggio 2003: (Estratto) (Rome) 255-64. Jaia, A. M. et al. 2007. “Viale delle Roselle: deposito votivo,” in A. M. Jaia and F. Di Mario (edd.), Capolavori ritrovati dal museo nazionale romano: villa Adèle - Anzio, 7 dicembre 2006 (Anzio) 9-30. Knapp, A. (ed.) 1992. Archaeology, Annales and Ethnohistory (Cambridge). Knoop, R. and P. Lulof 2007. “L’architettura templare,” in M. Gnade (ed.), Satricum: trenta anni di scavi olandesi (Amsterdam) 32-42. Lanjouw, T. J. R. 2011. Territoria en territorialiteit


in Latium Vetus (Centraal Italië) gedurende de Archaïsche periode tot en met de Republiek en de ruimtelijke kenmerken en organisatie van het territorium van Latijnse stadsstaat/kolonie Norba. (MA thesis, Rijksuniversiteit Groningen). Louwaard, M. 2007. “L’acropoli: l’edificio di età medio-repubblicana,” in M. Gnade (ed.), Satricum: trenta anni di scavi olandesi (Leuven) 75-77. Lugli, G. 1926. Regio 1.: Latium et Campania vol. 1: 1: Ager Pomptinus: “Pars prima”Anxur Tarracina (Rome). Lugli, G. 1928. Regio 1.: Latium et Campania vol. 1: 2: Ager Pomptinus: “Pars secunda” Circeii (Rome). Lulof, P. 2011. “The Late Archaic miracle: roof decoration in Central Italy between 510 and 450 B.C.,” in P. Lulof and C. Rescigno (edd.), Deliciae Fictiles IV: architectural terracottas in ancient Italy. Images of gods, monsters and heroes (Oxford) 23-31. Maaskant Kleibrink, M. 1987. Settlement excavations at Borgo le Ferriere ‘Satricum’, vol. 1: The campaigns 1979, 1980, 1981 (Groningen). Maaskant Kleibrink, M. 1992a. Settlement excavations at Borgo le Ferriere ‘Satricum’, vol. 2: The campaigns 1983, 1985 and 1987 (Groningen). Maaskant Kleibrink, M. 1992b. “Gli scavi più recenti svolti a Borgo le Ferriere (Satricum),” in S. Quilici Gigli (ed.), I Volsci (Rome) 53-64. Maassen, J. 2006. “Posta di Mesa - een Republikeins heiligdom langs de Via Appia (Italië),” Paleoaktueel 17, 122-29. Mangani, E. 2004. “Le stipi votive di Roma e del Lazio nel museo Pigorini,” in G. Papi (ed.), Religio Santuari ed ex voto nel Lazio meridionale (Terracina) 59-83. Mater, B. 2005. Patterns in pottery: a comparative study of pottery production in Salento, Sibaritide and Agro Pontino in the context of urbanization and colonization in the first millennium BC (Ph.D thesis, Vrije Universiteit Amsterdam). Melis, F. and S. Quilici Gigli 1972. “Proposta per l’ubicazione di Pometia,” ArchCl 24, 219-47. Mengarelli, R. and R. Paribeni 1909. “Norma: scavi

sulle terrazze sostenute da mura poligonali presso l’Abbazia di Valvisciolo,” NSc 6, 241-60. Nijboer, A. 1998. From household production to workshops: archaeological evidence for economic transformations, pre-monetary exchange and urbanization in central Italy from 800 to 400 BC (Ph.D thesis, Rijksuniversiteit Groningen). Nijboer, A. et al. 1995. “Notes on artifact and pottery production at Satricum in the 5th and 4th centuries BC,” Meded 54, 1-38. Oakley, S. P. 1997. A commentary on Livy: books VIX, vol. 1: Introduction and book VI (Oxford). Ogilvie, R. 1965. A commentary on Livy, books I-V (Oxford). Palombi, D. 2000. “Intorno alle mura di Cori,” in L. Quilici and S. Quilici Gigli (edd.), Fortificazioni antiche in Italia età repubblicana (Rome) 91-102. Palombi, D. 2003. “Cora: bilancio storico e archeologico,” ArchCl 54, 198-252. Pelgrom, J. 2008. “Settlement organization and land distribution in Latin colonies before the Second Punic War,” in L. de Ligt and S. J. Northwood (edd.), People, land and politics: demographic developments and the transformation of Roman Italy 300 BC-AD 14 (Leiden) 333-72. Pelgrom, J. 2012. Colonial landscapes: demography, settlement organization and impact of colonies founded by Rome (Ph.D thesis, Leiden University). Piccarreta, F. 1977. Astura (Florence). Quilici Gigli, S. 1983. “Sistemi di cuniculi nel territorio fra Velletri e Cisterna,” ArchLaz 7, 112-23. Quilici Gigli, S. 1988. “Insediamenti nel territorio di Norba, il poggio di Serrone di Bove,” ArchLaz 9, 227-32. Quilici Gigli, S. 1996. “Appunti di topografia per la storia di Norba,” AttiPontAcc 66, 285-301. Quilici Gigli, S. 2003. “Trasformazioni urbanistiche ed attività edilizia in epoca repubblicana: il caso di Norba,” Orizzonti 4, 23-32. Quilici Gigli, S. 2004. “Circumfuso volitabant milite volsci Dinamiche insediative nella zona pontina,” in L. Petacco and S. Quilici Gigli (edd.), Viabilità e insediamenti nell’italia antica (Rome) 235-75. Quilici Gigli, S., S. Ferrante and C. Caputo 2003.




“Norba: l’acropoli minore e i suoi templi,” in L. Quilici and S. Quilici Gigli (edd.), Santuari e luoghi di culto nell’Italia antica (Rome) 288-327. Quilici, L. 1991. “Un esempio di ingegneria stradale sulle montagne di Norba,” in M. Gnade (ed.), Stips Votiva: papers presented to C. M. Stibbe (Amsterdam) 149-55. Quilici, L. 2004. “Caprifico di Cisterna di Latina: una città arcaica nella pianura pontina,” Ocnus 12, 247-62. Quilici, L. and S. Quilici Gigli 1987. “L‘abitato di Monte Carbolino,” ArchLaz 8, 259-77. Quilici, L. and S. Quilici Gigli 1988. “Ricerche su Norba,” ArchLaz 9, 233-56. Quilici, L. and S. Quilici Gigli 2001. “Sulle mura di Norba,” in L. Quilici and S. Quilici Gigli (edd.), Fortificazioni antiche in Italia età repubblicana (Rome) 181-244. Quilici, L. and G. Tognon 2001. “Sul calcestruzzo della strada che da Norba scende alla piana Pontina,” in L. Quilici and S. Quilici-Gigli (edd.), Fortificazioni antiche in Italia età repubblicana (Rome) 245-50. Rescigno, C. 2003. “Norba: santuario di Giunone Lucina, appunti topografici,” in L. Quilici and S. Quilici Gigli (edd.), Santuari e luoghi di culto nell’Italia antica (Rome) 329-51. Ross Taylor, L. 1960. The voting districts of the Roman Republic: the thirty-five urban and rural tribes (Rome). Savignoni, L. and R. Mengarelli 1901. “Norba: relazione sopra gli scavi eseguiti nell’estate dell’anno 1901,” NSc, 514-59. Savignoni, L. and R. Mengarelli 1903. “Norba: relazione sopra gli scavi eseguiti a Norba nell’estate dell’anno 1902,” NSc, 229-62. Schmiedt, G. and F. Castagnoli 1957. L’antica città di Norba: documentazione aerofotogrammetrica (Florence). Solin, H. 1999. “Epigrafia repubblicana: bilancio, novità, prospettive,” in XI Congresso



internazionale di epigrafia greca e latina, Roma, 18-24 settembre 1997, vol. 1 (Rome) 379-404. Stibbe, C. M. 1987. “Satricum e Pometia: due nomi per la stessa città?,” Meded 47, 7-16. Stobbe, J. 2007. “L’architettura intorno ai templi,” in M. Gnade (ed.), Satricum: trenta anni di scavi olandesi (Amsterdam) 43-50. Termeer, M. K. 2010. “Early colonies in Latium (ca. 534-338 BC): a reconsideration of current images and the archaeological evidence,” BABesch 85, 43-58. Tol, G. et al. 2014. “Minor Centres in the Pontine Plain. The cases of Forum Appii and Ad Medias,” PBSR 82, 109-34. Torelli, M. 1999. “Religious aspects of early Roman colonization,” in M. Torelli (ed.), Tota Italia: essays in the cultural formation of Roman Italy (Oxford/New York) 14-42. Van ‘t Lindenhout, E. 2010. Bouwen in Latium in de Archaïsche periode (Ph.D thesis, Rijksuniversiteit Groningen). Van Leusen, M. 2010. “Tratturo Caniò (Tempio di Giunone),” Fasti Online, http://www.fastionline.org /micro_view.php?fst_cd=AIAC_2454&curcol=sea _cd-AIAC_3262. Van Leusen, M. et al. 2003-2004. “Protohistoric to Roman settlement on the Lepine margins near Ninfa (south Lazio, Italy),” Palaeohistoria 4546, 301-46. Van Royen, R. 1992. “Appendix B: ancient sources on the first decade of the Volscian presence in fifth-century Latium (509-483 B.C.),” in M. Gnade (ed.), The southwest necropolis of Satricum (Amsterdam) 437-53. Walsh, K., P. Attema and T. de Haas 2014. “The Pontine Marshes (Central Italy): a case study in wetland historical ecology,” BABesch 89, 27-46 Waarsenburg, D. J. 1995. The Northwest necropolis of Satricum (Amsterdam). Zaccheo, L. and F. Pasquali 1972. Sezze dalla preistoria all’età romana (Sezze).


Le colonie di Luceria e Venusia. Dinamiche insediative, urbanizzazione e assetti agrari Maria Luisa Marchi

Tra romanizzazione e colonizzazione: problemi di identità culturale Le dinamiche legate ai processi della prima romanizzazione hanno costituito, nell’ultimo ventennio, un grande tema di studio,1 e alla luce dell’ampio dibattito che ne è scaturito appare chiaro soprattutto che la comprensione del fenomeno dal punto di vista storico non può prescindere dalla ricostruzione archeologica del sistema insediativo e del paesaggio antico. Nello specifico per l’ambito italico, il tema della conquista romana e dell’assimilazione dei fattori culturali legati a essa è stato oggetto di grande interesse per storici e archeologi,2 ma l’ampiezza delle problematiche sembra lasciare ancora ampio spazio ad analisi di diverso tipo, in particolare in riferimento alla comprensione di evoluzione e trasformazioni dell’assetto territoriale e insediativo (si vedano i contributi di Attema et al. e Bellini et al. in questo volume) La ricostruzione storica di un periodo così complesso non può infatti esaurirsi in termini di semplice opposizione tra due sistemi culturali statici e compatti: Magna Grecia e mondo indigeno da un lato e Roma dall’altro; al contrario si deve porre sempre più attenzione alle trasformazioni progressive, agli elementi di cambiamento e di persistenza, all’individuazione delle situazioni intermedie, evitando generalizzazioni interpretative riconducibili a una drastica antitesi tra ‘continuità’ e ‘discontinuità’. Questi temi sono stati oggetto di un ampio dibattito internazionale, che si è focalizzato sul tentativo di destrutturazione del principio della romanizzazione, e la critica più recente, perlopiù di matrice anglosassone, ha investito il termine stesso, proponendo variabili che attenuassero il concetto dell’acculturazione romana,3 ma in una certa misura anche la dinamica e la portata del fenomeno, con l’introduzione della definizione di bricolage per indicare il fenomeno dell’assimilazione romana, a sottolinearne la frammentazione e la casualità.4 In definitiva l’obiettivo è stato di ribaltare il ruolo subalterno delle popolazioni indigene

1

2

L’introduzione nella storiografia sulla romanizzazione delle tematiche di tipo antropologico e l’arricchimento della ricerca con le indagini sul campo hanno consentito una progressiva evoluzione di tale analisi. Nel Convegno di Pontignano del 1969 (atti in Incontro di studi 1970-71) si trovano a confronto archeologi e storici, e si introduce così un filone di studi che nei decenni successivi troverà ampio spazio. Sulla romanizzazione vista come acculturazione e scambio culturale, cfr. Modes de contacts et processus 1983. Per una sintesi vedi Desideri 1991, 577-626. Un punto di riferimento fondamentale su queste problematiche è costituito dal convegno sulla romanizzazione del Sannio nel 1988 (atti in La Romanisation 1991). Un vero caposaldo, ma anche

3

4

punto di partenza per le analisi nei contesti meridionali, è offerto dal Convegno svoltosi a Venosa nel 1987 (Salvatore 1990) e sulla stessa linea si pongono quello di poco successivo, tenutosi ad Acquasparta (Tagliente 1990) e il Convegno di Bruxelles e Roma (Mertens e Lambrechts 1991). Un ampio livello di approfondimento si ha con le sintesi David 1994 e del Vallat 1995. In alcuni casi la discussione si è focalizzata sulla destrutturazione del concetto di ‘romanizzazione’ puntando l’attenzione sull’accento culturale delle popolazioni indigene: Curti 2001, con ampia bibliografia precedente; cfr. in generale Keay e Terrenato 2001; Van Dommelen e Terrenato 2007. Per una sintesi vedi i contributi in Colivicchi 2011a. Terrenato 1998.




sottovalutando la portata dell’organizzazione e della pianificazione politica dei Romani.5 Si è andato inoltre sempre più affermando anche un modello di analisi cosiddetto ‘antropologico’ legato alla rivalutazione delle entità locali mai del tutto cancellate. Dall’analisi dell’impatto del mondo romano con le popolazione preesistenti, e soprattutto riguardo l’acquisizione dei modelli romani da parte delle popolazioni locali secondo parametri ‘spontanei’, è stato messo in evidenza un fenomeno comunemente definito di ‘autoromanizzazione’, o meglio self-romanisation.6 Non c’è peraltro dubbio che il problema è stato giustamente visto dal punto di vista opposto, e cioè dell’assimilazione degli elementi indigeni da parte dei Romani su linee di reale inversione dei termini della questione.7 Per quanto riguarda il mondo dauno, una lunga tradizione di studi ci offre un esauriente panorama storico del problema.8 Nell’ambito di studi più generali è noto come Arnold Toynbee tracci un quadro della crisi sociale ed economica, particolarmente grave nell’Apulia postannibalica,9 e Peter Brunt si sofferma, a ragione, sulle questioni di base legate alle problematiche economiche e produttive.10 Ma solo l’analisi delle dinamiche archeologiche ha permesso di avviare un filone di riflessioni sulle problematiche del processo di penetrazione romana in Daunia, soprattutto dal punto di vista sociale e culturale.11 All’ambito più schiettamente storico-archeologico della Daunia richiama il volume di Enzo Lippolis e Marina Mazzei, che privilegiano l’ottica dello studio dell’area dal punto di vista cronologico e offrono un quadro documentario dall’ellenizzazione all’età romana.12 L’intervento di Marina Mazzei, Joseph Mertens e Giuliano Volpe al Convegno di Venosa del 1990 ribadisce l’attenzione sul processo di romanizzazione, affrontato sui due piani paralleli delle trasformazioni urbanistiche e dell’articolazione del paesaggio agrario.13 Una sintesi assai esauriente è senza dubbio offerta dal lavoro di Volpe, che costituisce un preciso punto di partenza14 nella ricerca nell’area, da integrarsi con successivi approfondimenti che si sono avuti con i lavori sul comprensorio venosino e lucerino di recente acquisizione.15 Le indagini avviate sul territorio negli ultimi anni16 stanno peraltro apportando dati assolutamente innovativi soprattutto per quanto riguarda la situazione precedente all’arrivo dei Romani, offrendo un ampio panorama proprio sul tema del passaggio e delle trasformazioni dall’abitato alla città.17 Appare quindi essenziale, alla luce delle più recenti acquisizioni e delle indagini condotte negli ultimi decenni,18 verificare come risulti mutato

5

6 7 8

9



Sul tema applicato all’Italia si veda la sintesi in Stek 2009, 9-16 con riferimenti bibliografici. Torelli 1990, 169-70. Coarelli 1991, 177-78; Marchi 2000. Si veda il dettagliato profilo storico, dedicato alla penetrazione romana in Apulia, tracciato da Lepore 1963 sulla base di fonti letterarie, spesso scarse e difficilmente interpretabili. Anche se ricca di spunti, risulta però ipotetica la ricostruzione dell’Apulia proposta da Sirago 1963 e Sirago 1993 basata unilateralmente sulla documentazione letteraria, che vede nella romanizzazione della regione un fattore di profonda crisi e irreversibile decadenza; i lavori più significativi sono comunque i numerosi contributi di Francesco Grelle (Grelle 1992; id. 1995; id. 2008). Esaurienti sintesi ha presentato in varie sedi e in diversi tempi Mario Torelli (Torelli 1984; id. 1990; id. 1992a; id. 1999). Toynbee 1965.

10 11 12 13 14 15

16

17 18

Brunt 1971. Torelli 1984, 325-29; id. 1991; id. 1992b; id, 1992c. Mazzei 1984. Mazzei, Mertens e Volpe 1990, 177-95. Volpe 1990. Marchi e Sabbatini 1996; Marchi e Salvatore 1997; Marchi 1999. Volpe, Romano e Goffredo 2003; iid. 2004; Goffredo 2008, 287-301; Goffredo 2011; Marchi 2010a; ead. 2008a, 26786; in riferimento a contesti limitrofi: per il territorio di Brindisi Burgers 1994; 145-54; id. 1995; id. 1998; Aprosio 2008. Per l’Oria Project si veda Boersma e Yntema 1982, iid. 1987; Yntema 1993; sul Valesio Project si veda Boersma, Burgers e Yntema 1991, 115-31. Marchi 2000; ead. 2008b; ead. 2010a. Per la valle dell’Ofanto e il territorio canosino cfr. Goffredo 2011, 287-301; per il territorio brindisino Aprosio, 2008, 107; per l’area melfese Marchi e Sabbatini 1996; Sabbatini 2000; Marchi 2010a.


il panorama delle conoscenze in relazione alle dinamiche di trasformazione e ai processi di acquisizione, sostituzione o persistenza con la possibilità di leggere il passaggio tra IV e III sec. a.C. come momento di trasformazione. L’intervento romano nei territori italici, rappresentato dalla colonizzazione, va analizzato in rapporto alle situazioni preesistenti e al loro adattamento al nuovo sistema di pianificazione introdotto dai Romani (fig. 1).19 Pur non potendo escludere ampie devastazioni ed emarginazioni delle preesistenze, è possibile verificare forme di persistenza nell’occupazione territoriale da parte delle popolazioni indigene. Si può pertanto definire un quadro insediativo in lenta evoluzione che sembra concludersi definitivamente solo dopo la guerra sociale, se non addirittura in età augustea, quando la sistematizzazione giuridica unifica le pianificazioni agrarie e territoriali all’interno di sistemi centuriali.20 FIG. 1. La Daunia con i principali centri indigeni: in evidenza le colonie latine di Venusia e Luceria.

L’intervento romano in Daunia L’espansione del dominio romano investe la Puglia settentrionale nel corso delle guerre sannitiche, a partire dal terz’ultimo decennio del IV secolo, e si estende progressivamente verso il mezzogiorno.21 La comparsa di Roma in Daunia avviene in modo progressivo, come premessa alla penetrazione in tutta la regione, ma la prima presenza ufficiale è concordemente indicata nel 326 a.C., anno in cui le fonti collocano una richiesta di intervento da parte dei principes dauni,22 a seguito della minaccia delle pressioni osche. L’alleanza con le popolazioni apule fu per i Romani l’occasione di aggirare il nemico sannitico. Nel 318 a.C. gran parte dei territori apuli occupati dai Sanniti era stata conquistata e in quell’anno Teanum Apulum e Canosa si arresero23 chiedendo un foedus, e nel 317 a.C.,24 essendo stata conquistata anche Forentum, Livio ci riferisce che tutta l’Apulia era in mano romana.25 In realtà una parte non piccola della regione era ancora sotto controllo dei Sanniti. Particolarmente articolata e difficile fu la presa di Luceria, considerata esplicitamente dalla tradizione di Diodoro come la città di maggior rilievo, la cui importanza è confermata dall’episodio della sconfitta delle forche caudine causata nel 321 a.C.26 dalla necessità di raggiungere rapidamente Luceria, minacciata dalla pressione sannitica. Un presidio vi era stato posto, a dire delle fonti già

19

20 21 22

Gabba 1977; Gabba 1979, 38-43; per un diverso punto di vista cfr. Colivicchi 2011a. Capogrossi Colognesi 2002, 6-14. Grelle 2008, 365. Narra a tal proposito Livio (8.25.3) che durante il consolato di C. Petelio e di L. Papirio Mugellino “Lucani atque Apuli, quibus gentibus nihil ad eam diem cum Romano populo fuerat, in fidem venerunt arma virosque ad bellum pollicentes”. L’alleanza sembra essere durata comunque poco, prima i Lucani e pochi anni dopo secondo le fonti anche gli Apuli ruppero

23 24 25 26

l’alleanza (Liv. 8.27. 6-11; 8.37.3-4). Anche se alcuni autori non accettano la defezione degli Apuli ritenendoli sempre alleati, la tradizione liviana sembra essere la più accreditata e a essa si richiamano anche i Fasti trionfali che ricordano nel 322 a.C. un trionfo di Q. Fabio Massimo Rulliano sui Sanniti e sugli Apuli (I.I 13.1, Fasti triumph. cap. 12). Liv. 9.20. 4. Liv. 9.20.7. Liv. 9.20. Liv. 9.2.6-8.




a

b

c

d

e

f

FIG. 2. Insediamenti dauni: confronto planimetrico tra l’area di occupazione in età arcaica (sinistra) e in età romana (destra). A: Arpi; B: Herdonia; C: Lavello-Forentum; D: Ascoli Satriano; E.: Canusium; F. Bantia (i blocchi della griglia misurano 1 x 1 km).

nel 325 a.C., ma la vera conquista della città, primo avamposto romano in Apulia, si realizzò solo nel 315-314 a.C. con la deduzione della colonia latina.27 L’obiettivo fu la creazione di un cordone di contenimento all’espansione sannitica, stabilito da Roma attraverso i rapporti di alleanze e le deduzioni coloniali, che si potrà considerare concluso solo qualche decennio più tardi, quando una colonia latina fu insediata a Venusia.28 La deduzione delle due colonie di diritto latino con la pianificazione delle città, l’organizzazione di vastissime partizioni territoriali assegnate ai coloni, la realizzazione di un complesso e articolato sistema stradale legato alla centuriazione produrrà profonde modificazioni nel sistema insediativo della regione. L’articolazione di questo modello insediativo risulta abbastanza particolare: si fonda su abitati, ormai ben conosciuti,29 ancora lontani dalle forme urbane ma non ricollegabili semplicemente al concetto di villaggio. Questi insediamenti risultano articolati in aggregati di varie dimensioni, che si alternano, in modo irregolare, a spazi vuoti e ad aree di sepolture (fig. 2).30 In definitiva sono organizzati come un insieme di nuclei abitativi, sparsi su vaste aree (tra 200 e 1000 ettari), all’interno dei quali si alternano i gruppi delle abitazioni, caratterizzate nelle fasi più antiche da capanne e sostituite poi da edifici a pianta quadrangolare, e di sepolture. Questi abitati sono generalmente privi di sistemi difensivi e solo in rari casi presentano aggeri che racchiudono spazi molto ampi.31

27 28 29 30



Liv. 9.26.1-5; Diod. Sic. 19.72.8-9. Marchi e Salvatore 1997. Marchi 2008a, 267-86; ead. 2009b, 327-67. Bottini 1982, 154; id. 2001, 109-16; id. 2013, per una sintesi Marchi 2009a; ead. 2010a.

31

De Juliis 1984, 183, n. 54; Marchi 2000; ead. 2008b; ead. 2009a. Di questi abitati solo alcuni erano dotati di sistema difensivo, come Arpi, forse Canusium, Tiati e Bantia; altri ne erano privi, basti ricordare i casi di Lavello-Forentum, Ascoli Satriano.


Le indagini archeologiche hanno messo in evidenza, oltre agli abitati più estesi già noti, quali Arpi, Ascoli Satriano, Teanum Apulum, Lucera, Canosa, Herdonia (fig. 2), un sorprendente numero di insediamenti minori, anch’essi diffusi su ampi sistemi collinari affacciati sulle principali vallate fluviali (Ofanto, Celone, Carapelle, Candeloro),32 finora ignoti. Il fenomeno sembra riscontrarsi in tutte le aree fin qui prese in esame, e appare piuttosto diffuso in tutta la Daunia. Tali centri, come ho già ipotizzato, potrebbero meglio chiarire la notizia di Appiano33 relativa a numerosi villaggi conquistati dai Romani. L’intervento romano sembra aver favorito il rapido emergere di classi dominanti locali, di cui abbiamo manifestazione negli ipogei canosini e nelle sontuose case di Arpi. A esse forse si devono le modificazioni degli insediamenti che tendono ad acquisire sempre più connotati urbani. Il passaggio alla definizione di veri e propri centri urbani si ebbe solo in alcuni insediamenti, probabilmente quelli interessati da un livello economico più elevato e da presupposti politici adeguati, mentre per alcuni centri minori si verifica un progressivo abbandono. Si può in realtà specificare che nei casi di più nuclei relativi a un unico agglomerato si verifica una concentrazione in corrispondenza di quello principale, da cui si genera il centro urbano, e la relativa scomparsa di quelli minori e più periferici. In sintesi appare piuttosto omogeneo il quadro dell’organizzazione insediativa dei principali centri dell’area dauna. A una situazione di abitato disorganico su ampie superficie sembra sostituirsi, perlopiù tra la fine del IV e nel corso del III secolo a.C., un generalizzato ridimensionamento delle aree insediative a cui corrisponde una organizzazione degli spazi anche a livello funzionale. Più in generale gli effetti della romanizzazione sono riscontrabili nell’assimilazione di modelli istituzionali, culturali, economici e architettonici verificabili a diversi livelli nei vari centri, in ambito insediativo nell’acquisizione di una forma insediativa di tipo urbano anche in modo autonomo, come nel caso di Herdonia e Bantia, con buona probabilità dietro la spinta della presenza delle colonie latine. Un chiaro indizio dell’assimilazione di elementi locali da parte dei Romani si riscontra nell’impostazione planimetrica degli edifici abitativi nelle colonie, ad esempio a Venosa, dove le case dei coloni sono del tutto identiche a quelle delle case indigene. In modo analogo, anche le tecniche costruttive di questi edifici sembrano risentire della tradizione locale che vede l’utilizzo anche nelle case dei coloni di sistemi contemporaneamente utilizzati negli abitati indigeni.

Dinamiche del popolamento e degli assetti agrari: le colonie di Luceria e Venusia I due contesti presi in considerazione in questa sede, cioè i territori delle città di Luceria e Venusia, si connotano come i capisaldi del processo di romanizzazione dell’area dauna. Nel paesaggio del Tavoliere, del subappennino dauno e delle colline del melfese, che di questa area sono parti fondamentali, la nascita delle nuove città e l’impianto della pianificazione centuriale ha rappresentato certamente l’elemento più incisivo della sua evoluzione, secondo solo alla grande riforma agraria degli anni ’50 del secolo scorso. L’ager Venusinus Il vasto territorio della colonia di Venusia34 (fig. 3) si estendeva dalla valle dell’Ofanto alle pendici del Vulture su un’area di circa 700 kmq35 la cui ampia varietà di paesaggio, dalle pianure della valle fluviale, ai sistemi collinari,

32

33 34

Favia, Giuliani e Marchi 2007; Marchi 2008a; Volpe, Romano e Goffredo 2003; iid. 2004; Goffredo 2008. App., Sam. 4.1. Questa sintesi è il risultato di un ampio progetto, condotto nel corso di quasi un ventennio (1989-2000) dall’Università di Roma Sapienza e l’Università di

35

Foggia con il coordinamento Paolo Sommella e di chi scrive. Per una sintesi Marchi e Sabbatini 1996; Sabbatini 2001; Marchi 2010a. La ricognizione è stata condotta con metodo estensivo e intensivo su tutto il territorio dell’antica colonia, con la segnalazione di oltre 2000 punti archeologici.




FIG. 3. Territorio della colonia di Venusia.

alle pendici montane, ha favorito il popolamento fin dalla preistoria.36 La colonia latina di Venusia viene dedotta in una zona di confine tra Lucania e Dauna.37 FIG. 4. Area ad Ovest di Venosa: insediamenti sannitici Le fonti ci riferiscono che la città fu assediata e (A. Casarotto). conquistata, essendo sottratta ai Sanniti,38 da uno dei consoli dell’anno 291 a.C., L. Postumio Megello, che provenendo da Cominium, in Irpinia, conquistò “molte altre città” e sterminò un ingente numero di uomini (diecimila).39 Nonostante ciò la deduzione della colonia fu, forse, affidata a uno dei Fabii,40 avversari di Postumio, e, se si può dar fede al passo di Dionigi, vi sarebbero stati inviati 20.000 coloni.41 È assai probabile che nelle varie fasi del conflitto sannitico questo territorio sia stato protagonista del passaggio degli eserciti e di combattimenti, se, come dimostrano i dati archeologici, la penetrazione sannita era giunta fino alle pendici del Vulture. D’altronde l’avanzata verso la Lucania interna, dopo la conquista di Forentum, era avvenuta attraverso percorsi – i precedenti delle vie Appia ed Herculia – che attraversavano questo territorio. Per quanto riguarda la distruzione di un sorprendente numero di insediamenti, di cui si fa cenno nelle fonti, i dati archeologici documentano molti centri minori, che possono essere attribuiti al popolamento sannitico. Questi villaggi, localizzati nella fascia occidentale del comprensorio venosino (fig. 4), dopo le vicende belliche e l’annientamento delle popolazioni sannitiche, vengono distrutti o abbandonati e non ripopolati in età repubblicana. Cessa di vivere anche l’abitato di località Grottapiana (fig. 5), localizzato a nord-est di Venosa, e quello individuato in corrispondenza della cittadina di Forenza. Ma l’esempio più significativo è il villaggio di località Casalini, insediamento daunio databile tra l’VIII e il IV secolo a.C., occupato dai Sanniti nel IV secolo a.C., dove la frequentazione termina, forse anche con una distruzione, nel III secolo a.C., in coincidenza della conquista romana (fig. 6).42

36

37

38 39

40



Per una ricostruzione del paesaggio antico cfr. Marchi 2010a. Sui territori coloniali si vedano i contributi di Pelgrom e di Stek in questo volume. Ancora nei versi di Orazio … Lucanus an Apulus anceps… (Hor., Sat. II.1.34) sembra conservarsene la memoria. Dion. Hal., Ant. Rom. 17-18.5. Cf. i contributi di Bradley e di Terrenato in questo volume. Torelli 1992c, 34 sgg.

41

42

Sul problema Marchi e Sabbatini 1996, 19; Marchi e Salvatore 1997, 8-10. Il villaggio di Casalini presenta un sistema insediativo di tipo daunio, ma nel IV secolo risulta essere stato interessato da presenze sannitiche e va, secondo recenti ipotesi, identificato con la Venosa conquistata dai Romani nel 291 a.C., che le fonti definiscono poluantropon. Cfr. Marchi e Sabbatini 1996, 92-97; Marchi 1999, 112-13; ead. 2004, 133, ead. 2010a.


FIG. 5. Principali insediamenti preromani dell’area melfese.

Una forte prevalenza di continuità insediativa nell’area settentrionale del territorio della colonia, dove parecchie fattorie preromane sopravvivono,43 potrebbe essere imputato al connotato etnico del popolamento; quest’area infatti può essere riferita all’abitato di Lavello-Forentum, e quindi occupata dalle popolazioni daune, alleate dei Romani. Potremmo quindi ipotizzare, facendo riferimento a quel numero elevato di coloni, che alcuni gruppi di abitanti indigeni fossero stati inseriti nel popolamento rurale e che ad essi fosse affidato l’uso di questa fascia di territorio. Analoga situazione si riscontra nell’area del comprensorio di Banzi, centro della Daunia interna, al confine con la Peucezia. Pur presupponendo quindi una persistenza nel territorio di gruppi di popolazione indigena,44 presumibilmente quella dauna, l’organizzazione delle campagne in questi anni è caratterizzata da

43

44

FIG. 6. Pianta dell’abitato di località Casalini (i blocchi della griglia misurano 1 x 1 km).

Il fenomeno appariva evidente anche in zone già precedentemente esaminate del vasto territorio coloniale, cfr. Marchi e Sabbatini 1996; Sabbatini 2001, 70 Marchi e Sabbatini 1996, 19-20.




FIG. 7. Piani di Camera: foto aerea con sovrapposizione delle aree che segnalano la presenza di edifici rurali di età repubblicana distribuiti lungo il percorso centuriale/via Appia.

una radicale ridistribuzione degli impianti rurali; infatti l’intervento romano rappresenta un momento di profonda cesura e di radicale trasformazione nel paesaggio, soprattutto in seguito alla realizzazione del nuovo sistema distributivo degli appezzamenti assegnati ai coloni e alla costruzione di una miriade di nuove fattorie. A tutto ciò non possono non agganciarsi inevitabili cambiamenti nella natura delle colture e del sistema produttivo. Ma se in genere in molti territori gli interventi legati agli impianti delle centuriazioni, dovuti alla realizzazione di grandi opere stradali e idriche, hanno lasciato segni intangibili sul paesaggio,45 in queste aree, così segnate dalle proprietà latifondistiche, da ampie aree boscose e in tempi relativamente più recenti dalle invasive riforme agrarie, rilevare tali tracce risulta piuttosto arduo.46 Risultano scarsamente attendibili i tentativi di ricostruzione degli assetti centuriali basati sulla sola lettura di tracce, peraltro scarsamente condivisibili, presenti su alcune foto aeree.47 Scarse in proposito anche le fonti: le notizie del Liber coloniarum48 ricordano una divisione graccana di difficile identificazione. All’assenza di tracce di assi centuriali intercettabili dalla lettura delle foto aeree o dalla persistenza nella cartografia, si affianca una difficoltà di selezione delle fattorie rinvenute, documentate dalle aeree di

45 46



Dall’Aglio 2004, 17-21. La storia agraria del territorio del comune di Venosa è ben documentata in Angelini, Di Vito e Groia 1990; Angelini 1999; in generale sul comprensorio venosino Marchi e Sabbatini 1996; Sabbatini 2001; Marchi 2010a da ultimo sul paesaggio agrario di queste regioni; Pepe 2005, 8-12 in cui si mette in evidenza come prima della riforma agraria degli anni ’50 del secolo scorso buona parte del territorio fosse boschivo o riservato al pascolo.

47

48

Scarsa attendibilità ha la ricostruzione attualmente priva di riscontri effettuata dal Mario Coppa (Catizzone, Giusteschi e Coppa 1979, 87-128; Coppa 1979, 119-28) già contrastata da Salvatore 1984, 12 che ricollega la divisione agraria individuata con una parcellizzazione della seconda metà del XIX sec. e sottolineata anche in Marchi e Sabbatini 1996, 123 e da ultimo per una sintesi sulla centuriazione graccana in area dauna Bonora Mazzoli 2001, 67. Lib. colon. I.210.7, L.


FIG. 8. Ipotesi di ricostruzione dell’asse centuriale sulla direzione della via Appia. indicato sono anche gli insediamenti di epoca Repubblicana (A. Casarotto).

frammenti fittili, là dove è assai complesso distinguere le stesse cronologicamente sulla base di pochi frammenti ceramici, spesso inquadrabili solo nell’ambito di un intero secolo. Le tracce di occupazione relativa alla prima fase coloniale si sono identificate nel settore a est della colonia, sul vasto altopiano dei Piani di Camera.49 Qui si è riscontrata una distribuzione con orientamento nord-est/sud-ovest all’interno della quale gli insediamenti sono localizzati a una distanza media di circa 200 m l’uno dall’altro. In un’area di circa 63 ettari, sono stati rintracciati diciassette nuclei rurali, organizzati lungo un percorso centuriale-stradale, identificabile con buona probabilità con il primo tracciato della via Appia (fig. 7).50 Il percorso della via pubblica segue cronologicamente di qualche anno quello della deduzione coloniale e quindi della distribuzione centuriale, come riscontrabile in molte altre colonie della penisola,51 confermando lo stretto rapporto tra colonizzazione e apertura delle grandi arterie, con un sistematico distacco cronologico tra le prime e le seconde, e il conseguente sfruttamento per le vie dei già esistenti assi centuriali.52 Poco più a est, su un’area di 56 ettari, sono distribuiti quattordici insediamenti, e successivamente su 88 ettari si segnalano ventidue fattorie.53 Si può ricostruire quindi una parcellizzazione su quote di assegnazione che oscillerebbero tra i 4 e i 5 ettari, pari a 16-20 iugeri.54 È possibile che anche nel settore occidentale, verso Melfi, la via consolare costituisse un asse della distribuzione delle fattorie dove, lungo tale percorso, che attraversa in modo abbastanza rettilineo tutto il sistema collinare, si trovano numerosi edifici rurali di piccole e medie dimensioni posti a una distanza non superiore ai 200 m (fig. 8). Diversi punti archeologici

49 50 51 52

53

Marchi e Sabbatini 1996, 112-13. Marchi 2004, 130-34. Muzzioli 2001, 10. Per una revisione critica di questo rapporto si veda comunque il contributo di Guy Bradley in questo stesso volume. Marchi e Sabbatini 1996, 112-15.

54

Queste cifre risultano piuttosto esigue anche per l’epoca, se per la colonia lucerina sono stati calcolati degli appezzamenti di 84-92 iugeri già nella fase repubblicana ( Jones 1980, 92-94) e Varrone calcolava che una famiglia di 4 persone poteva vivere con non meno 30 iugeri (Varro, Rust. 1.20).




di età repubblicana si distribuiscono lungo il percorso viario identificato con la via Appia;55 alcuni di essi sono senza dubbio aree di sepolture o monumenti funerari posti lungo la via consolare. A nord, intorno all’abitato di Lavello, il territorio è condizionato dai numerosi affluenti dell’Ofanto, primo fra tutti l’Olivento, ed è possibile che la distribuzione fosse organizzata seguendo l’andamento dei percorsi fluviali. I sistemi collinari occidentali sono occupati in età repubblicana con una densità inferiore alla zona orientale.56 È assai probabile che in questi settori fossero collocati gli insediamenti con quote di assegnazione più elevate; le fattorie o ville si distribuiscono infatti a una distanza di 400-500 m, così come avviene, nelle zone più periferiche tra Palazzo S. Gervasio e Spinazzola, sulle colline settentrionali affacciate sulla valle dell’Ofanto, vicine al confine con il territorio canosino, dove si riscontra una concentrazione lungo il percorso viario che collegava Venusia con Canusium.57 La distanza tra le fattorie del settore sudoccidentale si aggira tra i 600 e i 700 m, attestandosi quindi su assegnazioni ancora più ampie. Il ricorrere di questa distanza in più settori sembra potersi ricondurre a quella della centuria canonica.58 Ritroviamo tale misura nel territorio di Montemilone, dove le ville sono localizzate su ampi pianori, e tra Spinazzola e Palazzo S. Gervasio con fattorie di medie dimensioni (200-600 mq) che occupano in prevalenza ampie aree pianeggianti. Presso Spinazzola, all’estrema periferia del vasto territorio coloniale, la grande villa della Santissima di epoca tardo repubblicana, che si svilupperà nel villaggio tardo antico più esteso del comprensorio, nasce su una piccola fattoria documentata al centro del pianoro.59 Lungo la Via Appia tra il Piano di Palazzo e quello di Banzi, invece, le distanze sembrano ridursi e la distribuzione appare più fitta. Il percorso dell’antica via consolare potrebbe costituire l’asse portante del sistema di organizzazione centuriale della colonia e lungo di esso si attesterebbero i lotti più antichi e quindi di minori dimensioni, con una costante che si mantiene in molti casi anche nei periodi successivi quando le fattorie si trasformano in ville e gli appezzamenti in latifondi. Un’ampia fascia a nord di Banzi compresa tra il Vallone del Serpente e il Vallone Marascione risulta completamente spopolata, il che ci fa ipotizzare, almeno in età repubblicana, la presenza di un’area boschiva o in ogni caso non divisa e distribuita, che poteva costituire la cesura tra il territorio centuriato della colonia venosina e quello della civitas Bantina. Gli edifici rurali costruiti nella fase repubblicana possono inquadrarsi genericamente tra il III e il I sec. a.C. Raramente i materiali ceramici consentono di distinguere tra strutture abitate dai coloni del 291 a.C. e quelle successive relative alla deduzione graccana nota dalle fonti60 e in verità assai poco documentata anche negli altri settori del territorio coloniale. Si deve pensare che le case dei coloni dovessero essere costituite da strutture mono- o bilocali. Gli edifici rurali segnalati da aree di affioramento di materiale di minori dimensioni, generalmente 200-400 mq, sembrerebbero identificare edifici caratterizzati da pianta piuttosto semplice, in genere da uno o due ambienti, con cortile interno o posto sul retro molto simile all’impianto della fattoria Nocelli di Lucera,61 o alle case documentate in ambiente coloniale sia a Venosa che a Cosa,62 ma in genere nell’ambito di un tipo piuttosto diffuso in area dauna63 e lucana64 tra il IV e III secolo a.C. Alle ristrette dimensioni delle aree di

55 56 57 58

59



Cfr. infra p. 239. Marchi e Sabbatini 1996, 112-14; Marchi 2004, 133-34. Sabbatini 2001, 70-1; Marchi 2004, 136-37. La forma classica della centuria è rappresentata da un quadrato con il lati di 20 actus, cioè circa 710 m. Secondo la definizione di Plinio (Plin., HN 18, 9) si diceva actus la distanza sul terreno per la quale i buoi potevano tirare in un solo tratto l’aratro, mentre iugerum la superficie che con un paio di buoi si poteva arare in un giorno. L’actus corrisponderebbe a 35,48 m e quindi la centuria oscillerebbe tra i 705-10 m: Gabba 1985, 265. Marchi 2010a; Marchi, Di Stefano e Leoni 2006.

60 61 62

63

64

Lib. colon. I.210. Jones 1980, 85-100. Marchi 2000, 266-73; per l’ager cosanus cfr. Cambi 2002, 137-45. Per recenti rinvenimenti in Daunia cfr. Corrente et al 2008, 353-59. In genere sul problema degli insediamenti rurali in Lucania: Russo 1992; Di Giuseppe 1996, 189-252; Marchi 2000. In una recente sintesi si sono individuate varie tipologie abitative che comprendono edifici a un solo vano fino a giungere a strutture complesse e articolate con superfici anche


frammenti fittili corrisponde in genere anche una scarsa presenza di materiale da costruzione spesso da ricondurre a edifici costruiti con materiali deperibili. Un buon numero delle aree individuate identifica costruzioni che si inquadrano nella tipologia dei piccoli edifici rurali, con dimensioni che non dovrebbero superare i 100 mq, genericamente riconducibili ai tuguria o alle casae menzionate anche dalle fonti per indicare le case rurali di modeste dimensioni.65 Dove è possibile riscontrare, attraverso una distribuzione degli insediamenti più diradata, la presenza di assegnazioni più ampie, come nella zona settentrionale e nord-orientale, si deve presupporre anche la presenza di edifici con maggiore consistenza, probabilmente ville o grandi fattorie. Le estensioni rilevate, sempre attraverso la dispersione del materiale di superficie,66 si presentano oscillanti infatti tra i 400 e gli 800 mq, e si possono pertanto ipotizzare edifici con planimetrie più complesse, a volte caratterizzati da diversi corpi di fabbrica. Anche nei casi di edifici più articolati, occorre sempre distinguere tra ville non particolarmente lussuose, derivate da ampliamenti di fattorie e con decorazione ancora limitate all’essenziale, diffuse nella media e tarda età repubblicana, e le ville più grandi di età imperiale, caratterizzate oltre che da una estesa superficie, anche da elementi decorativi di un certo pregio.67 Nella fascia settentrionale del territorio le fattorie di III secolo a.C., cioè collegate alla deduzione della colonia, alcune delle quali hanno una continuità di vita dalla fase preromana, sembrano essere distribuite, in alcuni casi, a una distanza di circa 500 m, in altri anche fino a 700 m l’una dall’altra, con proprietà che raggiungevano quindi quote di assegnazione abbastanza elevate. Occorre inoltre tener presente che gli edifici dovevano avere tutti medie o grandi dimensioni. Le superfici calcolate si aggirano infatti tra i 300 e i 1500 mq. Se si confrontano con quelle stimate nelle aree orientali del territorio che oscillano tra 50 e 200 mq, si può evidentemente desumere che in queste zone si attestassero le proprietà più ampie con i complessi più imponenti, come si verifica anche nella zona tra il Lampeggiano e il Torrente Loconcello.68 A partire dagli anni della deduzione coloniale del 43 a.C., e poi nel corso dell’età imperiale, il numero dei complessi rurali tende a diminuire, le ampie aree pianeggianti sono occupate da grandi ville isolate affiancate da edifici minori localizzati a volte al centro, a volte alle estreme propaggini di amplissime proprietà. La diffusione della villa, fenomeno spesso ricollegato alla crisi della piccola proprietà,69 si connota come un processo di lenta trasformazione delle precedenti fattorie, con uno sviluppo planimetrico più esteso e articolato e un ampliamento delle proprietà dovuto all’accorpamento dei fundi e alla diffusione della produzione schiavistica;70 a essa, secondo alcuni, si deve la sostituzione delle proprietà dominate dalla cerealicoltura con vaste ed uniformi distese di vigneti, piantagioni di olivi e sterminati arborei e frutteti.71 La zona della campagna dauna, infatti, come attestano le ricerche nel Tavoliere, si presenta non soltanto utilizzata per colture cerealicole, né soltanto come zona dedita alla pastorizia, come la tradizione la disegnava, ma con un paesaggio vario, con estesi vigneti e uliveti.72 Il nostro comprensorio, come area di frontiera tra Daunia e Lucania, si raffronta con un vasto panorama che accomuna i due territori, in età romana ormai completamente omologati.

65 66

di 400 mq come gli edifici di Viggiano e Montemurro, nel comprensorio potentino, che a volte sembrano ancora ispirarsi al tipo della casa con portico, e spesso si trasformano nel tipo con cortile/atrio: Russo 2006, 188-203. Liv. 3, 13; 3, 26; 42, 4; 5, 53.8; Varro, Rust. 2.10.6. Cfr. i parametri stabiliti e presentati in Marchi 2010a, 25-27.

67

68 69

70 71 72

Per un inquadramento tipologico degli edifici rurali cfr. Lafon 2001, 15-40. Sabbatini 2001, 76-77. Gabba 1977, 269-84; Torelli 1990, 125-32; Per una sintesi in area laziale Valenti 2003, 143-46. Toynbee 1965. Capogrossi Colognesi 2002. Jones 1980; Volpe 1990; Marchi 2010a.




Fig. 9. Comprensorio della colonia di Luceria (siti archeologici individuati).

L’ager Lucerinus Anche il territorio della colonia di Lucera, oggetto di una recente indagine73 doveva estendersi per oltre 700 kmq74 (fig. 9) interessando un ampio comparto compreso tra il Fortore a nord e la pianura foggiana a sud. Dal punto di vista geomorfologico il paesaggio è ora, come doveva esserlo in antico, assai vario. Si passa dall’ampia pianura limitrofa all’abitato lucerino che fa parte del Tavoliere, alla fascia collinare, sulla quale si localizza anche il sito altomedievale di Montecorvino, alle alture del subappennino che raggiungono quote quasi montane.75 Questo settore si connota nell’antichità come zona di frontiera tra area dauna e frentana.76 Infatti verso il Tavoliere gli insediamenti di Arpi e Lucera costituiscono i caposaldi della cultura dauna, mentre dall’altro lato nel subappennino l’abitato di Carlantino77 rappresenta forse l’estrema propaggine meridionale della occupazione sannitica in questo settore, anche se ben sappiamo che la penetrazione osca, a partire dalla fine del V secolo, investe tutta la Daunia da Teanum Apulum a Lucera raggiungendo le zone più interne di Lavello-Forentum e Venosa.78 Anche in quest’area della Daunia ai grandi abitati si affiancano numerosi insediamenti di minori dimensioni, concentrati su ampi sistemi collinari che dominano la pianura.

73



A partire dal 2006 l’Università di Foggia conduce un vasto e articolato progetto che prevede lo scavo del sito medievale di Montecorvino a cura di Pasquale Favia e Roberta Giuliani, mentre le ricerche di survey sono dirette da chi scrive. L’indagine di survey estensivo e sistematico, che prevede la copertura totale dell’area prescelta, si è concentrata in particolare nel settore della colonia compreso fra il Tavoliere e il Subappennino dauno, e dovrebbe arrivare a interessare l’intera valle del Fortore. Le ricognizioni sul terreno per ora si sono concentrate nei territori dei comuni di Pietramontecorvino, Motta Montecorvino, Volturino,

74 75

76

77 78

Lucera, Casalnuovo Monterotaro Castelnuovo della Daunia, e Carlantino. Per una sintesi dei dati preliminari relativi a questa ricerca: Favia, Giuliani e Marchi 2007, 233-62; Marchi 2008a; Marchi e Buffo 2010; Marchi e Forte 2012. Grelle 2008. Sugli aspetti geomorfologici della Daunia cfr. Volpe 1990, 15-17; Marchi 2008b, 425-42. Sui confini tra daunia e area frentana cfr. Marchi 2000; ead. 2008a; ead. 2009b. De Benedittis 2006. Per una sintesi sul problema Marchi 2009a; ead. 2010a.


FIG. 10. Lucera carta archeologica e ipotesi di ricostruzione assi principali impianto urbano.

L’abitato di Lucera (fig. 10), localizzato nel sito della futura colonia, risulta essere senza dubbio il principale abitato, com’è ricordato dalle fonti, per le alterne vicende durante il conflitto sannitico, come insediamento minacciato, occupato dai Sanniti o passato al nemico, più volte ripreso dai Romani.79 Di questo insediamento, scarsa è la documentazione e ancora poco chiara l’articolazione abitativa; una presenza dell’Età del Ferro sembra documentata dal famoso carrello di Lucera e da una frequentazione sul Monte Albano. A un periodo più recente si ricollegano le necropoli, inquadrabili tra il V e il IV secolo a.C., sulle colline di Piano dei Puledri e di Carmine Vecchio,80 unici indizi del sistema abitativo articolato su più colline secondo i parametri degli abitati dauni.81 A questo articolato sistema si deve ricollegare l’area individuata, su una collina a sud del Monte Albano, nella zona delle Fornaci, che il materiale ceramico consente di inquadrare tra il VI e il IV secolo a.C. Questo nucleo insediativo consente di ridefinire l’estensione dell’abitato lucerino precedente la colonia. Un certo numero di centri minori è noto dalle notizie delle fonti, ancora di dubbia identificazione, come Gereonium82 e Acuca;83 altri invece sono documentati solo dai rinvenimenti archeologici e recenti indagini hanno permesso di mettere in evidenza un popolamento abbastanza diffuso con piccoli insediamenti disposti su sistemi collinari affacciati sulle principali vallate fluviali (Ofanto, Celone, Carapelle, Candeloro).84 Nell’area a ovest del moderno villaggio di Carignano, l’insediamento di Selva Piana si estende su un’area complessiva di circa 30 ettari; le concentrazioni di materiale fittile testimoniano uno stanziamento caratterizzato da numerosi edifici con elevati in argilla e coperture fittili inquadrabile tra il VI e il IV secolo a.C. Al centro del

79 80 81 82

Liv. 9.15, 3; Liv. 9.24 e 25. Lippolis 1999, 1-2; Marchi 2008b, 477-79. Marchi 2009b; ead. 2012. Polyb, 3.100.1-8; Polyb. 3.101.1-4 e 8-10; Polyb 3.107.1-4; Liv. 22.44.5-6.

83 84

Liv. 24.20.8. Favia, Giuliani e Marchi 2007; Marchi 2008b; Volpe, Romano e Goffredo 2003; iid. 2004; Goffredo 2008, 478; Marchi 2009; ead. 2010a; ead. 2012.




FIG. 11. Abitato di località Chiancone (siti archeologici individuati).

pianoro la foto aerea permette di leggere la traccia di un edificio rettangolare che suggestivamente possiamo assimilare a un oikos.85 La struttura identificata potrebbe essere simile all’edificio, recentemente indagato più a sud, in località Casanova,86 caratterizzato da una pianta rettangolare articolata in due vani e con una fascia di pavimentazione in ciottoli che permette di confrontarlo con il ben noto edificio di Ascoli Satriano. Su un ampio pianoro di circa 200 ettari, difeso naturalmente da vallate, è localizzato l’abitato di località Chiancone87 (fig. 11), identificato attraverso la presenza di varie aree di concentrazione di materiale. Tegole e coppi attestano la presenza di edifici abitativi, e l’abbondante ceramica permette di collocare tra il VII-IV secolo a.C. Al centro del pianoro, nell’area di maggiore concentrazione di elementi fittili, tra i quali spicca un numero notevole di frammenti di antefisse di vario tipo,88 prospezioni geomagnetiche89 hanno mostrato la presenza di un edificio costituito da vari ambienti, collocabile tra V e IV secolo a.C. L’insediamento sembra configurarsi, secondo le caratteristiche tipiche degli insediamenti dauni del periodo, con un’alternanza di nuclei di abitazione e sepolture. Il rinvenimento della sepoltura di un individuo armato deposto supino, collocabile nel V secolo a.C., conferma la presenza di elementi oschi anche in questo sito.90

85 86

87 88



Marchi e Buffo 2010; Marchi 2010b. Lo scavo è stato condotto dalla Soprintendenza per i beni archeologici della Puglia nell’ambito di una operazione di archeologia preventiva. Per una prima presentazione Corrente et al 2008, 376. Marchi 2008a; ead. 2010b; ead. 2012. Di particolare rilevanza il tipo pentagonale con cavaliere armato simile a un esemplare proveniente dalla stipe votiva di Lucera (D’Ercole 1990) e una di tipo nimbato paragonabile ai tipi etrusco-campani presenti ad Arpi, Teano e Lucera, nonché una matrice del medesimo tipo, che testimonia la

89

90

presenza di un centro di produzione nell’area dell’insediamento. Sono state condotte in collaborazione tra Università di Foggia e Scuola di Specializzazione in Beni Archeologici di Matera. Ringrazio la Dott.ssa Rizzo per aver messo a disposizione i dati della sua tesi. I sondaggi che hanno intercettato la sepoltura sono stati condotti dalla Soprintendenza per i Beni Archeologici della Puglia in occasione di lavori preliminari alla realizzazione di impianti eolici. Sono grata al funzionario responsabile Dott. Muntoni e al soprintendente La Rocca per avermi messo a disposizione il dato.


FIG. 12. Distribuzione delle fattorie repubblicane nel settore ad Ovest di Lucera.

A nord-est, a Castelnuovo della Daunia in località Finocchito, si segnala il più vasto degli abitati presenti in questa zona.91 Il sito, posto lungo un’antica strada che proveniva da Larinum e si biforcava per Arpi e Lucera, risulta di grande interesse per la sua alterna identificazione con l’abitato di Gereonium, ricordato dalle fonti per essere stato occupato da Annibale, e per la famosa battaglia nella quale Minucio risultò vittorioso nel 217 a.C.92 Se Gereonium sopravviverà alla riorganizzazione del territorio seguita alla deduzione della colonia lucerina, molti degli altri insediamenti verranno abbandonati, spesso a favore della realizzazione delle piccole fattorie dei coloni. Le profonde modifiche prodotte dall’intervento romano furono anticipate dagli eventi probabilmente traumatici che lo produssero e che avviarono il processo di trasformazione del paesaggio. Proprio la difesa di Lucera, i cui abitanti sono definiti “soci boni ac fidelis”,93 minacciata dai Sanniti, spinse i Romani in Daunia,94 ma solo successivamente, nel 315 o 314 a.C.,95 si concretizzeranno la deduzione coloniale con l’invio di 2500 coloni, la pianificazione urbana del sistema collinare già occupato dall’insediamento daunio96 e la suddivisione agraria di un vasto territorio (fig. 12). Per tutto il III secolo a.C. Luceria risultò essere un importante punto strategico per i Romani: resistette infatti ai Sanniti nel 294 a.C. e restò fedele a Roma nella guerra contro Taranto. La colonia, come primo avamposto romano nell’Italia meridionale, introduce nell’area nuovi modelli a livello sia urbanistico, sia monumentale, sia della pianificazione territoriale. Il vasto territorio fu diviso e assegnato, e di questa operazione abbiamo notizie dalle fonti; due passi del Liber coloniarum riferiscono delle divisioni agrarie lucerine.97 Finora i tentativi di ricostruzione della centuriazione lucerina, basati essenzialmente sulla lettura delle foto aeree98 hanno portato all’individuazione di varie divisioni.

91

92 93 94 95

96 97

Marchi e Buffo 2010; Marchi e Forte 2012; Gravina 1982; Russi 1982; Volpe 1990. Polyb. 3.93; Liv. 2.17. Liv. 8, 25, 3. Vell. Pat. 1, 14, 4. Sulla deduzione la prima data è riportata in Diod. Sic. 19.72.8; la seconda ci è riferita in Liv. 9.26, 1-5. Lippolis 1999, Marchi 2008b. Lib. colon. 1, 210, 15-19 L: “Ager Lucerinus Kardinibus

98

et decimanis est adsignatus: sed cursus solis secuti sunt et costituerunt centurias contra cursum orientalem n. LXXX et contra meridianum actus n. XV: efficiuntur n. DCXL. Iter populo non debetur.” Lib. colon. 2, 261, 9-11 L: “Lucerinus ager KK et dd est adsignatus: sed cursum solis sunt secuti et constituerunt centurias contra cursum orientalem. Finitur sic uti ager Ausculinus.” Bradford 1949; id. 1950; id. 1957; per una sintesi Radcliffe 2006; Schmiedt 1985; id. 1989.




A nord-ovest della città, a nord del torrente Salsola e a cavallo del corso del Triolo, è stata individuata una limitatio caratterizza da centurie di 20 x 20 actus di lato, riferita già da Toynbee99 alla fine del II secolo a.C. Una vasta area è stata individuata a est della città, che secondo Giulio Schmiedt avrebbe origine a circa tre chilometri dal centro urbano,100 e che comprende al suo interno le tracce di tre diverse divisioni.101 Sono stati messi in evidenza 12 assi paralleli orientati da sud-ovest a nord-est e costituiti da strade affiancate da fossati che terminano in corrispondenza di un tracciato viario identificato con la via Tiati-Arpi già di origine preromana. La distanza tra gli assi non è regolare e oscilla tra i 15,5, i 18 e i 21 actus.102 Nel settore più meridionale di questa fascia, Schmiedt segnala anche la presenza di kardines, ortogonali ai primi, posti a una distanza di circa 26.76 actus l’uno dall’altro. Quindi a un’ipotesi che proponeva una divisione per decumanos solos, in genere attribuita alle colonie più antiche,103 si affianca una scamnatio costituita da rettangoli di 13,38 x 26,76 actus. Oltre questo tracciato il sistema muta radicalmente con diverso orientamento; sono stati notati altri due assi che sembrano far proseguire la centuriazione verso nord-est per due ulteriori chilometri, attestandosi su un intervallo di circa 20 actus.104 Anche le ricerche di Geraint Jones hanno rilevato lo stesso sistema di suddivisione che attraversa per circa 9 km la terrazza sottostante la città con l’impianto di strade parallele: Jones ricostruisce un sistema di allineamenti basati su un angolo di 61° e disposti a intervalli leggermente irregolari, all’interno dei quali la ricerca sul terreno ha permesso di individuare una serie di fattorie. La contemporaneità delle fattorie con gli assi centuriati è confermata dalla presenza o di interruzioni nei fossati laterali del decumano, o in altri casi di vere e proprie strade di accesso dal decumano ai nuclei delle fattorie. È stato calcolato che le fattorie potessero occupare un fronte lungo il decumano di circa 400 metri, cioè 16 actus, e si sono ricostruiti appezzamenti di 86 iugera fino a un massimo di 92 iugera. Considerando le variazioni minime delle dimensioni citate e prendendo il decumano come divisione, si possono ritenere le cifre proposte da Jones rappresentative dell’appezzamento medio dato in dotazione ai coloni di Lucera.105 È stato ricostruito anche il tipo di colture praticate; la lettura delle tracce nella fotografia aerea lascia infatti trapelare la presenza di macchie relative a vigneti e uliveti in circa due terzi del territorio, mentre il rimanente poteva essere riservato a cereali e pascolo. Lo scavo della fattoria di Nocelli, contemporanea all’impianto, ha permesso di ancorare quest’ultimo nell’ambito della fine del II secolo a.C., con una vita fino all’età tiberiana. È stata ipotizzata anche una centuriazione per soli decumani o piuttosto con cardini non facilmente rilevabili perché di secondaria importanza, e una a sud con un sistema di distribuzione organizzato con una divisione per rettangoli, che possono essere interpretati come divisioni relative a due distinti momenti cronologici, ma anche come due reticoli separati ancorché conviventi.106 Secondo Daniele Manacorda107 i 13,36 actus si possono leggere in 1600 piedi, e la misura di base non dovrebbe riferirsi come multiplo del piede all’actus, ma al vorsus (una misura lineare di 100 piedi)

99 100 101 102

103 104 105



Toynbee 1965, 700. Schmiedt 1985, 263-69. Volpe 1990. Sono inoltre segnalate misure discordanti anche tra i vari studiosi John Bradford indica una distanza per gli assi A G di 15 actus; Schmiedt indica invece 17,5 e tra gli assi H-I di 20,5 e nel settore meridionale indica una distanza di 13, 5 actus. Castagnoli 1958, 24. Jones 1980. Secondo quanto specificato da Jones, questa stima dedotta da fattori esclusivamente archeologici è in se stessa molto significativa perché induce a ulteriori

106 107

considerazioni. Varrone calcolava che i terreni da 100 iugera (in particolare vigneti) fossero un’unità agraria considerevole e aggiungeva che una famiglia di coloni di quattro persone poteva vivere dei prodotti di 30 iugera (Varro, Rust. 1.20.4). Ne deriva che, anche tenendo conto dell’inclemenza del clima pugliese e del suo effetto negativo sulle colture, le dimensioni che abbiamo attribuito alle fattorie esaminate dovevano essere sufficienti non solo per la sussistenza dei coloni ma anche per la necessità del mercato. Castagnoli 1984, 241-52. Manacorda 1991, 49-66, cf. Pelgrom 2008.


testimoniato da alcune fonti.108 Si tratterebbe quindi di centurie di 16 x 32 vorsus. Tracce di un reticolo con centurie di 20 actus di lato si rinvengono anche a sud, nella zona tra i torrenti Vulgano e Celone,109 attribuite per un’analogia metrica con la limitatio della zona settentrionale, agli ultimi decenni del II secolo a.C.110 Una distribuzione centuriale si può forse individuare nel settore a ovest della città, dove è stata individuata una fitta maglia di piccole fattorie. Esse si concentrano perlopiù nell’area pianeggiante lungo la Strada Statale 17. Le aree di frammenti fittili permettono di identificare piccoli edifici probabilmente di 100-200 mq. Si possono ipotizzare strutture caratterizzate da pianta piuttosto semplice sempre con uno o due ambienti, e con parte degli elevati in materiali deperibili, del tutto simili a quelle ipotizzate nell’ager venusinus. Abbastanza suggestive anche le tracce che si individuano a nord tra il torrente Triolo e la località Fornelli, dove una fitta rete di fattorie si inserisce all’interno di un gruppo di tracce individuabili attraverso la lettura della foto aerea e interpretabili come assi di una divisione centuriale disposti a una distanza di 700-710 m circa (circa 20 actus), che quindi vanno a riallacciarsi alla divisione identificata dall’altro lato del corso fluviale. Per queste assegnazioni potremmo ipotizzare una collocazione cronologica tra il II e il I secolo a.C., in relazione forse a interventi graccani e cesariani.111 Mentre appare piuttosto difficile identificare significative tracce legate al passaggio dallo stato municipale a quello della deduzione coloniale di età augustea, alla quale potrebbero agganciarsi alcune delle ville individuate in più settori del comprensorio, più certa è una riorganizzazione di piena età imperiale, quando scompaiono molte delle fattorie e molte si trasformano in grandi ville inserite in più grandi proprietà nate dall’accorpamento dei fondi.

108

Varro, Rust. 1.10.1: “Versum dicum centum pedes quoquo versum quadratum”, vorsus equivale alla lunghezza del solco compiuto dal giogo prima di girare l’aratro per dare inizio a un nuovo solco, ma funge anche da misura di superficie equivalente a un quadrato di 100 piedi di lato.

109 110 111

Volpe 1990, 214. Bonora Mazzoli 2001, 70. Jones 1980.




Bibliografia Angelini, G. 1999. “Appunti per una storia del paesaggio rurale,” in A. Maurano (ed.), Venosa diecimila ettari di storia (Venosa) 13-5. Angelini, G., L. Di Vito e A. Groia 1970. “Venosa: saggio per una carta storica del territorio comunale,” Storia della città 49, 89-124. Aprosio, M. 2008. Archeologia dei Paesaggi a Brindisi. Dalla romanizzazione al medioevo (Bari). Boersma, J., G-J. Burgers e D. G. Yntema, 1991. “The Valesio Project. Final interim report,” BABesch 66, 115-31. Boersma, J. e D. G. Yntema 1982. “The Oria Project: first interim report,” BABesch 57, 213-16. Boersma, J. e D. G. Yntema 1987. “The Oria Project: second interim report,” BABesch 62, 1-19. Bonora Mazzoli, G. 2001. “Testimonianze di centuriazione nel Liber Coloniarum,” in S. Alessandrì e F. Grelle (edd.), Dai Gracchi alla fine della Repubblica (Galatina) 61-78. Bottini, A. 1982. “Il melfese tra VII e V secolo a.C.,” DialArch 4, 2, 152-60. Bottini, A. 2001. “Gli Italici della mesogaia lucana ed il loro sistema insediativo,” in M. L. Lazzarini e P. Poccetti (edd.), Il mondo enotrio tra VI e V secolo a.C. (Napoli) 109-16. Bottini, A. 2013. “Il modello insediativo ‘nord-lucano’,” in M. Osanna e M. Vullo (edd.), Segni del Potere. Oggetti di lusso dal Mediterraneo nell’Appennino lucano di età arcaica (Venosa) 17-21. Bradford, J. 1949. “Buried landscapes in southern Italy,” Antiquity 23, 58-72. Bradford, J. 1950. “The Apulia expedition: an interim report,” Antiquity 24, 84-95. Bradford, J. 1957. Ancient landscapes. Studies in field archaeology (London). Bradley, G. J. 2006. “Colonization and identity in Republican Italy”, in G. J. Bradley e J. P. Wilson (edd.), Greek and Roman Colonization. Origins, Ideologies and Interactions (Swansea) 161-87. Brunt, P. A. 1971. Italian Manpower 225 B.C.-A.D. 14 (2nd edn., Oxford). Burgers, G-J. 1994. “The Salento Isthmus Project. Second Interim Report,” BABesch 69, 145-154. Burgers, G-J. 1995. “Ricognizioni sistematiche nel Salento: il caso di Muro Tenente,” StSalent 72, 124-41.



Burgers, G-J. 1998. Constructing Messapian landscapes. Settlement dynamics, social organization and culture contact in the margins of Graeco-Roman Italy (Amsterdam). Cambi, F. 2002. “La casa del colono e il paesaggio (III-II a.C.),” in A. Carandini e F. Cambi (edd.), Paesaggi d’Etruria tra la Fiora e l’Albegna, valle d’Oro, valle del Chiarone, valle del Tafone (Roma) 137-45. Camodeca, G. 1982. “Ascesa al senato e rapporti con i territori d’origine. Italia: regio I (Campania, esclusa la zona di Capua e Cales), II (Apulia et Calabria), III (Lucania et Bruttii),” in S. Panciera (ed.), Epigrafia e ordine Senatorio 2 (Roma) 10163. Capogrossi Colognesi, L. 2002. Persistenza e innovazione nelle strutture territoriali dell’Italia romana (Napoli). Castagnoli, F. 1958. Ricerche sui resti della centuriazione (Roma). Castagnoli, F. 1984. “Sulle più antiche divisioni agrarie romane,” RAL 39, 241-57. Catizzone, B. A., G. Giusteschi e M. Coppa 1979. “L’assegnazione di Venosa attraverso l’analisi della struttura ambientale, produttiva e storica,” in M. Coppa e G. Alvisi (edd.), Fotografia aerea e storia urbanistica (Roma) 87-128. Coarelli, F. 1992. “Colonizzazione e municipalizzazione: tempi e modi,” DialArch 10, 21-30. Colivicchi, F. (ed.) 2011a. Local cultures of South Italy and Sicily in the Late Republican period between Hellenism and Rome (Portsmouth, RI). Colivicchi, F. 2011b. “The long goodbye: the local élites of Daunia between continuity and change (3rd-1st c. B.C.),” in F. Colivicchi (ed.), Local cultures of South Italy and Sicily in the Late Republican period between Hellenism and Rome (Portsmouth, RI) 112-37. Corrente, M. et al. 2008. “Le diverse esigenze. Paesaggio rurale, archeologia preventiva e fattorie del vento”, in Atti del 28 °Convegno di Preistoria, Protostoria e Storia della Daunia San Severo 2007 (San Severo) 341-74. Coppa, M. 1979. “Formazione della struttura territoriale dell’agro di Venosa,” in M. Coppa e G.


Alvisi (edd.), Fotografia aerea e storia urbanistica (Roma) 119-28. Curti, E. 2001. “Toynbee’s legacy. Discussing aspects of the Romanization of Italy,” in S. Keay e N. Terrenato (edd.), Italy and the West: comparative issues in Romanization (Oxford) 17-26. D’Ercole, M. C. 1990. La stipe del Belvedere a Lucera (Roma). David, J. M. 1994. La romanisation de l’Italie (Parigi). Dall’Aglio, P. L. 2004. “Perché studiare la Centuriazione,” Agri Centuriati 1, 17-21. De Benedittis, G. 2006. Carlantino. La necropoli di Santo Venditti (Campobasso). Desideri, P. 1991. “La romanizzazione dell’impero,” in Storia di Roma, vol. 2: L’impero mediterraneo. 2, I principi e il mondo (Torino) 577-626. De Juliis, E. M. 1984. Gli Japigi. Storia e civiltà della Puglia preromana (Milano). Di Giuseppe, H. 1996. “Insediamenti rurali della Basilicata interna tra la romanizzazione e l’età tardoantica: materiali per una tipologia,” in M. Pani (ed.), Epigrafia e territorio politica e società (Bari) 189-252. Favia, P., R. Giuliani e M. L. Marchi 2007. “Montecorvino: note per un progetto archeologico. Il sito, i resti architettonici, il territorio. La ricognizione,” in Atti del 27° convegno di preistoria, protostoria e storia della Daunia (San Severo) 233-62. Franchin Radcliffe, F. (edd.), 2006. Paesaggi sepolti in Daunia. John Bradford e la ricerca archeologia dal cielo (1945-1957) (Foggia). Gabba, E. 1977. “Considerazioni sulla decadenza della piccola proprietà contadina nell’Italia centro-meridionale del II secolo a.C.,” Ktema 2, 269-84. Gabba, E. 1978. “Il problema dell’unità dell’Italia preromana,” in La cultura italica. Atti del convegno della società italiana di glottologia (Pisa) 11-27. Gabba, E. 1979. “Sulle strutture agrarie dell’Italia romana tra il III e il I sec. a.C.,” in E. Gabba and M. Pasquinucci (edd.), Strutture agrarie e allevamento transumante nell’Italia romana (IIIII sec. a.C.) (Pisa) 15-73. Gabba, E. 1985. “Per una interpretazione storica della centuriazione romana,” Athenaeum 73, 20-7. Goffredo, R. 2008. “Persistenze e innovazioni nelle

modalità insediative della valle dell’Ofanto tra fine IV e I sec. a.C.,” in G. Volpe, M. J. Strazzulla and D. Leone (edd.), Atti delle giornate di studio sulla Daunia antica in memoria di Marina Mazzei (Bari) 287-301. Goffredo, R. 2011. Aufidus. Storia archeologia e paesaggi della vale dell’Ofanto (Bari). Gravina, A. 1982. “Contributo per una carta topografica nel bacino del basso Fortore dall’età romana al medioevo,” in Atti del 4° Convegno sulla Preistoria, Protostoria e Storia della Daunia (San Severo) 49-90. Grelle, F. 1992. “Il municipio e la colonia,” in R. Cassano (ed.) Principi, imperatori e vescovi. 2000 anni di storia a Canosa (Venezia) 683-691. Grelle, F. 1994. “La centuriazione di celenza Valfortore, un nuovo cippo graccano e la Romanizzazione del Subappennino dauno,” in Ostraka 2, 249-58. Grelle, F. 1995. “Ordinamento municipale e organizzazione territoriale nella Puglia romana,” in Studi in memoria di Ettore Lepore (Napoli) 241-60. Grelle, F. 1999. “Forme insediative, assetto territoriale e organizzazione municipale nel comprensorio del Celone,” in Atti del 17° convegno sulla Preistoria, Protostoria e Storia della Daunia (San Severo) 387-401. Grelle, F. 2008. “Le colonie latine e la romanizzazione della Puglia,” in G. Volpe, M. J. Strazzulla and D. Leone (edd.), Atti delle giornate di studio sulla Daunia antica in memoria di Marina Mazzei (Bari) 365-87. Guaitoli, M. e V. Cazzato (edd.) 2005. Lo sguardo di Icaro, insediamenti del Salento dall’antichità all’età moderna (Roma). Incontro di studi 1970-71. Incontro di studi su “Roma e l‘Italia fra i Gracchi e Silla” (Siena, Certosa di Pontignano, 18-21 settembre 1969): atti, DialArch 4. Keay, S. J. e N. Terrenato (edd.) 2001. Italy and the West. Comparative issues in Romanization (Oxford). Jones, G. B. D. 1980. “Il Tavoliere romano. L’agricoltura romana attraverso l’aerofotografia e lo scavo,” ArchCl 32, 85-100. La Romanisation 1991. La Romanisation Du Samnium Aux Iie Et Ier siècles Av. J.-C: Actes Du Colloque (Napoli).




Lafon, X. 2001. Villa maritima: recherches sur les villas littorales de l’Italie romaine (3. siècle av. J. C.-3. siècle ap. J. C) (Roma). Lepore, E. 1963. Ricerche sulla penetrazione romana in Apulia e Lucania (Bari). Lippolis, E. 1999. “Lucera: impianto e architettura della città romana,” in E. Antonacci (ed.), Lucera. Topografia storica Archeologia Arte (Bari) 1-25. Manacorda, D. 1991. “La centuriazione di Lucera,” in Profili della Daunia antica, 7° ciclo di Conferenze (Foggia) 49-66. Marchi, M. L. 1999. “Effetti della penetrazione romana nel comprensorio venosino, in La Daunia romana,” in Atti del 17° Convegno di Preistoria, Protostoria e Storia della Daunia (San Severo) 111-28. Marchi, M. L. 2000. “Effetti del processo di romanizzazione nelle aree interne centromeridionali. Acquisizione, innovazioni ed echi tradizionali documentati archeologicamente,” Orizzonti 1, 227-42. Marchi, M. L. 2004. “Fondi, latifondi e proprietà imperiali nell’ager venusinus,” Agri Centuriati 1, 129-56. Marchi, M. L. 2008a. “Dall’abitato alla città. La romanizzazione della Daunia attraverso l’evoluzione dei sistemi insediativi,” in G. Volpe, M. J. Strazzulla and D. Leone (edd), Atti delle Giornate di Studio sulla Daunia Antica in memoria di Marina Mazzei (Bari) 267-86. Marchi, M. L. 2008b. “Nuovi dati per una ricostruzione storica del paesaggio del Subappenino dauno: dall’Ager Lucerinus a Montecorvino,” in Atti del 28° convegno di preistoria, protostoria e storia della Daunia (San Severo) 475-499. Marchi, M. L. 2009a. “La romanizzazione della Daunia. Il quadro storico-archeologico,” in “La monetazione pugliese dall’età classica al medioevo (Bari) 21-41. Marchi, M. L. 2009b. “Modi e forme dell’urbanizzazione della Daunia,” in M. Osanna (ed.), Verso la città. Forme insediative in Lucania e nel mondo italico fra IV e III secolo a.C. (Venosa) 327-67. Marchi, M. L. 2010a. Ager Venusinus II (Firenze). Marchi, M. L. 2010b. “Villaggi, fattorie e ville: tracce del popolamento antico nel territorio di Luceria,”



Archeologia Aerea 4, 185-90. Marchi, M. L. e D. Buffo 2010. “Tra la valle del Fortore e il subappenino daunio: nuovi dati per la ricostruzione storica del paesaggio antico,” in Atti del 30° convegno sulla Preistoria, Protostoria e Storia della Daunia (San Severo) 407-26. Marchi, M. L. e G. Forte 2012. “Paesaggio e storia della Daunia antica: l’ager Lucerinus,” in Atti del 32° Convegno di Preistoria, Protostoria e Storia della Daunia (San Severo) 271-90. Marchi, M. L. e G. Sabbatini 1996. Venusia (Firenze). Marchi, M. L. e M. Salvatore 1997. Venosa. Forma e urbanistica (Roma). Marchi, M. L., V. Di Stefano e G. Leoni 2006. “Paesaggi rurali della Daunia interna. Nuovi dati dall’agro di Spinazzola (BA), il complesso in località ‘La Santissima’,” in Atti del 26° convegno sulla Preistoria, Protostoria e Storia della Daunia (San Severo) 425-42. Mazzei, M., J. Mertens e G. Volpe 1990. “Aspetti della romanizzazione della Daunia,” in M. Salvatore (ed.), Basilicata: l’espansionismo romano nel sud-est d’Italia: il quadro archeologico: atti del Convegno: Venosa, 23-25 aprile 1987 (Venosa) 177-95. Mazzei, M. (ed.) 1984. La Daunia Antica dal Paleolitico all’alto Medioevo (Milano). Mertens, J. e R. Lambrechts (edd.) 1991. Comunità indigene e problemi della romanizzazione nell’Italia centro-meridionale, IV-III secolo a.C. (Bruxelles). Modes de contacts et processus 1983. Modes de contacts et processus de transformation dans les societes anciennes: actes du Colloque de Cortone (24-30 mai 1981) organisé par la Scuola normale superiore et l’Ecole francaise de Rome, avec la collaboration du Centre de recherches d’histoire ancienne de l’Universite de Besancon (Pisa/Roma). Morizio, V. 2007. “I Lutatii Catuli in Daunia: una importante famiglia romana a Luceria,” in Contributi all’epigrafia d’Età Augustea. Actes della 13° rencontre franco-italienne sur l’épigraphie du monde romain (Tivoli-Roma). Muzzioli, M. P. 2001. “Sui tempi di insediamento dei coloni nel territorio,” ATTA 10, 7-20. Pani, M. 1977. “Su nuovo cippo graccano dauno,” RendIstlomb 111, 389-400.


Pelgrom, J. 2008. “Settlement organization and land distribution in Latin colonies before the Second Punic War,” in L. de Ligt and S. J. Northwood (edd.), People, land and politics: demographic developments and the transformation of Roman Italy 300 BC-AD 14 (Leiden) 333-72. Russi, A. e A. Valvo 1997. “Note storiche sul nuovo termine graccano di Celenza Valfortore,” in Quinta miscellanea greca e romana (Roma) 225-49. Russo, A. 2006. Con il fuso e la conocchia. La fattoria lucana di Montemurro e l’edilizia domestica nel IV secolo a.C. (Milano). Russo Tagliente, A. 1992. Edilizia domestica in Apulia e Lucania. Ellenizzazione e società nella tipologia abitativa indigena tra VIII e III secolo a.C. (Galatina). Sabbatini, G. 2001. Ager Venusinus I: Mezzana del Cantore (IGM 175 2. SE) (Firenze). Salvatore, M. 1984. Venosa: un parco archeologico e un museo. Come e perché (Taranto). Salvatore, M. R. (ed.) 1990. Basilicata. L’espansionismo romano nel Sud-Est d’Italia. Il quadro archeologico (Venosa). Schmiedt, G. 1985. “Le centuriazioni di Lucera e Aecae,” L’Universo 65, 2, 260-304. Schmiedt, G. 1989. Atlante aerofotografico delle sedi umane in Italia, vol. 3: La centuriazione romana (Firenze). Stek, T. D. 2009. Cult places and cultural changes in Republican Italy. A contextual approach to religious aspects of rural society after the Roman conquest (Amsterdam). Tagliente, M. (ed.) 1990. Italici in Magna Grecia: lingua, insediamenti, strutture (Venosa). Toynbee, A. J. 1965. Hannibal’s legacy: the Hannibalic war’s effects on Roman life (London). Terrenato, N. 1998. “The Romanization of Italy. Global acculturation or cultural bricolage?,” in C. Forcey, J. J. Hawthorne and R. Witcher (edd.), TRAC 97 (Oxford) 20-7. Torelli, M. 1984. “Aspetti storico-archeologici della romanizzazione della Daunia,” in La civiltà dei Dauni nel quadro del mondo italico (Firenze) 325-36. Torelli, M. 1988. “Aspetti ideologici della colonizzazione romana più antica,” DialArch 6, 65-72.

Torelli, M. 1991. “La fondazione di Venosa nel quadro della romanizzazione dell’Italia meridionale,” in M. Salvatore (ed.) Il Museo Archeologico Nazionale di Venosa (Matera) 18-26. Torelli, M. 1992a . “Aspetti materiali e ideologici della Romanizzazione della Daunia,” DialArch 10, 4764. Torelli, M. 1992b. “Il quadro materiale e ideale della Romanizzazione,” in R. Cassano (ed.), Principi, imperatori, vescovi. Duemila anni di storia a Canosa (Venezia) 608-19. Torelli, M. 1992c. “Venosa romana,” in P. Fedeli (ed.), Venosa (Venosa) 33-79. Torelli, M. 1996. “La Romanizzazione del Sannio,” in L. Del Tutto Palma (ed.), La Tavola di Agnone nel contesto italico (Firenze) 27-44. Torelli, M. 1999. Tota Italia. Essays in the cultural formation of Roman Italy (Oxford). Valenti, M. 2003. “Il rapporto tra la città e il territorio. Strutture dell’economia e della residenza,” in P. Sommella (ed.), Atlante del Lazio (Roma) 141-80. Vallat, J. P. 1995. L’Italie et Rome (Paris). Van Dommelen, P. e N. Terrenato (edd.) 2007. Articulating local Cultures. Power and identity under the expanding Roman republic (Portsmouth, RI). Volpe, G. 1990. La Daunia nell’età della Romanizzazione (Bari). Volpe, G. 2001. “Linee di storia dela paesaggio dell’Apulia Romana: San Giusto e la valle del Celone,” in E. Lo Cascio e A. Storchi Marino (edd.), Modalità insediative e strutture agrarie nell’Italia meridionale in età romana (Bari) 315-61. Volpe, G.,V. Romano e R. Goffredo 2003. “Archeologia dei paesaggi della valle del Celone,” in Atti del 23° Convegno sulla Preistoria, Protostoria, Storia della Daunia (San Severo) 349, 391. Volpe, G., V. Romano e R. Goffredo 2004. “Il Progetto valle del Celone”: ricognizione, aerofotografia, GIS,” AAAd 58, 181-220. Yntema, D. G. 1993. In search of an ancient countryside. The Amsterdam Free University Field Survey at Oria. Province of Brindisi, South Italy (1981-1983) (Amsterdam).




Roman colonial landscapes: Interamna Lirenas and its territory through antiquity Giovanna R. Bellini, Alessandro Launaro and Martin Millett “ποταμοῖσι τοῖσιν αὐτοῖσιν ἐμβαίνουσιν ἕτερα καὶ ἕτερα ὕδατα ἐπιρρεῖ ” “to those who step in same rivers ever different waters flow” (Heraclitus DK22b12)1

Introduction Archaeological research is constantly evolving. This is true not just of methodology, but also – perhaps especially – of the questions methodologies address. Archaeological survey in Italy developed from the awareness that ancient remains scattered across the landscape were being threatened with significant destruction with the spectacular growth of urbanization.2 Field-survey was the best way to record vanishing ancient settlement patterns. However, the amount of evidence collected has far exceeded its scope in both quantity and quality and, even in the initial stages, it was obvious that this new ‘ploughsoil assemblage’ had the potential to address wider archaeological and historical questions.3 Pioneering work was done in the South Etruria Survey organized by the British School at Rome (BSR), which paved the way for many more survey projects in Italy (cf. the contributions of Attema et al., Vermeulen, and Marchi in this volume).4 Landscape archaeology, which is actively engaged with other disciplines (e.g. the natural sciences) and with ‘new’ technologies (e.g. GIS, geophysical prospection), has since become a fully established discipline and is making an essential contribution to illuminating the development of people and places throughout the Mediterranean.5 The questions the surveyors have raised have evolved with the development in its method. It has even been possible to revisit earlier fieldwork and frame it within new research perspectives. The BSR Tiber Valley Project is an excellent example: benefiting from important advances in both urban survey (e.g. geophysics) and material analyses (e.g. distribution patterns), it has endeavoured to address important issues which have emerged in the decades since the South Etruria Survey was originally published.6 Debates move forward and new questions arise, but landscapes also evolve. Before the widespread introduction of mechanized agriculture in the 1950s ploughed them back to the surface, all but the most grandiose remains of the rural settlement of ancient Italy were practically invisible.7 The landscape is still a

1 2

3 4

Translation by the authors. The scope was well-summarized by John Ward-Perkins: “The purpose of this series of notes being to record the extant or recently destroyed remains of antiquities within the several cities of South Etruria and their associated territories” (Ward-Perkins 1961, 1). Compare with similar remarks by Lorenzo Quilici: “Da allora ho continuato a percorrere la regione, che ho visto man mano deperire ed inglobare in un’immensa periferia, che è tra le più squallide dell’infelice esperienza urbana di Roma” (Quilici 1974, 11). E.g. Frederiksen 1970-71. On the role of the British School at Rome see Potter and Stoddart 2001, 10-23. There is no comprehensive

5

6

7

list of projects to rely on, but an overview is provided by Patterson 2006, 72-88 and Launaro 2011b, 103-45. See the 5 volumes of the POPULUS Project: Barker and Mattingly 1999-2000; review by Cherry 2002. Patterson and Millett 1998. Contrast Potter 1979 with Patterson, Di Giuseppe and Witcher 2004. This awareness developed slowly over time as a much quoted – and perhaps by now abused – passage by Moses I. Finley demonstrates: “The Graeco-Roman world […] was a world of cities. Even the agrarian population, always a majority, most often lived in communities of some kind, hamlets, villages, towns, not in isolated farm homesteads” (Finley 1977, 305).




FIG. 1. Ancient features across the Liri Valley: urban settlement, road network and field systems (strigatio/scamnatio and centuriatio) (dark grey = areas over 200 m AMSL).

home to a large population and is a significant economic resource (e.g. urbanization, infrastructures): it keeps changing, but the scale of such change is often more rapid than the ability of archaeology to keep pace with it. Most importantly, this means that it is rarely possible to survey the same landscape twice. These remarks serve to explain why the Faculty of Classics of the University of Cambridge, in collaboration with the British School at Rome and the Soprintendenza per i Beni Archeologici del Lazio, has now launched the Roman Colonial Landscapes field project in the Lower Liri Valley (Southern Lazio, Italy), its focus being the archaeological context of Interamna Lirenas and its territory. This region is certainly not terra incognita as the amount of earlier and current research makes abundantly clear (see below). However, it is precisely by building on this work that it is possible to introduce new approaches and hence address questions which are crucial to current debates. Colonization is obviously a central issue in our analysis, especially as it underpinned various important territorial developments like urbanization, road-building and land allotment in the area between the 4th and 1st c. B.C. The area is traditionally seen as having played a key part in the development of Roman colonial practice in this crucial period. Consequently it exemplifies issues which extend well beyond the geographical compass of the Liri Valley. Moreover, it is increasingly recognized that Roman colonization was not an entirely one-sided process as colonizers had to negotiate specific relationships in particular local situations (e.g. religious cults, settlement patterns and material culture). This necessity to negotiate generated a tension on different scales of analysis (i.e. general vs. local), which can be better understood by contrasting the ‘colonial period’ with both what came before and what followed, adopting the long-term perspective which is the strength of landscape archaeology. As this essay will try to demonstrate, returning to Interamna Lirenas is not merely a repetition of a past effort, it is ‘stepping into different waters’.




Context The Liri Valley (fig. 1) lies about 100 km south of Rome, in the province of Frosinone, near its border with Campania. The river basin is part of an area known today as Ciociaria and is best described as a wide sedimentary valley floor extending NW-SE and bordered by steep limestone mountains to both the north (Le Mainarde) and the south (Monti Aurunci).8 Throughout history, it has represented a natural inland corridor between central Lazio and Campania, running parallel to the route along the Tyrrhenian coast (from which it is separated by the Monti Aurunci range).9 This aspect is very evident today as the valley is the route of both the A1 motorway and of the ordinary and the high-speed railway lines. Hence it represents the primary connection link between Rome and Naples, between central and southern Tyrrhenian Italy. This connective role was undoubtedly a significant factor in Roman expansion into the area (cf. Bradley in this volume). In contrast to the totally new Via Appia along the coast (built 312 B.C.), in the context of the Roman operations of the second half of the 4th c. B.C., the Via Latina formalized a much earlier route.10 The area was inhabited by the Volsci (NW) and the Ausones/Aurunci (SE), and later became the focus of the clash between Rome and the Samnites in their struggle to control Campania.11 The foundation of the Latin colonies of Fregellae (328 B.C.) and Interamna Lirenas (312 B.C.), both located along the Via Latina, was imposed at the expense of earlier settlements12 and sounded the death knell of Samnite presence in the region. The origins of Interamna Lirenas (also known as Interamna Sucasina) can be traced to its role as one bridgehead of southward Roman expansion.13 Its name reflects its location: between two rivers (Interamna) near the River Liri (Lirenas) and the town of Casinum (Sucasina).14 Although it had originally been assigned 4,000 colonists,15 its strategic role was lost after the Roman subjugation of southern Italy (completed in 270 B.C.) and the importance of Interamna ebbed considerably, and it was only mentioned in passing by ancient authors.16 Like most Latin communities south of the River Po, it probably became a municipium around 90 B.C., at which time its inhabitants would have been granted full Roman citizenship. The Liber coloniarum mentions a triumviral intervention (i.e. the building of the city wall), which might also imply some partial re-arrangement of the rural landscape, possibly as a consequence of the extensive centuriatio centred on Aquinum.17 By this time, this latter – as had Casinum – achieved greater prominence, evident from the new route of the Via Latina, which ran directly between the two centres, effectively bypassing Interamna.18 Following its probable abandonment at the time of the Lombard invasion (second half of the 6th c. A.D.), a process of spoliation began and, with the exception of its SE tip during the Middle Ages, the site

8

9

10

11

12

13 14

A thorough presentation and discussion of the local geology and geomorphology is provided by Martini 1994. But contrast this idea with remarks by Wightman, who argues that “Conditions during the last two centuries before the Roman conquest […] were inimical to traffic. The valley served as frontier rather than corridor, and while not of course hermetically sealed, was no highway” (Wightman 1994a, 30). Cagiano de Azevedo 1947, 37-39; Wightman 1994a, 30-32; Coarelli 1998, 49-51; Ceraudo 2003, 442-43; Ceraudo 2004d, 29-30. Wightman 1994c, 26-27. On pre-Roman settlement in the mountains south of Interamna Lirenas see Lauria 2010. Fregellae: Coarelli 1998, 45-46. Interamna Lirenas: Cagiano de Azevedo 1947, 8; id. 1949, 17; Carettoni 1948, 18. Cagiano de Azevedo 1947, 7-8. Ibid., 11. Strabo assumed one of the rivers to be the

15 16

17

18

Liri, an assumption which makes little sense in terms of the local topography (Strab. V.3.9). Liv. 9.28. A comprehensive list of relevant passages is reported in Cagiano de Azevedo 1947, 47-49. Lib. colon. L 234 = C 182.36-37; Campbell 2000, 42021 n. 124. Concerning the nature, dating and extent of such a centuriatio see Chouquer et al. 1987, 127-30 and Chiocci 2004 (both with previous bibliography). Ceraudo 2004d, 31-36. On the relationship between a Via Latina vetus (passing through Interamna) and a Via Latina nova (bypassing it) see Cagiano de Azevedo 1947, 39-43 and Coarelli 1998, 50. Note that neither the old road course nor Interamna itself is reported in the relevant parts of either the Itinerarium Antonini (It. Ant. 302.1-304.4 ed. Cuntz 1929) or the Tabula Peutingeriana (Segmentum V).




was never extensively re-occupied.19 Today only a few poorly preserved standing structures survive, very pale reflections of the ancient city.20 The area is now occupied by a few farm-buildings and a series of arable fields whose boundaries and access tracks clearly respect elements of the ancient street system. Decontextualized architectural fragments and inscriptions are to be found scattered all over the site and its neighbourhood, providing evidence of several cults and – more specifically – the probable existence of a temple dedicated to Jupiter Optimus Maximus.21 On the basis of the recovery of fragments of votive terracottas, an extramural sanctuary has also been tentatively located immediately to the north of the town.22

Previous archaeological work The whole of the Liri Valley has been the focus of significant studies by various institutions.23 Excavations and geophysical prospection have been carried out at Fregellae, Fabrateria Nova and Aquinum.24 The same is true of (more or less) systematic field surveys across the valley,25 which have benefited greatly from the invaluable contribution of aerial archaeology.26 What is remarkable, however, is the relative underrepresentation of Interamna Lirenas in recent fieldwork, especially given the significance even recently ascribed to it as a relevant case study which could shed light on important issues like imperial urban development and colonial settlement.27 As often happens, ‘new’ interpretations have to rely on ‘old’ data. Town Following a first topographic synthesis by Michelangelo Cagiano de Azevedo,28 over the decades local enthusiasts re-awakened interest in investigating the urban area, which by then had been heavily impacted by the introduction of mechanized agriculture. Early unsystematic collections of archaeological finds from the ploughsoil provided data for the creation of the very first map of the distribution of dated potsherds across the site.29 As the spread of Black Gloss pottery was so much wider and denser than that of any other class of material, it was suggested that the town flourished in the Republican period only to stagnate and enter a decline in the Early Empire. The overall picture also benefited from the analysis and processing of aerial photographs by French scholars, who established that the current field boundaries were largely informed by the original Roman layout.30

19 20

21

22 23



Cagiano de Azevedo 1947, 10-11. These include some cisterns (one in opus reticulatum with an aqueduct feeding it) and a (probable) sewer. Outside of the urban area proper, the most relevant are a bridge on the road to Casinum (over the Rio Spalla Bassa) and what remains of a bath complex (immediately south of Masseria Morra) (ibid., 14-32; Lena 1982). On whether or not a Capitolium: Cagiano de Azevedo 1947, 30-31. Hayes and Wightman 1984, 143. Both Italian (Soprintendenza per i Beni Archeologici del Lazio, Università degli Studi di Cassino, Università degli Studi di Perugia and Università del Salento) and non-Italian (British School at Rome, Centre de recherches d’histoire ancienne de Besançon, Deutsches Archäologisches Institut, McMaster University and University of Cambridge). To this list should be added

24

25

26

27

28 29 30

the activities promoted by the local comuni. Fregellae: Coarelli 1986; id. Coarelli 1998; Crawford and Keppie 1984; Crawford, Keppie and Vercnocke 1985; Ferraby et al. 2008. Fabrateria Nova: Beste et al. 2010; Ceraudo 2004c. Aquinum: Bellini 2007b; Bellini and Trigona 2011; Ceraudo 2004b; id. 2008; Nicosia and Ceraudo 2007. Ager Fregellanus: Crawford et al. 1986; Monti 1998. Ager Aquinas: Bellini 2007a; ead. 2008; Bellini and Trigona 2010; Ceraudo 2008. Lower Liri Valley: Bellini 2010; Hayes and Martini 1994. Chouquer et al. 1987, 124-30, 263-68, 271-80; Ceraudo 2004a. Patterson 2006, 101-6 (on imperial urban development); Pelgrom 2008, 347-49 (on colonial settlement). Cagiano de Azevedo 1947. Lena 1982, 68 Table II. Chouquer et al. 1987, 265; also Scardozzi 2004, 91-93.


The Canadian project led by the late Edith M. Wightman produced a pioneering integrated interpretation of the urban development of Interamna Lirenas which, although based on a new set of surface collections, also obviously profited from earlier work.31 The urban area was divided into 26 sectors of variable size (i.e. as defined by current field boundaries): diagnostic potsherds within each were then grab-sampled and their date ranges used to derive a dynamic view of occupation over time.32 The surface distribution did indeed conform to a picture of slow and irreversible decline, with the area of occupation contracting from about 30 ha in the second to first c. B.C. to fewer than 10 ha by the 1st c. A.D.33 This view was in a way remarkable as it seemed at variance with a rich epigraphic record, mainly Imperial in date, which attests individual acts of public munificence as well as the existence of a lively urban community well into the 4th to 5th c. A.D.34 However, closer scrutiny of the Canadian dataset could be taken to suggest a rather different reading of the very same evidence.35 Indeed, the total number of Republican Black Gloss potsherds (247) not only exceeds that of Early Imperial terra sigillata (91), or Mid- to Late Imperial African Red Slip (60), but constitutes about 60 per cent of the total finewares collected and is almost equal to the total number of fragments of amphorae (ca. 275). If pots were people – and they are not36 – this pattern would imply a Republican urban population which had declined by 64 per cent in the Early Imperial period and fallen even farther by Late Antiquity. Such a reconstruction makes the contrast with Imperial epigraphic evidence even harder to understand, but at this point two related issues should be noted. First, there is very clear evidence to demonstrate that the volumes of supply of ceramics varied considerably both through time and across the landscape.37 Although this issue has been explored for Gaulish sigillata38 and African Red Slip ware,39 it has not yet been examined in the context of either Black Gloss wares or Italic sigillata, although Roman Roth’s study of the former generally supports the contention that its production and distribution did vary in this way.40 Such a finding sounds a sharp note of caution against rushing into any very direct reading of the ceramic evidence from the Canadian survey. Second, in the absence of evidence of the volumes of pottery supply locally, it should be noted that, except where it has been possible to identify the precise form of a pot, the classes of Black Gloss and terra sigillata represent rather different chronological ranges: 290 and 160 years respectively.41 Therefore, by rule of thumb, the former could have been almost twice as likely to end up into an archaeological deposit and therefore is roughly twice as visible archaeologically. Taking into consideration that the potters at Interamna Lirenas are likely to have begun producing their own Black Gloss fineware as early as 250 B.C, whereas both terra sigillata and African Red Slip had to be imported,42 extreme caution is advisable and we should avoid placing too much reliance on absolute numbers of sherds per se. As these surface collections were not made systematically, even more

31 32 33 34

35

Hayes and Wightman 1984. Ibid., 137-38. Ibid., 143-45. As remarked by Wightman and Hayes 1994, 38. It should be noted, however, that this local pattern needs to be calibrated against the background of increased epigraphic activity in the Imperial period. A comprehensive list of inscriptions is (again) provided by Michelangelo Cagiano de Azevedo, and has now been supplemented by Kajava (Cagiano de Azevedo 1947, 49-52; Kajava 1996, 191-96). The following considerations are based on the original survey dataset presented in Hayes and Wightman 1984, 142 Table I and at p. 144 Table II.

36 37

38 39 40

41

42

See Sbonias 1999. See originally Millett 1985, developed in id. 1991. Many commentators have not appreciated that the original observations were concerned with locally produced wares, not the more widely traded finewares. Marsh 1981. Fentress and Perkins 1989. Roth 2007. A systematic study of the volumes of production of Black Gloss ware and Italic sigillata is clearly a fundamental priority for future research. Black Gloss: 330-40 B.C.; terra sigillata: 30 B.C. - A.D. 130. See Launaro 2011b, 86 Table 4.1. Hayes and Wightman 1984, 145.




circumspection about their interpretation is in order.43 Furthermore, a rather less emphatic shift between Late Republic and Early Empire is perhaps also suggested by the relative proportions of Republican Dressel 1 and Early Imperial Dressel 24 amphorae.44 If numbers per se are of limited value, the overall presence of dated potsherds across the survey sectors might offer a better insight into urban development. Pottery classes have therefore been grouped into three main periods (Republic, Early Empire and Mid- to Late- Empire)45 and their occurrence mapped (fig. 2). Remarkably enough, out of the 26 sectors, 19 (73%) produced Republican potsherds, 20 (73%) Early Imperial and 17 (65%) FIG. 2. The presence of Roman fineware classes across the Mid- to Late Imperial ones; 14 sectors (54%) Canadian survey sectors as evidence of possible occupaproduced pottery from all three periods. This tion at Interamna Lirenas (based on Hayes and Martini evidence does not seem to suggest Early Imperial 1984, 137 fig. 1, 142, table I and 144, tab. II). stagnation or a sharp decline into Late Antiquity. Taken by themselves, these considerations do not prove anything, except perhaps the fact that the evidence on which scholars have based their interpretation is – at best – rather impressionistic and at worst fundamentally flawed. It would make better sense of the other evidence available (e.g. imperial inscriptions) to assume a certain degree of stability (if not growth) in the urban development of Interamna Lirenas between Republic and Empire. This impression is reinforced when the possible triumviral (colonial?) intervention mentioned by the Liber coloniarum is taken into account. Territory One topic closely related to Roman colonization is the question of the parallel development of urban and rural landscapes. The French contribution to reconstructing patterns of colonial development in the valley is certainly controversial,46 but should not be dismissed out of hand. Building on an integrated reading of both aerial photographs and the Corpus Agrimensorum, the équipe of the University of Besançon endeavoured to look closely at how the Roman theory of land allotment intersected with the landscape evidence on the ground, concluding that the original land allocation (4th c. B.C.) took the form of a strigatio/scamnatio – that is a system laid out in oblong parallel bands – rather than the more familiar chessboard-style centuriation.47

43 44



Hayes and Wightman 1984, 138. Notional data ranges are 150-30 B.C. / 120 years (Dressel 1) and 50 B.C. - A.D. 125 / 175 years (Dressel 2-4) (Launaro 2011b): 86 Table 4.1). It is worth noting that the relevant Table conflates Dressel 1 and Dressel 2-4 amphorae of Campanian origin into one total figure (23), effectively inflating the actual number of Republican amphorae. If, only 5 of those 23 Campanian amphorae were Dressel 2-4, this would result in 18 (=23-5) Dressel 1 over 130 years and 24 (=19+5) Dressel 2-4 over 175 years, both yielding an

45

46 47

average of 0.14 sherds per year (Hayes and Wightman 1984, 144 Table II). With reference to Hayes and Wightman 1984, 142 Table I and at p. 144 Table II: Republic (‘v. nera’ and ‘[anfore] campane’); Early Empire (‘terra sig.’, ‘Dressel 2-4 diverse’, ‘[locali] Dressel 2-4’); Mid- to Late Empire (‘sigillata africana’, ‘[anfore] africane’). Note that ‘[anfore] campane’ in fact include some Dressel 2-4 (see n. 44). Quilici 1994, 130-31. Chocquer et al. 1987, 124-25, 263-68.


PERIOD Early Republic (300-200/150 B.C) Late Republic (200-1 B.C.) Early Empire (A.D. 1-100) Middle Empire (A.D. 200-300) Late Empire (A.D. 300-600)

SCATTERS

SITES

6 52 25 27 7

TOTAL

MINOR

MAJOR

11 31 28 28 7

15 12 17 17 6

32 95 70 72 20

Tab. 1. Surface finds within the territory of Interamna Lirenas as recorded by the Canadian survey (based on Hayes and Martini 1994, 173-236; territorial definition as per Cagiano de Azevedo 1947, 38 fig. 3).

Consequently, the Canadian survey team of the territory of Interamna Lirenas was rather surprised by the remarkably low numbers of smaller rural sites dated to the earlier phases of the colony (i.e. mainly 3rd c. B.C.),48 especially in contrast to their presence in later periods (see tab. 1) and given the extensive evidence of contemporaneous urban occupation (see above).49 This is at variance with the apparent existence of a strigatio/scamnatio (see above). Tentative explanations of this pattern have postulated a primarily urban population,50 lower levels of archaeological visibility51 or a more nucleated settlement pattern made up of few, but fairly populous villages.52 This putative initial discrepancy between urban and rural development would have gradually leveled out as both town and country have been assumed to have peaked archaeologically in the second to 1st c. B.C., only to commence a slow, irreversible decline in the Early Imperial period.53 As already discussed the evidence for urban development is open to debate – at least as much as the interpretation based on it. A similar argument can be adduced for the surrounding landscape. The very limited recovery of early colonial farms (3rd c. B.C.) might be a result of the comparatively low intensity of coverage in the Canadian survey.54 Moreover, it is also relevant that there might have been some clustering around the two main nucleated settlements in the area (i.e. Interamna Lirenas and a putative village to the west) and nothing in between.55 However, given the likely role towns and villages played in supplying and distributing the categories of material culture which modern archaeologists have since largely relied on, it is also possible to explain this pattern as the result of decreasing archaeological visibility the farther the investigator moves away from the principal settlements. Furthermore, one needs to consider the fact that before the mid-3rd c. B.C. Black Gloss pottery – the primary diagnostic evidence for this period – had to be imported from outside the region: the relatively higher number of sites from the 2nd c. B.C. could have been caused by a distortion of the dataset arising from the initiation of local Black Gloss production and hence its increased availability. Again, all these observations are not meant to prove that the traditional interpretations are wrong, but merely to show that evidence produced by methodologies common in the 1970-1980s is not necessarily best suited to finding an answer to questions raised forty years later. In a nutshell, it is not just a matter of renewing the research agenda, but data suitable to addressing current questions has to be produced to complement, integrate and improve the existing evidence. These two interlinked aspects are the subjects of the following sections.

48 49

50 51

Wightman and Hayes 1994, 36. In contrast, pre-Roman phases are comparatively well attested (Wightman 1994b), although the density of sites does seem rather low, perhaps the result of the modest intensity of the survey. Hayes and Wightman 1984, 143. Wightman and Hayes 1994, 36.

52 53 54

55

Pelgrom 2008, 368. Wightman and Hayes 1994, 36-38. The approach adopted by the Canadian team involved “archaeologists […] spaced at intervals of 15-25 m, zigzagging slightly” (Wightman, Hemphill and Hayes 1994, 3). Pelgrom 2008, 348-49.




FIG. 3. A view of the site of Interamna Lirenas.

Current research questions One characteristic makes Interamna Lirenas stand out among Roman towns in Italy: it was abandoned at the end of Antiquity, never to be reoccupied again or overbuilt by later phases (fig. 3) and, despite later spoliation and some probable plough damage, its buried archaeology is still likely to preserve a largely complete urban layout and intact stratification. This situation is not uncommon, but what makes Interamna Lirenas all the more remarkable is that it has almost never been the site of an excavation so it remains a virtually untapped archaeological resource.56 Furthermore, its surroundings in the valley are composed of an almost uninterrupted series of open fields (fig. 4), whose characteristics (e.g. geomorphology, ploughing practice) make them highly suitable for field survey. In other words, Interamna Lirenas and its territory have the potential to offer a unique window onto the past landscape, so good in fact that urban and rural archaeology can finally be approached in a fully integrated manner. This would be no little achievement and would make a cogent contribution to current debates.57

Roman colonization Recent years have witnessed some growing dissatisfaction with established views about Roman colonization in Italy and the concomitant beginnings of its re-appraisal. 58 Whereas some earlier scholars focused on broad similarities and therefore assumed long-term uniformity in Roman colonial practice, 59 a new wave of studies, inspired by both a closer scrutiny of the evidence and current theoretical debates about identity, has put renewed stress on specific differences, chronological discrepancies and local influences.60

56

57 58 59



E.g. Bispham 2006; Bradley 2006. One aspect whose peculiarity was already remarked on by Cagiano de Azevedo: “[…] mai infatti, non dico una campagna sistematica, ma neppure saggi sporadici sono stati eseguiti in quel luogo pur così promettente” (Cagiano de Azevedo 1947, 21). Millett 2010. See also contributions in this volume. Focus of such a re-assessment is usually Salmon 1969 (standard work on the subject in English) and Brown

60

1980 (usually presented as an ‘archaeological counterpart’ to the former). Cf. Pelgrom and Stek in this volume. The debate about ‘Romanization’ is explicitly evoked (Bispham 2006, 75). For the Republican phase of Roman expansion see especially contributions in Van Dommelen and Terrenato 2007. For the vigour of the ‘identity/imperialism’ debate consider most recent contributions by David Mattingly and Martin Millett (Mattingly 2011; Millett 2010).


FIG. 4. A view of the landscape across the Liri Valley (Monti Aurunci in the background).

What is especially being questioned is the nature and value of assumptions about Roman colonization and the ways these have created normative expectations in what has been read and excavated. Such assumptions do not take full account of the time factor: while Roman colonial practice is thought to have evolved through time, later literary evidence seems to have led to the imposition of later assumptions on the understanding of earlier phases, thereby seeking regularities which might be illusory.61 This possibility has been raised with explicit reference to a centralized, dirigiste view of colonization, which assumes that it emanated from Rome with the intention of reproducing the original political structures of that city and its underlying spatial layout (cf. Sewell in this volume). Interestingly, several pieces of evidence from earlier phases do not seem to be consistent with this hypothesis. These discrepancies are thought to pertain especially to religious aspects and the presence of temples dedicated to the Capitoline Triad (Jupiter, Juno and Minerva) as an intentional emulation of the Capitolium in Rome (cf. Boos and Torelli in this volume). Edward Bispham argues that earlier colonization (until the late 3rd c. B.C.) had placed much more emphasis on other divinities – namely Hercules – as a useful vehicle for local integration (i.e. indigenous peoples and their cults) at a time when Roman success in Italy was anything but certain (cf. Liberatore in this volume).62 This situation would have changed once Roman supremacy over Italy had been sanctioned by events, after which it would have been “unthinkable that a Roman colony should not have a temple to the Capitoline triad”.63 Whereas previous scholarship has put much more emphasis on 338 B.C. (the dissolution of the Latin League) as a more crucial caesura in the development of Roman colonization,64 this recent view of the evidence points to a significantly later date. What changes is the way the nature of (early) Roman imperialism in Italy is framed – and interpreted – especially in relation to indigenous peoples. At the same time, the exploration of urban sites with the help of extensive geophysical surveys has also raised questions about the simple categorization of sites as colonies or otherwise (cf. Vermeulen in this volume). A good example is the survey of Falerii Novi, a site which although founded de novo after Rome

61 62

E.g. Gell., NA 16.13.8-9. Bispham 2006, 113-22.

63 64

Ibid., 122. Also Coarelli 1992, 24.




Fig. 5. The layout of the Roman town of Falerii Novi.

razed Falerii Veteres in 243 B.C. is generally agreed not to have been the site of a colony at that period.65 Nevertheless, a close study of its primary grid suggests that it was divided into regular allotments not dissimilar to those identified at Cosa where these divisions are identified with the differential status of different groups of colonists (fig. 5).66 However, at Falerii Novi it is also possible to see the way in which the plan of the town seems to have been influenced by the layout and sacred topography of its predecessor, Falerii Veteres, in a manner which appears entirely at odds with Roman traditions. Despite an historical rhetoric which emphasizes that the new town was built in a less defensible location than its predecessor, 67 it is evident that it was designed in such a way that the landscaping gave its walls considerable prominence, and their course does not reflect the urban grid but was built instead to mirror the plan of Falerii Veteres. Finally, the distribution of a series of temples around the peripheral road inside the walls seems to create a sacred route which literally connected the new town with the old.68 However these issues in understanding Falerii are resolved, the evidence surely casts doubt on common preconceptions about the exclusive regularity of colonial town planning, and indeed core concepts about Roman Republican urban planning in general. Therefore, the Liri Valley does indeed provide a privileged observatory from which to engage with many of these issues. Both Fregellae and Interamna Lirenas were founded only 25 km and 16 years apart, in the decades immediately following the key date of 338 B.C. The lay-outs of both towns were bisected by the Via Latina69 and the topographical positions they occupy are very similar,70 and their foundations are linked to the same historical context: the southward expansion of Rome during the Samnite Wars.71 Indeed a unitary plan (a dirigiste strategy?) might be seen to have been in operation here – as clearly suggested by Filippo Coarelli.72 Furthermore, the

65 66 67 68 69



Keay et al. 2000; Hay et al. 2010. Millett 2007, 73-75. Zonar. 8.18; see Keay et al. 2000, 1-2. Millett 2007, 77-81. The Via Latina constitutes the main axis of the urban layouts of both Fregellae (Coarelli 1998, 55-56) and Interamna Lirenas (Cagiano de Azevedo 1947, 21-24).

70

71 72

Both are located on relatively flat plateaus, bounded on two sides by steep slopes descending into adjacent river valleys, arguably betraying either their military origin in terms of control and defence of the area or a desire to occupy visible locations in the landscape. See Wightman 1994c. Coarelli 1988, 39-41.


associated strigatio/scamnatio (see above) is generally assumed to have borne no relationship to previous settlement patterns (although these are almost unattested in past surveys across the valley floor).73 Therefore, it has often been taken to represent the quintessential mark of a new Roman landscape. As yet no Capitolium has been identified at Fregellae, but one cult place has been tentatively ascribed to Hercules.74 On the other hand, an Early Imperial (?) inscription from Interamna records the restoration of a temple to Jupiter Optimus Maximus (see above), possibly a Capitolium (although the two things are not necessarily one and the same). What is remarkable is that local devotion to Hercules is also attested in two other inscriptions,75 possibly hinting at the existence of some sort of sacred place or sanctuary to the god within the territory of the colony (for the existence of a Via Herculanea see below). The local popularity of this cult is also implied by the presence at Aquinum of a specific religious collegium, the cultores Herculis.76 Although this local popularity might be interpreted as evidence of an earlier devotion to Hercules as a ‘god of colonization’,77 it might be explained alternatively in relation to the role of such a god in a pastoral society.78 To sum up, several key elements in current debates are highly relevant to this area, specifically to Interamna, and therefore they certainly deserve further investigation.

Town and country During the last few years the relationship between population density and settlement patterns has attracted considerable scholarly attention.79 As noted above, inferring population levels from survey evidence is definitely a tricky exercise. In its efforts to move from site numbers to absolute population figures, the seminal paper by Robert Witcher describing the suburbium of Rome and its population has intersected with the wider debate on the population of Roman Italy.80 Despite its consequent popularity, his approach suffers from several critical difficulties which mean that its conclusions cannot be accepted as definitive.81 The key debate hinges on whether the Italian free population between Late Republic and Early Empire was relatively small and in decline, or larger and rising. Given the coeval and unprecedented urban population growth (attested by the increasing number of towns), any decline in the free Italian population would have been matched by a similar decline in its rural component, readily supplanted by slaves.82 An examination of the relative trends for farms and villages (settlement types usually related to a primarily free population) extrapolated from 27 survey areas across Italy reveals that more often than not their number actually increased between 200 B.C. and A.D. 100.83 In the lower Liri Valley, this approach has shown that rural settlement in the area was also subject to such an increase.84 However, if Interamna was indeed shrinking, this later process might be taken to have mitigated – or even negated – any overall population growth in the area. In other words, Interamna Lirenas and its territory provide a key test case for the above interpretation – and the methodology which produced it – by looking at town and country in an integrated project and as part of the same landscape. Any research on settlement patterns in Roman Italy cannot avoid engaging in the debate on the nature of

73 74

75

76 77 78

Wightman 1994b, 14. Coarelli 1998, 61-62. It is of uncertain (Republican) date and must not to be confused with the other extramural sanctuary dedicated to Aesculapius (built in the 2nd c. B.C.). CIL X 5366 reports both inscriptions. Their description was significantly revised by Kajava 1996, 191-93 (also AE 1996, 333-34). CIL X 5386. Bispham 2006, 113-17. Coarelli 1998, 62.

79 80

81

82 83 84

E.g. Bintliff and Sbonias 1999; Osborne 2004. Witcher 2005. A recent presentation of this most farreaching debate is Scheidel 2008, to be integrated with Lo Cascio 1999. Also Launaro 2011b, 11-50. Wilson 2008; Yntema 2008; Fentress 2009; Attema and De Haas 2011. Detailed critiques of Witcher’s approach are provided by Scheidel 2008, 49-54 and Launaro 2011b, 55-70. Hopkins 1978, 68-69 table 1.2. Launaro 2011b, 149-64. Ibid., 131-33.




what is called the ‘villa economy’, often equated with a ‘slave mode of production’, which has been assumed to constitute a universal norm.85 Nevertheless, it is becoming increasingly evident that this equation is far from universally justifiable. Problems arise from how to categorize sites on the basis of the archaeological evidence, and how to relate these to the term villa as used in texts, but there is also a growing recognition that the architectural elaboration which characterizes villas is a matter of complex cultural choices not linked in any straightforward manner to any particular mode of production.86 Not only was there a structural need for seasonal (free) labourers,87 but we also observe a chronological discrepancy between the peaks in numbers of villas and in volumes of Italian exports.88 More recently a debate has arisen about the evidence for slave barracks.89 Taken in conjunction with the archaeological evidence for a growth in the numbers of farms and villages (and the underlying free population), these factors suggest a much more varied settlement pattern and economic system, whose character is currently very much open to debate. It seems clear that tenancy agreements constituted a much more widespread alternative to slave production than has usually been assumed.90 Although the ‘villa-economy’ is also generally related to intensive agriculture based on cash crops for Mediterranean export, researching the town of Interamna Lirenas in an integrated study of its territory – some of it a centuriated territory for at least part of its history – provides an excellent opportunity to investigate the nature of the economy in a more differentiated landscape and thereby contribute to the more general debate. Previous Canadian survey work in the area has already suggested that this landscape was dominated by small farmsteads, interspersed with only a small number of villas and villages.91 To refine present understanding, a more complete knowledge of how the size and distribution of sites changes through time, how this relates to patterns of colonial land allotment and how settlements were linked to local and longer distance networks of production, distribution and consumption is required. The area also provides a rare opportunity to investigate the utilization of a Roman landscape through a survey in the context of its centuriation. Past approaches to centuriated landscapes have treated land allotment from an idealized perspective, leaving the actual character of settlement in the individual landscape units unquestioned and largely uninvestigated.92 The results of the recent large-scale geophysical survey on the Isola Sacra also hint that the reality of land allotment on the ground might have run along rather less neat, clear-cut lines than those superimposed on small-scale maps might imply.93

Local networks In the last few decades, there has been a growing awareness of a critical limitation in the present understanding of archaeological assemblages, which is still too heavy reliant on particular classes of pottery, namely finewares and amphorae. Their broad distribution (all over the Mediterranean) and physical qualities have made them a most suitable diagnostic material by which to identify and date a wide array of archaeological sites. Unsurprisingly, sites at different levels of the settlement hierarchy and in different places with respect to the geography of distribution had a differential access to these products. This factor not only affects the chances of identifying less affluent or more marginal sites (the vast majority), but also shapes the interpretation of social and economic relationships across the landscape, generally over-emphasizing export-oriented production and longdistance distribution at the expense of less visible local consumption patterns.94 This almost invisible network can only be identified if the very basic forms of material culture, by their very nature more common at all levels

85

86 87 88 89 90



The foundational work on the subject is Giardina and Schiavone 1981. Millett 2010. Rathbone 1981, 15. Terrenato 2001, 26-27. Marzano 2007, 148-53. Launaro 2011a; id. 2011b, 170-77. This is subject of

91 92

93 94

the research by Alessandro Launaro as part of his British Academy Postdoctoral Fellowship (2009-12). Wightman and Hayes 1994, 36-39. See now, however, the valuable contribution by De Haas on the Pontine region (De Haas 2011). Germoni et al. 2011, 238, figs 12.4-12.5. Witcher 2006, 49-52.


of society, such as locally-produced coarseware pottery are taken into account. Nowadays their high information potential is widely recognized,95 although their analysis and classification still poses difficulties on account of their local distribution and a dearth of well-dated local sequences. The latter is primarily a consequence of the fact that such material has traditionally been overlooked in finds-processing from both survey and excavation. Roman Italy is generally well-served by knowledge of the major classes of pottery. Production centres of both finewares (Black Gloss and terra sigillata) and amphorae throughout the Republican and Early Imperial periods are known, and local production sites are often well defined,96 even the lower Liri Valley probably produced its own Black Gloss pottery (see above).97 In contrast to both earlier and later phases, this activity is likely to have impacted on the archaeological visibility of the Late Republican period – therefore it should be properly accounted for by calibrating finds quantification (see above). Moreover, as Italian products were eventually supplanted by provincial imports in the course of the 2nd c. A.D., especially from Northern Africa, the situation changed. It has been established that African Red Slip relied on a different pattern of both production and distribution,98 not always resulting in an even supply and ready accessibility across Italy. Hence, Mid- to Late Imperial landscapes are also likely to suffer from the problems of unreliable diagnostic evidence as outlined above. This is precisely the point at which the potential of coarsewares emerges as an aid to assist in illuminating the patterns of local and regional networks. Although it is necessary to be aware that the volume of their supply and use is also likely to have varied over space and time, it is reasonable to assume that it probably did so to a much lesser extent than did the finewares imported from the shores of North Africa. Another aspect of this subject is communications. Although the Via Latina has certainly not been overlooked, knowledge of the existence of secondary routes is the vital clue which would provide the best complement to ceramics in pursuing any investigation and understanding of local networks. Again, although Interamna seems literally to have lost its place along the main route through the Liri Valley after the construction of the Via Latina Nova, it still remained a point of convergence from both Aquinum and Casinum, providing a connection across the Aurunci mountains to the Via Appia and the Tyrrhenian coast near Minturnae. Although this route is traditionally known as the Via Herculanea,99 it had nothing to do with the road of the same name mentioned by Cicero, a point made by Angelo Nicosia.100 Whether such an appellation is original (perhaps related to the presence of a sanctuary to Hercules along its route: see above) or a modern, erudite reconstruction remains open to debate. What matters most here is the probable existence of a secondary road network centred on Interamna Lirenas, whose discovery might shed pertinent light on the nature of production, distribution and consumption on a local scale.

Methodology Now that the main issues have been addressed, the time has come to turn to the particular study of Interamna Lirenas and its territory. At the outset it is essential to discuss the methodology most suitable to succeed in this endeavour. The first prerequisite of our research strategy involves a thorough and systematic (re-)consideration of what is already known (both published and unpublished) and the second is the production/gathering of fresh evidence. In an area with such an impressive research tradition, it is impossible not to engage actively and constructively with all the actors involved – both now and in the past. In an examination of the past research, the expectation is that the different questions which can now be put will uncover a different landscape. However, only new fieldwork can provide the fresh evidence needed and, accordingly, three main tasks have been identified which will all be integrated within a single GIS:

95

96

E.g. Olcese 1993; Bats 1996; Santoro Bianchi and Fabbri 1997; Cortese 2005. Also MacDonald 1995 and Hayes 2000 with specific reference to landscape archaeology. Launaro 2011b, 89-91.

97 98

99 100

Also Williams 1994, 161. E.g. Fentress et al. 2004, 148-49; Launaro 2009, 29; Pasquinucci and Launaro 2009, 190. Cagiano de Azevedo 1947, 43-44. Cic., Leg. agr. 2.36; Nicosia 2008, 209-11.




I) a geophysical prospection and topographic survey of the whole site of the urban area of Interamna Lirenas; II) an intensive field-survey of a representative area within its territory (ager Interamnas); III) a thorough analysis of the local material culture (especially coarseware and building materials) recovered from systematic surface collection. The settlement patterns so mapped will be complemented by a critical appraisal of linear landscape features (e.g. roads, ditches) as possible remains of Roman field-boundaries and land allotment schemes (tentatively identified as such by previous archaeological research). In the longer term, funding permitting, the aim is to develop an extensive program of geophysical surveying to explore this rural landscape following the model of such survey work developed in northern England by Dominic Powlesland.101 For the first time this would permit a large-scale empirical evaluation of settlement sites and land divisions in a centuriated landscape. Interamna Lirenas As already noted, virtually nothing certain is known of the urban layout of Interamna Lirenas. Earlier studies have outlined its extent and noted some of its features and suggest hypothetical reconstructions of the street grid. They have also intimated the identification of specific functional areas (e.g. productive quarters). Nonetheless, much better quality data are needed if the questions set out above and, specifically, the concept of centralized direction in the establishment and planning of such colonies are to be tackled. Thanks to extensive excavation and some limited geophysical prospection, something is known of the plan of the contemporaneous center of nearby Fregellae. It will therefore be illuminating to compare this and other Mid-Republican colonial plans with that of Interamna Lirenas, highlighting differences and similarities in planning, and noting the presence or absence of distinctive building types such as the circular comitium or tripartite-cell temple. Thanks to a long history of successes in the study of urban Roman sites, a full coverage including geophysical prospection and a detailed topographical survey seem to offer the solution.102 Compared to other possible techniques (like georadar or electrical resisistivity survey), magnetometry is that best suited to produce an initial overall view of Interamna Lirenas, especially given the local geology and the nature of the known archaeological remains. It is a swift, efficient technique which works by measuring minor changes in the Earth’s magnetic field, leading to the detection of different types of features including kilns, hearths, ovens, ditches and walls, especially when fired ceramic material or tuff have been used in their construction. The magnetometry will be complemented by the focused use of other geophysical prospection techniques either to investigate individual structures or to answer particular questions posed by the site. Results will be optimized when the prospection grid is laid out at an angle to the orientation of known archaeological features.103 A precise survey of the micro-topography will complement the interpretation of the results of the geophysical survey, since the situation of individual buildings can be understood in relation to the local surface topography and the overall townscape can be appreciated within the scope of the broader topography. This latter fact means that aspects of the town planning will be better understood, especially in the context of phenomenological approaches. The goal is to cover the whole intramural area (ca. 30 ha) and, given the supposed existence of an extramural sanctuary – like that attested at Aquinum – which might shed further light on local political, social and cultural relationships as mediated by religion and cults,104 it is also intended to survey the surrounding area.105 This will be integrated with fieldwalking across the same areas. The analysis of the distribution of surface finds will provide evidence of the chronological development (e.g. changing size), and might also suggest the function(s) of specific sectors (as the Canadians have already done), or reveal differential distributions of

101 102 103



http://www.landscaperesearchcentre.org/. For comparative work see Johnson and Millett 2012. It reveals features parallel to the street grid more clearly.

104 105

E.g. Stek 2009. A similar approach was applied at Falerii Novi (Hay et al. 2010).


specific artefacts (indicative of status or cultural choices). Such evidence will complement the results of the magnetometry, for instance, by showing the materials used in the construction of particular buildings. More specifically, it might also be possible to identify recurring patterns in the ploughsoil assemblage in relation to particular buildings identified through the geophysical survey, perhaps allowing an investigation of the very early phases of the colonial urban landscape. At this stage, limited test excavation can be used to provide data at a higher resolution. The reconstruction of different stages in the development of the urban plan should also contribute to the broader demographic debates, especially as it is sometimes suggested that many sectors within the walled area of Roman towns were actually devoid of buildings.106 Ager Interamnas The issue of population leads to the contribution of field-survey and landscape archaeology. The general patterns which provide only very limited evidence of early settlement in the territory of a colony have been interpreted in several ways. One traditional explanation has ascribed its dearth to the nature of the evidence itself (i.e. weak traces of small farms), not very visible across the landscape.107 A variation on this explanation is that smaller farms were less integrated into the pottery-using economy, and therefore intrinsically less visible in a survey than even marginally larger sites.108 More recent contributions have suggested that the absence might instead reflect the original settlement pattern itself with an earlier colonial occupation phase which relied on a nucleated system of villages (hence the limited occurrence of isolated rural houses).109 This issue is fundamentally linked to the level of intensity and coverage of any survey. Field practice has confirmed that the more intensive a survey, the higher the numbers of sites recovered and this is especially true with reference to smaller sites (which might otherwise be heavily under-represented). However, any approach which utilizes very high intensity survey to counter these problems can only reduce the overall area covered, meaning that in this smaller sample area there is less likelihood that a nucleated settlements (like towns or villages) will be found as they are fewer and more widely spaced from each other than smaller sites. Ironically, despite being highly visible when surveyed, an intensive survey has a fairly high chance of missing them.110 Consequently, there is an unresolved tension between the resolution and scale of the analysis,111 and, while some questions might be better answered by opting for one or the other, the ‘farms vs villages’ problem falls perfectly in between. It would seem that the only way to address this issue is to undertake a new intensive field-survey (i.e. with walkers on parallel lines 5 m apart covering the whole landscape), across a very sizeable transect across the landscape. This will facilitate the recovery of smaller sites and lower density concentrations (e.g. associated with pre-Roman and Early Medieval settlement), which might otherwise be missed, and will also allow an identification of those crucial – fewer and widely spaced – larger sites if they exist. In order to maximize recovery and to make the sampling of sites more consistent and better suited to comparative analysis, by and large only ploughed fields will be surveyed. Furthermore, if sites from lower levels of the settlement hierarchy and at the very margins of distribution networks are to be identified, special emphasis will have to be devoted to the study of regionally or locally produced coarseware pottery.112

Concluding remarks The aim of this paper has been to explore the many issues involved in the study of Roman colonization from the point of view of settlement and landscape, and also to show how this field of study is continuously

106 107 108 109 110

De Ligt 2008, 147-54. Rathbone 1981, 20-21; also id. 2008. Millett 1991. Pelgrom 2008. Launaro 2011b, 154-55.

111 112

E.g. Terrenato 2004. This is facilitated by the fact that there is a growing literature on the subject and regional syntheses relevant to the area have since begun to appear (e.g. Bats 1996; Hayes 1994; Olcese 1993).




developing by identifying new questions and employing new data to address them. These factors have been taken into account in the design of a new field project, and fieldwork from the first seasons (in 2010-11) has adequately shown the feasibility of such an undertaking, not to mention the high potential of the methodology as applied to such research questions.113 One of the inherent strengths of the study area is that it allows a full integration of urban and rural landscape archaeology in a way which has rarely been achieved in the past. Even though such an approach was previously adopted by our Canadian predecessors, their work preceded – and in this sense missed out on – fundamental developments in the practice of remote sensing and the use of Geographical Information Systems as well as important improvements in the study of material culture. By going back to Interamna Lirenas, it is hoped that the same spirit can be carried into the third millennium, acknowledging the fact that we are just part of an ongoing discourse and that ever different waters will keep flowing.

Acknowledgements The Roman Colonial Landscapes project is being made possible by the generous contribution of the British Academy, the Faculty of Classics (University of Cambridge), the McDonald Institute for Archaeological Research (University of Cambridge) and the Comune di Pignataro Interamna. It also benefits from extensive support from the British School at Rome, the Archaeological Prospection Services of the University of Southampton and the Soprintendenza per i Beni Archeologici del Lazio. Invaluable expert supervision has been provided by Sophie Hay (geophysical prospection) and Ninetta Leone (finds processing). Our deepest gratitude goes to the cittadinanza of Pignataro Interamna for their wonderful hospitality.

113



Preliminary results in Hay et al. 2012; Hay et al. forthcoming; Bellini et al. 2012; Bellini et al. 2013. See also www.classics.cam.ac.uk/faculty/rcl/.


References Attema, P. A. J. and T. C. A. de Haas 2011. “Rural Settlement and Population Extrapolation. A Case Study from the ager of Antium, central Italy (350 BC - AD 400),” in A. Bowman and A. Wilson (edd.), Settlement, Urbanization, and Population (Oxford) 97-140. Barker, G. and D. J. Mattingly (edd.) 1999-2000. The archaeology of Mediterranean landscapes, 5 vols. (Oxford). Bats, M. 1996. Les ceramiques communes de Campanie et de Narbonnaise (Ier s. av. J.-C.-IIe s. ap. J.-C.): la vaisselle de cuisine et de table (Naples). Bellini, G. R. 2007a. “L’ager di Aquinum. Ricerca e tutela nel 2005,” in G. Ghini (ed.), Lazio e Sabina, vol. 4: Quarto incontro di studi sul Lazio e la Sabina: atti del Convegno: Roma 2931 maggio 2006 (Rome) 241-44. Bellini, G. R. 2007b. “La necropoli occidentale di Aquinum,” in G. Ghini (ed.), Lazio e Sabina, vol. 4: Quarto incontro di studi sul Lazio e la Sabina: atti del Convegno: Roma 29-31 maggio 2006 (Rome) 245-50. Bellini, G. R. 2008. “Il progetto ‘L’ager di Aquinum – Conoscere per tutelare’,” in C. Corsi and E. Polito (edd.), Dalle sorgenti alla foce (Rome) 157-65. Bellini, G. R. 2010. “L’attività di ricerca e tutela negli anni 2003-2008 lungo la bassa valle del Liri. Materiali per una carta archeologica,” in G. Ghini (ed.), Lazio e Sabina, vol. 6: Sesto incontro di Studi sul Lazio e la Sabina: atti del convegno: Roma 4-6 marzo 2009 (Rome) 471-74. Bellini, G. R. et al. 2012. “Archaeological Fieldwork Reports: Interamna Lirenas,” PBSR 80, 358-60. Bellini, G. R. et al. 2013. “Interamna Lirenas (Comune di Pignataro Interamna, Provincia di Frosinone, Regione Lazio),” PBSR 81, 358-60. Bellini, G. R. and S. L. Trigona 2010. “L’ager di Aquinum. L’attività di ricerca e tutela nel 2007 e 2008,” in G. Ghini (ed.), Lazio e Sabina, vol. 6: Sesto incontro di Studi sul Lazio e la Sabina: atti del convegno: Roma 4-6 marzo 2009 (Rome) 475-79. Bellini, G. R. and S. L. Trigona 2011. “La necropoli occidentale di Aquinum. IV secolo a.C.-IV sec.

d.C.,” in G. Ghini (ed.), Lazio e Sabina, vol. 7: Settimo incontro di Studi sul Lazio e la Sabina: atti del Convegno: Roma 9-11 marzo 2010 (Rome) 479-92. Beste, H.-J. et al. 2010. “San Giovanni Incarico (Frosinone): ricerche topografiche e archeologiche sul sito di Fabrateria Nova,” in G. Ghini (ed.), Lazio e Sabina, vol. 6: Sesto incontro di Studi sul Lazio e la Sabina: atti del convegno: Roma 4-6 marzo 2009 (Rome) 45769. Bintliff, J. and K. Sbonias (edd.) 1999. Reconstructing Past Population Trends in Mediterranean Europe (Oxford). Bispham, E. 2006. “Coloniam Deducere: How Roman was Roman Colonization during the Middle Republic?,” in G. J. Bradley and J.-P. Wilson (edd.), Greek and Roman Colonisation: Origins, Ideologies and Interactions (Swansea) 73-160. Bradley, G. J. 2006. “Colonization and identity in Republican Italy,” in G. J. Bradley and J.-P. Wilson (edd.), Greek and Roman Colonisation: Origins, Ideologies and Interactions (Swansea) 161-87. Brown, F. 1980. Cosa: the making of a Roman town (Ann Arbor, MI). Cagiano de Azevedo, M. 1947. Interamna Lirenas vel Sucasina (presso Pignataro Interamna): Regio I: Latium et Campania (Rome). Cagiano de Azevedo, M. 1949. Aquinum (Rome). Campbell, J. B. 2000. The writings of the Roman land surveyors: introduction, text, translation and commentary (London). Carettoni, G. F. 1948. Casinum (Rome). Ceraudo, G. 2003. “Via Latina: da Fabrateria Nova a Casinum,” in M. Guaitoli (ed.), Lo sguardo di Icaro (Rome) 442-45. Ceraudo, G. (ed.) 2004a. Ager Aquinas. Aerofotografia archeologica lungo la valle dell’antico Liris (Marina di Minturno). Ceraudo, G. 2004b. “Aquinum: la città romana,” in G. Ceraudo (ed.), Ager Aquinas. Aerofotografia archeologica lungo la valle dell’antico Liris (Marina di Minturno) 13-23. Ceraudo, G. 2004c. “Fabrateria Nova,” in G.




Ceraudo (ed.), Ager Aquinas. Aerofotografia archeologica lungo la valle dell’antico Liris (Marina di Minturno) 81-84. Ceraudo, G. 2004d. “Via Latina: da Fabrateria Nova a Casinum,” in G. Ceraudo (ed.), Ager Aquinas. Aerofotografia archeologica lungo la valle dell’antico Liris (Marina di Minturno) 29-36. Ceraudo, G. 2008. “Progetto Ager Aquinas. La carta archeologica di Aquinum e territorio,” in C. Corsi and E. Polito (edd.), Dalle sorgenti alla foce (Rome) 145-55. Cherry, J. F. 2002. “Vox POPULI: Landscape archaeology in Mediterranean Europe,” JRA 15, 561-73. Chiocci, P. F. 2004. “Centuriazione di Aquinum nella media valle del Liri,” in G. Ceraudo (ed.), Ager Aquinas. Aerofotografia archeologica lungo la valle dell’antico Liris (Marina di Minturno) 4347. Chouquer, G. et al. 1987. Structures agraires en Italie centro-méridionale. Cadastres et paysages ruraux (Rome). Coarelli, F. (ed.) 1986. Fregellae, vol. 2: Il santuario di Esculapio (Rome). Coarelli, F. 1988. “Colonizzazione romana e viabilità,” DialArch 6, 35-48. Coarelli, F. 1992. “Colonizzazione e municipalizzazione. Tempi e modi,” DialArch 10, 21-30. Coarelli, F. 1998. “La storia e lo scavo,” in F. Coarelli and P. G. Monti (edd.), Fregellae 1. Le fonti, la storia, il territorio (Rome) 29-69. Cortese, C. 2005. “Le ceramiche comuni: problemi generali e criteri di classificazione,” in D. Gandolfi (ed.), La ceramica e i materiali di età romana. Classi, produzioni, commerci e consumi (Bordighera) 325-38. Crawford, M. H. and L. Keppie 1984. “Excavations at Fregellae, 1978 - 1984. An interim report on the work of the British team,” PBSR 52, 21-35. Crawford, M. H., L. Keppie and M. Vercnocke 1985. “Excavations at Fregellae, 1978-84. An interim report on the work of the British team. Part II,” PBSR 53, 72-96. Crawford, M. H. et al. 1986. “Excavations at Fregellae 1978-1984. Part III. The Territory,” PBSR 54, 40-68. De Haas, T. 2011. Fields, farms and colonists:



intensive field survey and early Roman colonization in the Pontine region, central Italy (Ph.D thesis, Rijksuniversiteit Groningen). De Ligt, L. 2008. “The population of Cisalpine Gaul in the time of Augustus,” in L. de Ligt and S. J. Northwood (edd.), People, Land and Politics. Demographic Developments and the Transformation of Roman Italy 300 BC-AD 14 (Leiden) 139-83. Fentress, E. 2009. “Peopling the countryside: Roman demography in the Albegna Valley and Jerba,” in A. Bowman and A. Wilson (edd.), Quantifying the Roman economy: methods and problems (Oxford) 126-61. Fentress, E. et al. 2004. “Accounting for RAS: fineware and sites in Sicily and Africa,” in S. E. Alcock and J. F. Cherry (edd.), Side-by-side survey. Comparative regional studies in the Mediterranean World (Oxford) 147-62. Fentress, E. and P. Perkins 1989. “Counting African Red Slip Ware,” L’Africa Romana 5, 205-14. Ferraby, R. et al. 2008. “Archaeological survey at Fregellae 2004 - 2005,” in C. Corsi and E. Polito (edd.), Dalle sorgenti alla foce. Il bacino del Liri-Garigliano nell’antichità. Culture, contatti, scambi. Atti del convegno, Frosinone - Formia, 10 - 12 novembre (Rome) 125-31. Finley, M. I. 1977. “The Ancient City: from Fustel de Coulanges to Max Weber and beyond,” Comparative Studies in Society and History 193.3, 305-27. Frederiksen, M. 1970-71. “The contribution of archaeology to the agrarian problem of the Gracchan period,” DialArch 4-5, 330-57. Germoni, P. et al. 2011. “The Isola Sacra: reconstructing the Roman landscape,” in S. Keay and L. Paroli (edd.), Portus and its hinterland (London) 231-60. Giardina, A. and Schiavone 1981. Società romana e produzione schiavistica (Rome/Bari). Hay, S. A. et al. 2010. “Falerii Novi: further survey of the northern extramural area,” PBSR 78, 138. Hay, S. A. et al. 2012. “Interamna Lirenas e il suo territorio: indagini archeologiche non invasive 2010,” in G. Ghini and Z. Mari (edd.), Lazio e Sabina, vol. 8: Ottavo incontro di studi sul Lazio e la Sabina: atti del convegno: Roma 30-


31 marzo, 1 aprile 2011 (Rome) 603-09. Hay, S. A. et al. forthcoming. “Interamna Lirenas e il suo territorio: indagini archeologiche non invasive 2011,” in G. Ghini and Z. Mari (edd.), Lazio e Sabina, vol. 9 (Rome). Hayes, J. W. 1994. “Appendix A. Pottery finds,” in J. W. Hayes and I. P. Martini (edd.), Archaeological survey in the Lower Liri Valley, Central Italy (Oxford) 127-28. Hayes, J. W. 2000. “The current state of Roman ceramic studies in Mediterranean survey, or handling pottery from surveys,” in R. Francovich and H. H. Patterson (edd.), Extracting Meaning from Ploughsoil Assemblages (Oxford) 105-09. Hayes, J. W. and I. P. Martini 1994. Archaeological Survey in the Lower Liri Valley, Central Italy (Oxford). Hayes, J. W. and E. M. Wightman 1984. “Interamna Lirenas: risultati di ricerche in superficie 19791981,” Archeologia Laziale 6, 137-48. Hopkins, K. 1978. Conquerors and slaves (Cambridge). Johnson, P. and M. Millett (edd.) 2012. Archaeological Survey and the City (Oxford). Kajava, M. 1996. “Nuove iscrizioni dal Lazio meridionale,” in H. Solin (ed.), Studi storicoepigrafici sul Lazio antico (Rome) 187-220. Keay, S. et al. 2000. “Falerii Novi: a new survey of the walled area,” PBSR 68, 1-93. Launaro, A. 2009. “La Val Polcevera dalla seconda Età del Ferro alla fine dell’Evo Antico,” Rivista di Studi Liguri 72-73, 5-34. Launaro, A. 2011a. “Investing in the countryside: villas and farms, landowners and tenants (200 BC to 100 AD),” Facta 5, 129-43. Launaro, A. 2011b. Peasants and slaves. The rural population of Roman Italy (200 BC to AD 100) (Cambridge). Lauria, M. 2010. “Il centro fortificato preromano di Colle S. Lucia-Maceralonga nei territori di S. Giorgio a Liri e Castelnuovo Parano (Frosinone),” in G. Ghini (ed.), Lazio e Sabina, vol. 6: Sesto incontro di Studi sul Lazio e la Sabina: atti del convegno: Roma 4-6 marzo 2009 (Rome) 481-85. Lena, G. 1982. “Interamna Lirenas. Note di topografia antica,” Quaderni museo civico

Pontecorvo 2, 57-66. Lo Cascio, E. 1999. “The population of Roman Italy in town and country,” in J. Bintliff and K. Sbonias (edd.), Reconstructing past population trends in Mediterranean Europe (3000 BC 1800 AD) (Oxford) 161-71. MacDonald, A. 1995. “All or nothing at all? Criteria for the analysis of pottery from surface survey,” in N. Christie (ed.), Settlement and economy in Italy, 1500 BC – AD 1500 (Oxford) 25-29. Marsh, G. 1981. “London’s samian supply and its relationship to the development of the Gallic samian industry,” in A. C. Anderson and A. S. Anderson (edd.), Roman Pottery research in Britain and NW Europe (Oxford) 173-238. Martini, I. P. 1994. “The genesis of an environment: elements of geology, geomorphology and land use,” in J. W. Hayes and I. P. Martini (edd.), Archaeological survey in the Lower Liri Valley, Central Italy (Oxford) 5-10. Marzano, A. 2007. Roman villas in central Italy. A social and economic history (Leiden). Mattingly, D. J. 2011. Imperialism, power and identity: experiencing the Roman empire (Oxford). Millett, M. 1985. “Field-survey calibration: a contribution,” in C. Haselgrove, M. Millett and I. Smith (edd.), Archaeology from the ploughsoil (Sheffield) 31-38. Millett, M. 1991. “Pottery: population or supply pattern? The Ager Tarraconensis approach,” in G. Barker and J. Lloyd (edd.), Roman landscapes: archaeological survey in the Mediterranean region (London) 18-26. Millett, M. 2007. “Urban topography and social identity in the Tiber valley,” in R. Roth (ed.), Roman by integration. Dimensions of group identity in material culture and text (Portsmouth, RI) 71-82. Millett, M. 2010. “Town and country in the early Roman West – a perspective,” in C. Corsi and F. Vermeulen (edd.), Changing landscapes: the impact of Roman towns in the western Mediterranean (Bologna) 17-25. Monti, P. G. 1998. “Carta archeologica del territorio,” in F. Coarelli and P. G. Monti (edd.), Fregellae 1. Le fonti, la storia, il territorio (Rome) 81-111.




Nicosia, A. 2008. “Le vie di comunicazione tra la media valle del Liri e la costa tirrenica,” in C. Corsi and E. Polito (edd.), Dalle sorgenti alle foci: il bacino del Liri-Garigliano nell’antichità. Culture, contatti, scambi (Naples) 205-14. Nicosia, A. and G. Ceraudo 2007. Spigolature Aquinati. Studi storico-archeologici su Aquino e il suo territorio (Aquino). Olcese, G. 1993. Le ceramiche comuni di Albintimilium: indagine archeologica e archeometrica sui materiali dell’area del Cardine (Florence). Osborne, R. 2004. “Demography and survey,” in S. E. Alcock and J. F. Cherry (edd.), Side-by-side survey. Comparative regional studies in the Mediterranean World (Oxford) 163-72. Pasquinucci, M. and A. Launaro 2009. “Researching ‘on the margins’: landscape archaeology in the Polcevera Valley,” in I. Holm and K. Stene (edd.), Liminal Landscapes. Beyond the concepts of ‘marginality’ and ‘periphery’ (Oslo) 183-201. Patterson, H., H. Di Giuseppe and R. Witcher 2004. “Three South Etrurian ‘crises’: first results of the Tiber Valley Project,” PBSR 72, 1-36. Patterson, H. and M. Millett 1998. “The Tiber Valley Project,” PBSR 66, 1-20. Patterson, J. R. 2006. Landscapes and cities: rural settlement and civic transformation in early imperial Italy (Oxford). Pelgrom, J. 2008. “Settlement Organization and Land Distribution in Latin Colonies before the Second Punic War,” in L. de Ligt and S. J. Northwood (edd.), People, Land and Politics. Demographic Developments and the Transformation of Roman Italy 300 BC-AD 14 (Leiden) 333-72. Potter, T. W. 1979. The changing landscape of South Etruria (London). Potter, T. W. and S. K. F. Stoddart 2001. “A century of prehistory and landscape studies at the British School at Rome,” PBSR 69, 1-34. Quilici, L. 1974. Collatia (Rome). Quilici, L. 1994. “Centuriazione e paesaggio agrario nell’Italia centrale,” in J. Carlsen, P. Ørsted and J. E. Skydsgaard (edd.), Landuse in the Roman empire (Rome) 127-33. Rathbone, D. W. 1981. “The development of Agriculture in the ‘Ager Cosanus’ during the



Roman Republic: problems of evidence and interpretation,” JRS 71, 10-23. Rathbone, D. W. 2008. “Poor peasants and silent sherds,” in L. de Ligt and S. J. Northwood (edd.), People, Land and Politics. Demographic Developments and the Transformation of Roman Italy 300 BC-AD 14 (Leiden) 305-32. Roth, R. 2007. Styling Romanisation: pottery and society in Central Italy (Cambridge). Salmon, E. T. 1969. Roman colonization under the Republic (London). Santoro Bianchi, S. and B. Fabbri (edd.) 1997. Il contributo delle analisi archeometriche allo studio delle ceramiche grezze e comuni. Il rapporto forma/funzione/impasto (Bologna). Sbonias, K. 1999. “Introduction to issues in demography and survey,” in J. Bintliff and K. Sbonias (edd.), Reconstructing Past Population Trends in Mediterranean Europe (Oxford) 1-20. Scardozzi, G. 2004. “Interamna Lirenas,” in G. Ceraudo (ed.), Ager Aquinas. Aerofotografia archeologica lungo la valle dell’antico Liris (Marina di Minturno) 91-93. Scheidel, W. 2008. “Roman population size: the logic of the debate,” in L. de Ligt and S. J. Northwood (edd.), People, Land and Politics. Demographic Developments and the Transformation of Roman Italy 300 BC-AD 14 (Leiden) 17-70. Stek, T. D. 2009. Cult places and cultural change in Republican Italy. A contextual approach to religious aspects of rural society after the Roman conquest (Amsterdam). Terrenato, N. 2001. “The Auditorium site in Rome and the origins of the villa,” JRA 14, 5-32. Terrenato, N. 2004. “Sample size matters! The paradox of global trends and local surveys,” in S. E. Alcock and J. F. Cherry (edd.), Side-by-side survey. Comparative regional studies in the Mediterranean World (Oxford) 36-48. Van Dommelen, P. and N. Terrenato (edd.) 2007. Articulating local cultures. Power and identity under the expanding Roman Republic (Portsmouth, RI). Ward-Perkins, J. 1961. Veii: the historical topography of the ancient city (London). Wightman, E. M. 1994a. “Communications,” in J. W. Hayes and I. P. Martini (edd.), Archaeological


survey in the Lower Liri Valley, Central Italy (Oxford) 30-33. Wightman, E. M. 1994b. “The Iron Age,” in J. W. Hayes and I. P. Martini (edd.), Archaeological survey in the Lower Liri Valley, Central Italy (Oxford) 13-17. Wightman, E. M. 1994c. “The Romanization of the valley,” in J. W. Hayes and I. P. Martini (edd.), Archaeological survey in the Lower Liri Valley, Central Italy (Oxford) 26-29. Wightman, E. M. and J. W. Hayes 1994. “Settlement patterns and society,” in J. W. Hayes and I. P. Martini (edd.), Archaeological survey in the Lower Liri Valley, Central Italy (Oxford) 34-40. Wightman, E. M., P. Hemphill and J. W. Hayes 1994. “Archaeology,” in J. W. Hayes and I. P. Martini (edd.), Archaeological survey in the Lower Liri Valley, Central Italy (Oxford) 2-3. Williams, D. F. 1994. “Petrological examination of

pottery and tile from the Lower Liri Valley,” in J. W. Hayes and I. P. Martini (edd.), Archaeological survey in the Lower Liri Valley, Central Italy (Oxford) 159-63. Wilson, A. 2008. “Site recovery and the ancient population of the Biferno Valley,” in G. Lock and A. Faustoferri (edd.), Archaeology and landscape in central Italy (Oxford) 233-53. Witcher, R. 2005. “The extended metropolis: Urbs, suburbium and population,” JRA 18, 120-38. Witcher, R. 2006. “Broken pots and meaningless dots? Surveying the rural landscapes of Roman Italy,” PBSR 74, 39-72. Yntema, D. G. 2008. “Polybius and the field survey evidence from Apulia,” in L. de Ligt and S. J. Northwood (edd.), People, Land and Politics. Demographic Developments and the Transformation of Roman Italy 300 BC-AD 14 (Leiden) 373-85.




PART III THE RELIGIOUS DIMENSION OF ROMAN COLONIZATION


Tutelary deities in Roman citizen colonies Marion Bolder-Boos

Although tutelary deities of individual urban settlements were common in the ancient orient,1 there are only few references for the Roman world. In some cases they can be detected either through the literary tradition, public representation or the appearance of a deity’s name in the official nomenclature of a town.2 Literary sources explicitly naming a particular deity as protector of a settlement are rare. There is one passage in Appian’s Bella civilia reporting that the inhabitants of Etruscan Perusia used to worship Juno but, after the town was destroyed by Octavian, the survivors chose Vulcan as their patron deity.3 In other instances, the tutelary god or goddess is referred to as ruler or custos of a town.4 The idea of communities being in tutela of one or several gods perhaps is most plainly evident in connection with the ritual of evocatio.5 Talking about ancient Roman beliefs, Pliny reports that the Romans used to evoke a settlement’s protective divinity prior to military actions and promised this deity greater veneration than those people, whose patron the god or goddess was, could offer. He also states that the fear of an enemy evocatio caused the Romans to keep their tutelary god secret.6 It is clear that Pliny is not referring to a contemporary ritual but to an old custom which had fallen into abeyance. The most prominent occasion on which this ritual was allegedly performed is the evocatio of Juno Regina of Veii, which, according to Livy, was employed by M. Furius Camillus in order to finally conquer the enemy city in 396 B.C.7 Another report of an evocatio comes from Macrobius, who is commenting on a passage in

1

2

3 4

5

Cf. Ishtar of Uruk, Marduk of Babylon, Sîn of Ur or Melqart of Tyre. Prominent examples are, for instance, Juno Curritis of Falerii, Fortuna of Antium or Fortuna Primigenia of Praeneste. In Herculaneum, the importance of Hercules is evident not only in the name of the town but also in the frequent representation of this god in public art. App., B Civ. 5, 49. Cf. Juno Regina of Veii. See also Hor., Carm. 1.28.29 on Neptune and Jupiter as protectors of Tarentum, Hor., Carm. 1.35.1 on Fortuna of Antium and Serv., Georg. 1, 498 on Minerva, patroness of Athens, and Juno, patroness of Carthage. There are only a few references to the ritual of evocatio in Roman literature. An inscription discovered at Isaura Vetus in Cilicia in 1970 (AE 1977, 816 = CIL I² 2954) reporting an evocatio in the years around 75 B.C. indicates that the ritual was practised at least until the 1st c. B.C. Modern scholarship therefore tends to accept the historicity of the ritual, cf. Alvar 1984; Rüpke 1990, 162-64; Gustafsson 2000,

6

7

46-62; Rüpke 2006, 134-36; Ferri 2010a; id. 2010b, 3349. On evocationes in general see also Wohleb 1927; RAC 6 (1966) 1160-65, s.v. “Evocatio” (F. Pfister); Bruun 1972. Plin., HN 28.18: “Verrius Flaccus auctores ponit, quibus credat, in oppugnationibus ante omnia solitum a Romanis sacerdotibus evocari deum, cuius in tutela id oppidum esset, promittique illi eundem aut ampliorem apud Romanos cultum. Et durat in pontificum disciplina id sacrum, constatque ideo occultatum, in cuius dei tutela Roma esset, ne qui hostium simili modo agerent”. Cf. Gustafsson 2000, 42-43; Ferri 2010b, 44. On the secret tutelary deity of Rome see Brelich 1949; Ferri 2010a; id. 2010b, 189-97. Liv. 5.21.3: “Iuno Regina, quae nunc Veios colis, precor, ut nos victores in nostrum tuamque mox futuram urbe sequare, ubi te dignum amplitudine tua templum accipiat”. On Camillus’ evocatio of Juno and the erection of a new temple to the goddess in Rome see also Liv. 5.22.3-7; Val. Max. 1.8.3; Plut., Cam. 6.1-2; Dion. Hal., Ant. Rom. 13.3.3.




the Aeneid referring to the gods of Carthage. He equally speaks of tutelary deities.8 While characterizing the ritual as an old Roman custom as well, he also states that it was a well-known fact that all cities were under the guardianship of a god, and that the Romans’ guardian deity had a secret name to prevent him or her from being evoked by enemies.9 Thus, the authors are describing a ritual they did not know very well, since this custom, no longer in use in imperial times, was a memory of a distant past.10 An earlier source, Vitruvius, shows no concern about keeping the name of the patron deity of a town or community secret. Speaking about the most suitable places for the erection of temples, he recommends placing the sanctuaries for the protective gods of a settlement on the highest ground, from where these gods could overlook most of their territory.11 This passage indicates that tutelary gods were not just a feature of enemy cities conquered by Rome, but existed for Roman communities as well. Therefore, while some aspects of the evocatio ritual and Rome’s secret guardian deity remain obscure,12 tutelary deities seem to have been an integral part of the local panthea of Roman towns. In the context of Roman colonization, this raises the question of which patron deities could be found in Roman citizen colonies, whose settlers, in contrast to those of Latin colonies, retained their Roman citizenship.

Roman citizen colonies and their gods Despite the story about the secret guardian deity mentioned above, the tradition of the city of Rome makes it quite clear that Jupiter used to be its most important god, and the first sanctuaries which were built soon after the foundation of the city were dedicated to him.13 As Rome grew and its influence in Latium increased, a more representative sanctuary was sought. Hence, during his struggle against the Sabini, Tarquinius Priscus vowed to build a great temple to Jupiter Optimus Maximus,14 which was duly erected on the Capitoline Hill and finally consecrated by the founders of the Roman Republic in 509 B.C.15 In the course of the Roman Republic, Jupiter and the Capitoline Triad became the protectors of both the city of Rome itself and the Roman State, and temples to Jupiter or the Capitoline Triad were built in several Roman citizen colonies.16 Nevertheless,

8

9



Macrob., Sat. 3.9.7-9: “Si deus, si dea est, cui populus civitasque Carthaginiensis est in tutela, teque maxime, ille qui urbis huius populique tutelam recepisti, precor veneroque veniamque a vobis petut vos populum civitatemque Carthaginiensem deseratis, loca templa sacra urbemque eorum relinquatis, absque his abeatis eique populo civitati metum formidinem oblivionem iniciatis, prodiitque Romam ad me meosque veniatis, nostraque vobis loca temple sacra urbs acceptior probatiorque sit, mihique populoque Romano militibusque meis praepositi sitis ut sciamus intellegamusque. Si ita feceritis, voveo vobis templa ludosque facturum”. It has been disputed whether the evocatio of the gods of Carthage, which is also mentioned by Serv., Aen. 12.841, was a historical event or a legend. Wissowa 1912, 374 and Latte 1960, 125 especially have interpreted it as a later Roman invention. However, more recently scholars have tended to accept the historicity of this evocatio, see Le Gall 1976; Berti 1990; Gustafsson 2000, 59-60; Ferri 2010b, 95-112. Macrob., Sat. 3.9.1: “Constat enim omnes urbes in alicuius dei esse tutela, moremque Romanorum arcanum et

10 11

12

13 14 15 16

multis ignotum fuisse ut, cum obsiderent urbem hostium eamque iam capi posse confiderent, certo carmine evocarent tutelares deos; quo daut aliter urbem capi posse non crederent, aut etiam si posset, nefas aestimarent deos habere captivos. Nam propterea ipsi Romani et deum in cuius tutela urbs Roma est et ipsius urbis Latinum nomen ignotum esse voluerunt”. Cf. Gustafsson 2000, 45. Cf. Rüpke 2006, 134-36. Vitr., De arch. 1.7.1: “Aedibus vero sacris, quorum deorum maxime in tutela civitas videtur esse, et Iovi et Iunoni et Minervae, in excelsissimo loco, unde moeniam maxima pars conspiciatur, areae distribuantur.” See also Boos 2011b. Gustafsson 2000, 140 suggests that the idea of Rome’s secret tutelary deity is probably a late invention. Liv. 1.10.5–7; 1.11.6. Liv. 1.38.7 Liv. 2.8.6-8. The role of Capitolia as tools of religious Romanization has rightly been questioned in recent years, cf. Bispham 2006, esp. 96-105 on the so-called Capitolium of Cosa;


neither Jupiter nor the Capitoline Triad automatically became the tutelary deities of every colony (cf. the contributions of Carini and Torelli in this volume). Ostia In Ostia, Vulcan was considered to be the patron deity. His temple has not yet been securely identified,17 but the extraordinary position of Vulcan is revealed by the Fasti Ostienses and other written sources, according to which the pontifex Volkani held the most important priestly office in Ostia and is recorded to have been superintendent of all the sanctuaries in the city.18 This includes the Capitolium situated in the center of the town on the northwestern side of the forum. Furthermore, the priest of Vulcan had several magistrates as his subordinates.19 As Ostia was founded ex novo, it is probable that the Roman settlers brought the god with them from Rome, although it has also been argued that there was an archaic sanctuary of Vulcan prior to the foundation of the colony.20 There is still disagreement about why Vulcan was chosen: as fire god protecting the city and its granaries;21 as “guarantor of the plenty, safety and continued navigability of the Tiber”;22 as a god who had a sanctuary near the comitium in Rome, where the founding of colonies was decided upon in Republican times and from where the Ostian settlers brought him to their new home;23 or perhaps as god of blacksmithing, making not only offensive but also defensive weapons?24 This last skill would have endowed him special protective abilities that would have been most welcome in a colony founded in order to protect the mouth of the Tiber. These protective aspects would have made Vulcan an ideal patron deity for all colonies established on the borders of enemy territory. However, it is only in Ostia and – from the later 1st c. B.C. – Perusia that he held this special position. In the colonies for which a patron god can be identified at all, quite different deities can be found.

17

see also Stek 2009b. However, while there is no evidence for Capitolia in Latin colonies, they do seem to have been an important feature of Roman citizen colonies, cf. Boos 2011a, 221-23. On the differentiation between the two types of colonies in general see DNP 3 (1997) 76-85, s.v. “Coloniae” (H. Galsterer) with further references. Many attempts have been made to locate the temple: Vaglieri 1914, 13 presumed it to have been sited outside the colony alongside the road leading to Rome. Vidman 1957, 86 suggests that the temple should be sought near the Porta Marina, where a large concentration of the Fasti Ostienses has been found (see also Zevi 1994, 408). Meiggs 1960, 342 deems it possible that Vulcan was the principal god of an earlier settlement by the salt pans which retained its importance when that site was absorbed by the Roman colony. Steuernagel 2004, 163, following Vaglieri, suggests that the temple was an extra-urban sanctuary near the mouth of the Tiber. Rieger 2004, 220-21 identifies the temple on the western side of the Republican Capitolium with the temple of Vulcan, proposing that, when the temple was built over by the so-called curia, the cult was transferred to a shrine in

18 19

20

21

22 23 24

this new building. However, a shrine has not yet been identified. Pensabene et al. 2007, 118-23 propose an identification of the temple of Vulcan with the ‘Tempietto Repubblicano’ situated at the intersection of the Decumanus Maximus with the Via dei Molini. The most recent attempt to find the temple of Vulcan has been made by Boin 2010, 260-1, who claims that the temple of Hercules in the ‘area sacra dei templi repubblicani’ outside the Porta Occidentale was not dedicated to Hercules but perhaps to Vulcan. Contra Boos 2011a, 69; ead. 2012. Therefore, the location of the temple of Vulcan remains disputed. AE 1968, 81; AE 1988, 216. CIL XIV 4553; CIL XIV 4625; AE 1986, 111-12; AE 1989, 125; cf. Meiggs 1960, 177-78, 337-38. Carcopino 1919, 42-44; Rose 1933, 51-52; Meiggs 1960, 342; Pellegrino 1986, 296-97. Wissowa 1912, 229-32; Roscher, Lex. 6 (1924-37) 35669, s.v. “Volcanus” (G. Wissowa); Latte 1960, 129-31; contra Rose 1960, 161-72. Bispham 2000, 163. Simon 1990, 250. Boos 2011a, 43.




FIG. 1. Luni, plan of the city (after Lackner 2008, 356).

Luni Luni, 17 km northwest of Cararra in Liguria, has a Capitolium at the forum in the center of town, but the tutelary deity was the moon goddess Luna who gave the colony its name.25 The question of why the moon goddess played such a significant role in this colony has not yet been answered satisfactorily. In the late 3rd or early 2nd c. B.C., the Romans had established a military camp there, which they named Portus Lunae,26 and Strabo records how, in a similar vein, the Greeks referred to this site as the harbour of Selene.27 Nothing of this camp has survived, as it was replaced by the colony which was founded roughly one generation later in 177 B.C.28

25



Only a few colonies founded in the 3rd and 2nd c. B.C. bore the name of a deity, but it is not entirely without a parallel. In 183 B.C. the citizen colony of Saturnia was founded in Etruria, and more than 100 years earlier, in 291 B.C., a Latin colony named Venusia was established in Lucania. In the case of Saturnia the relationship with Saturn is rather a puzzle – the nomenclature might have been linked to the legend of Italy as the refuge of Saturn after his defeat at the hands of Jupiter (see. Dion. Hal., Ant. Rom. 1.34). The Latin colony Venusia was said to go back to the old

26

27 28

town Aphrodisias, founded by the Greek hero Diomedes as a peace-offering to Aphrodite (Serv., Aen. 11.246). On the possible connection between the Latin colony and an important Venus cult in this region see Koch and Seel 1960, 63-64; Torelli 1995, 150-54. Ennius Ann. 16; cf. also Liv. 34.8.4, who reports that in 195 B.C. Cato the Elder set out for “Lunae portum” with twentyfive warships to join his army which had mustered there for the campaigns in Spain. Strab. 5.2.5. Liv. 41.13.4.


FIG. 2. Luni, area of the temple of Luna (so-called “Grande Tempio”).

Two Republican temples have been identified at Luni: the Capitolium situated at the northeastern end of the forum and what is known as the Grande Tempio in the northern part of the city (fig. 1).29 The latter is a tetrastyle temple on a high podium which measures ca 16 x 20.5 m (fig. 2). It is built of local stone and, like the Capitolium, was erected soon after the foundation of the colony. Until the temple at the forum was discovered, the Grande Tempio was thought to be the Capitolium because of its tripartite cella. Although no inscriptions which might reveal the deity worshipped there have been found, a group of terracotta statues from the temple pediment suggests that it was dedicated to Luna: in the center of the group is an enthroned female figure in a long robe which leaves one breast uncovered. Unfortunately, the head is missing. To her right stands a naked young man with a kithara, to her left another man in a cloak, a cornucopia in his left arm. Two female figures in long robes flank the group. The kithara player can easily be identified as Apollo, while the man with the cornucopia is perhaps Dionysos/Liber.30 The goddess on the throne was originally interpreted as Minerva,31 but a bare-breasted Minerva seems rather unlikely. As Apollo is at her side, the most plausible suggestion is that the goddess is somehow connected with him (cf. Carini in this volume). Anna Maria Durante and Lucia Gervasini have therefore postulated that the goddess is either Diana or Luna,32 but a Diana in a long robe would be an unusual depiction of Apollo’s sister. However, Luna is often represented in this manner, sometimes even

29 30

Bonghi Jovino 1973, 653-91; ead. 1977, 413-52. Durante and Gervasini 2000, 70.

31 32

Milani 1884, 89-112. Durante and Gervasini 2000, 70; see also Coarelli 1985-87.




FIG. 3. Terracina, plan of the city (after Lackner 2008, 381).

with one breast bare.33 A connection between Luna and Dionysos is indicated by Cicero, according to whom Dionysos was sometimes regarded as the son of Jupiter and the moon goddess.34 Furthermore, several fragments of gold-plated bronze torches have been found in one of the favissae of the temple,35 which were most likely part of a votive offering or even belonged to the cult statue of the temple. Among the deities depicted with torches are Apollo, Vulcan, Diana and Luna. Given these facts, Luna seems to be the most plausible deity here, which would make the Grande Tempio the sanctuary of the tutelary deity of Luni. The t