Page 1


Plato’s Laws R.F. STALLEY

Basil Blackwell

© R .F . S ta lk y 1983

F irst p u b lish ed 1983 Basil Blackwell P u b lish er L im ited 108 Cowley R oad, O xford 0 X 4 1JF, E ngland All rights reserved. N o p a rt o f this publication may be repro d u ced , stored in a retrieval system , or tra n sm itte d , in any fo rm o r by any m eans, electronic, m echanical, photocopying, recording or otherw ise, w ith o u t th e p rio r perm ission of th e publisher.

B r itish L ibrary C ataloguin g in P u b lica tio n D ata Stalley, R .F . An in tro d u ctio n to P lato ’s Laws 1. Plato. L aw s— C riticism and in terp re tatio n 2. L aw 3. Ju risp ru d en ce 1. t i t l e 340’. 109 IS B N 0 -6 3 1 -1 3 3 9 9 -2

T y p eset b y K errypress L td , L u to n . P rin te d in G reat B ritain by T h e P itm an Press, Bath.


Preface 1

Reading the Laws 1

2 3 4 5 6




1 1

2 4 7 8 10

Plato’s Political Philosophy


1 The philosopher king 2 The impracticality o f the Republic 3 The Statesm an 4 The letters 5 P lato’s politics

14 16 17 19 21

The Nature o f Law


1 Law in the Greek city 2 Law and purpose 3 The role o f reason 4 The role o f the legislator 5 Plato and natural law

23 24 28 31 33

The Aims o f Legislation


1 2 3 4


The importance o f the L a w s Date and style The structure o f the dialogue The character o f the constitution From the Republic to the Laws English versions


Critique o f the Dorian legislation Virtue and peace as objects o f legislation P lato’s legal moralism Law as education

35 37 40 42



1 M oral psychology




M oral responsibility Self-control and the possibility o f akrasia The harmony o f passion and judgement Sophrosune: the mainspring o f the Laws? The unity o f virtue

2 3 4 5 6


Pleasure and the G ood L ife Book I: pleasure and self-restraint Book II: pleasure and happiness Pleasure and the choice o f lives Book V: divine goods and human pleasures Pleasure and choice

1 2 3 4 5


Political Analysis 1 2 3 4 5 6


T h e Sovereignty o f Law 1 2 3 4


In what sense can law be sovereign? How can the sovereignty o f law be ensured? Is the sovereignty o f law realistic? S ophrosune and the sovereignty o f law

Towards the Ideal State 1 2 3 4 5


The origin o f states Peloponnesian history Seven claims to rule S parta and the mixed constitution Persia and A thens The politics o f due measure

Ideal states and utopias The Cretan colony The conditions fo r legislation The second-best state P la to ’s purposes

Society 1 2 3 4 5

The problem P la to ’s remedies The position o f women Slaves and aliens Class distinctions and the division o f labour

48 50 52 54 56

59 60 62 64 66 67

71 71 72 73 74 76 77

80 80 81 83 84

87 87 89 90 92 94

97 97 99 104 106 108






The System of Government



1 Magisterial power 2 Checks and balances 3 A mixed constitution 4 Democracy

113 115 116 120

Education and the Arts


1 The aims o f education 2 Music 3 Education and play 4 Organization and curriculum 5 Higher things

123 125 130 132 133



1 The penalties 2 The aims o f punishment 3 Punishment as cure 4 The theory in practice

137 139 143 146



1 2 3 4 5



The problem The solution The sources o f injustice Responsibility in practice Responsibility and the treatment o f criminals

152 154 157 160 164



1 The dangers o f atheism 2 The priority of the soul 3 From soul to god 4 The significance o f the proof 5 The gods’ concern fo r man 6 The law against impiety

167 169 172 173 175 111

The Closed Society o f the Laws


1 Plato and Popper 2 P lato’s principles 3 The afterlife o f the'Laws

179 182 184

v iii


A P P E N D IX : T he C onstitution Further Reading Select Bibliography Index

186 192 198 204



T o most students of philosophy Plato ’s Laws m ust appear som ething o f a m ystery. I t is obviously an im portant work b u t, by com parison with dialogues like the Republic or the Phaedo, it has received little attention. T h e reasons for this neglect are not far to seek. T h e style is difficult and sometimes obscure b u t even greater problem s stem from the structure of th e dialogue. T he philosophical content is interw oven with a mass of legal m aterial that is now largely of historical interest. So, although an adequate appreciation of P lato’s th ought m ust take full account of the Laws, the average reader cannot, I think, be expected to make very m uch of it w ithout some kind of guide. This introduction is intended to fulfil that role. In w riting it I have em phasized those aspects of th e Laws th at have m ost significance for the problem s of philosophy as we now understand th e m . I have treated Plato as a thinker with whom one can argue, not as a figure to be regarded with aw estruck adm iration nor m erely as a source for the understanding o f G reek culture and society. I have tried to initiate debates and to stim ulate the reader’s own philosophizing rath er than to give conclusive answers. Given these aim s, it has not been possible to follow closely the order of discussion as it develops through the twelve books of the Laws. Plato’s m anner is to retu rn repeatedly to the same subjects. I have therefore discussed th e Laws topic by topic rather than taking it book by book, though there is a rough correspondence betw een the order I have adopted and the order in w hich different them es are brought to prom inence in th e text. At the head of each chapter there is a list of relevant passages from the Laws and from other dialogues. These lists are not exhaustive, b u t they indicate the passages w hich have most influenced my interpretations and which I would recom m end as starting points for the reader’s own reflections. T here is very little m odern literature about the Laws and m ost of what there is deals w ith historical and philological m atters rather than with the philosophy of the dialogue. I have listed those m odern



discussions w hich I feel able to recom m end in the suggestions for F u r th e r R eading. I have also included some m odern discussions of related topics and som e o f th e m any passages from A ristotle that bear o n th e Laws. T h e latter may prove m ore helpful than m ost of the m o d e rn works. L ike m ost academ ic authors I owe m uch to the colleagues and stu d e n ts w ith w hom , over the years, I have discussed problem s of a n c ie n t philosophy. In p articular I wish to record the d eb t th at I and ev ery o th e r E nglish-speaking stu d e n t o f the Laws m ust owe to P ro fe sso r T .J. Saunders. H is translation has m ade th e Laws accessible to th e ordin ary reader in a quite rem arkable way and he has c o n trib u te d enorm ously to o u r u n d erstan d ing of the dialogue by his scholarly articles and bibliography. I am grateful too for the friendly in te re st he has show n tow ards my own work.


Reading the Laws 1

The importance o f the Laws

The Laws is P lato’s longest as well as his last work. O f its twelve books the first three suggest the basic principles w hich should underlie any satisfactory system of law. A ttention then shifts to a schem e for the foundation o f a new city som ew here in C rete. Books IV and V deal with various prelim inaries to legislation, and the rem aining seven books describe in considerable detail a constitution and system of law for this new city. I t is obvious th at a work o f this kind m ust deal, at least by im plication, w ith most of the fundam ental questions o f political and legal philosophy. B ut since Plato takes th e province of law to include every aspect of public and private life, the scope o f the dialogue is wider than one m ight at first suppose. F or exam ple, it contains a full account of the aims and m ethods o f education, a systematic treatm en t of the theory of punishm ent and responsibility and an elaborate proof of the existence of a god or gods. T h e discussions of the moral basis of the state include P lato’s final thoughts on topics, such as the natu re of virtue and the role of pleasure, th at had concerned him th ro u g h o u t his life. Since Plato is, by any reckoning, one of the greatest of political philosophers, we m ight expect the L a m to occupy a prom inent place in all histories o f political th o u g h t. B ut in practice it has often been ignored. In the eyes o f m ost students and teachers o f political philosophy the only real source for Plato’s doctrine is the Republic. This tendency to forget about the Laws can infect even the finest scholars. Sir E rnest Barker, for exam ple, gives us one of the fullest and most sym pathetic treatm ents o f th e dialogue, yet he is apt to leave it out of account when he generalizes about Plato or about the history of Greek political thought (1960, pp. 14, 207-8, 226-7; cf. 42-6). T h ere are two m ains reasons for the neglect of the Laws. In the first place it is m uch less attractive than the Republic as a work o f literature.



I n th e second place there is a long tradition o f scholarship that id e n tifie s the authentic Plato w ith th e doctrines of the Republic and of a fe w closely related dialogues (the dialogues w hich are now ascribed to P la to ’s m iddle period and w hich give prom inence to the theory of tra n s c e n d e n t Form s). T h e Laws adopts a quite different approach to th e pro b lem s of politics from th at of the Republic, it says virtually n o th in g ab o u t the F orm s, and its doctrine appears inconsistent with th e Republic o n m any points o f detail. It is not surprising therefore that th o s e w ho saw th e au th o r o f th e Republic as the only true Plato tended to play dow n the Laws. So far as logic, m etaphysics and epistem ology are concerned, the tra d itio n a l R epublic-centred view o f Plato is now extinct, at least a m o n g E nglish-speaking scholars. T h o se trained in the analytic tra d itio n o f philosophy have found th a t they can learn as m uch, if not m o re , fro m late dialogues such as th e Sophist as they can from those of th e m iddle period. As yet there has been no corresponding change with r e g a r d to political philosophy. A tten tio n i | still focussed on the R epublic, w ith th e Law s treated as a kind of appendix. In so far as the p re s e n t volum e has any single aim it is to argue that this approach is m ista k e n and th a t a re-evaluation o f th e Laws is overdue. It is as one­ sid e d to focus exclusively on the Republic w hen discussing P lato’s p o litic a l philosophy as it is w hen we consider his m etaphysics or epistem ology. 2

D ate and style

T h e accep ted view of th e Laws as P lato ’s last work is based on external a s w ell as in tern al evidence. D iogenes L aertiu s, the m ost im p o rtan t of th e ex tern al sources, gives the following account of the Laws and o f its s h o rt com panion-w ork the Epinomis: Som e say th at Philip of O pus transcribed the Laws w hich were in th e wax. T hey also say th at the Epinomis is his. ( I I I .37) T h e n a tu ra l reading of th is is th at w hen Plato died in 347 B C he left the L a w s as a rough d raft, p resum ably w ritten on wax tablets. H is pupil, P h ilip , p rep ared this d ra ft for publication. T h e Epinomis, on the other h a n d , is said to be P hilip ’s own work. ^ I t is n o t clear from this account w hether P h ilip ’s contribution to the L a w s consisted sim ply in copying o u t w hat Plato had w ritten or w h e th e r he edited it at all extensively. T h e form er is likely to be nearer th e tru th . T h e text contains errors and discrepancies th at could easily



have been rem oved by an editor had he wished to do so. T heir presence suggests that Philip generally reproduced P lato’s w ords as he found them. So far as the Epinomis is concerned, scholars are divided quite evenly betw een those who follow Diogenes L aertius in ascribing it to Philip and those who regard it as a genuine work o f Plato. F ortunately we can, for present purposes, ignore this controversy. T he Epinomis is not an integral part of the Laws and does not cast m uch fresh light on any of the philosophical questions I shall consider. I shall m ention it only in passing. W ithin the text o f the Laws, 638b may possibly refer to the defeat of the Locrians by the Syracusans in 356 B C . If this is right, Book I, at least, m ust have been w ritten during the last eight years or so of P lato’s life. M ore subjectively, m ost readers agree that the m ain bulk of the work m ust have been composed after the period of P lato ’s involvem ent in Syracusan politics (i.e. after about 360 B C ). T h e frequent harping on the virtues o f old age also suggests that this is an old m an’s work. T h e m ost striking confirm ation o f the lateness o f the Laws comes from the style of w riting and m anner of com position. On a superficial reading, at least, the work often appears ram bling and ill-structured. This is particularly apparent in th e first two books w ith th eir long digression on the subject of drinking parties. T he style is characterized by long ram bling sentences in which the author him self sometimes seems to lose track of th e gram m atical structure. T here are inconsistencies of detail (e.g. about the age o f the th ird chorus (664d, 812b-c) and the age at which a m an should m arry (721b, 772d-e, 785b)). T he characterization is generally weak, at least when com pared with dialogues o f the early and middle periods, and in Book V the dialogue form is dropped altogether. Book X I and the early part of Book X II read like a collection of disconnected fragm ents. Some of these anomalies are m ost naturally explained on the assum ption that Plato died before he could give the Laws its final polish. O thers may result from a decline in his literary powers. W hichever way one takes it, they support the view that the Laws is the product o f P lato’s old age. Scholars hostile to the Laws have sometimes em phasized th at it is an old m an’s work, w ith the im plication that it is likely to show a decline in philosophical power. T here is, I think, a straightforw ard answer to this. It would indeed be rem arkable, if not inconceivable, th at a man of, say, seventy-nine should suddenly develop new and original ideas of lasting significance. But no one would suggest that this is what happened in the case of the Laws. Plato m ust have been working on the Laws for several years before his death. M oreover he had been deeply



co n cern ed w ith political and ethical problem s th ro u g h o u t his life. In settin g dow n his final th o u g h ts on these questions, he obviously drew o n th em es and argum ents developed th ro u gh m any years of reflection. So, even if there are some elem ents in the Laws th at are feeble or even em b arrassin g , we may still expect to find w ithin it m aterial w orthy o f a really g reat philosopher. T h e only way to approach th e dialogue is to re a d it w ith an open m ind, being prepared to adm ire its strengths as well as to criticize its weaknesses.


The structure o f the dialogue

M o st readers o f th e Laws at tim es find it difficult to follow the thread of th e arg u m en t. T his is p artly because the work is long and lacks aids, su ch as ch ap ter headings, th at the m odern reader takes for granted, bu t also because the underlying stru ctu re is often obscured by surface featu res. T h ere are lengthy digressions, and in some cases the same g ro u n d seems to be w orked over again and again. N evertheless, on a careful reading, the stru c tu re does becom e apparent. In m any ways the L aw s is a m ore single-m inded work th an , say, the Republic. C ertainly it n ev er strays very far from its m ain topics. I f it seems m ore diffuse than earlier dialogues th a t is largely because th e transitions are less clearly m arked and less skilfully handled. T h e opening pages of th e Laws leave the reader in no doubt about th e su b ject or th e basis on w hich it is to be treated. T h e participants in th e dialogue are th ree old m en, an unnam ed A thenian who takes the lead, M egillus th e S partan and Cleinias the C retan. T hey are walking th ro u g h th e C retan countryside to th e cave w here Zeus is supposed to have given instructions to M inos, th e legendary lawgiver of Crete. S p arta an d C rete, b o th o f them D orian cities, were m uch adm ired fo r th e stability o f th e ir laws and th e ir strict m ilitary discipline, b u t th ey were distinctly backw ard in philosophy and art. A thens, the home o f dem ocracy, was m uch less stable b u t was know n for the intelligence a n d cu ltu re o f its citizens. In keeping w ith this picture, the two D o rian s b oth attrib u te the legal system s of their own states to divine a u th o rsh ip . T h e A thenian does n o t deny this — he certainly regards legislation as a m ore th a n hu m an m atter — b u t w ithin a few pages he has attacked th e basic assum ptions underlying the C retan and Spartan system s. C rete and S parta were fam ous for th eir em phasis on m ilitary tra in in g and for th e practice w hereby male citizens took meals in co m m u n al dining room s rather th a n in their own households. Both p ractices are d irected tow ards success in w ar, b u t w ar is waged for the

â– f



sake of peace, and the A thenian shows that the D orian states have no institutions to secure the inner concord on which true peace depends. A lthough it is difficult for us to make m uch of the contrast betw een the Athenian and D orian ways o f tho u g h t it m ust have seemed highly significant to P lato’s contem poraries. Plato leaves us in no do u b t as to his respect for the D orian way of life, which was m uch adm ired by conservative elem ents th ro u g h o u t G reece, b u t at the same tim e he makes it clear that the D orian trad itio n is fundam entally defective and needs to be reviewed in the light o f A thenian wisdom. He thus distances him self from the orthodox conservatism o f his day. Once the defects of the D orian institutions are established, the Athenian moves on to describe institutions which m ight rem edy the situation. T he rest o f Book I and th e whole o f Book II are therefore occupied by a discussion of drinking parties, w hich are supposed to achieve beneficial effects by subm itting the young to tem ptations under controlled conditions, and o f the educational value of m usic and dance. M any readers have found this section tedious; Plato, him self apologizes for it (642a, 645c, 890e). From the philosophical point of view, it is im portant to attend to the underlying motives rather than to the proposals themselves. T h e aim o f the law is virtue. A prerequisite of all virtue is sophrosune, self-control, self-discipline, or tem perance. T h e achievem ent of sophrosune depends on being brought up to experience pleasures and pains in th e right kind of way. T he proposals about drinking, m usic and dance are designed to ensure this. T hus Books I and I I set out what is, in fact, the m ain direction o f the whole work. We are concerned w ith a system of laws designed to produce the virtue of self-control; lawgiving is largely a m atter of education. T h e beginning of Book I I I looks like a com pletely new start. W e are given a survey of the history o f hum an society from its first beginnings after the flood. T his culm inates in a description o f the D orian League which looked like a marvellous institution b u t in fact collapsed (682e689e). T he relevance of this whole section becomes clear only w hen we learn the reason for this collapse: it was due to a lack o f the kind of selfcontrol whose im portance has been established in the first two books. Sparta alone survived the collapse because it alone had the good fortune to evolve a constitution w hich com bined different elem ents, each of which acted as a restraint on the others (691b-693c). H ere we are introduced to the idea o f a mixed constitution. T he two fundam ental form s of constitution, m onarchy and dem ocracy, are exem plified by Persia and A thens. Each of these states has declined because of an unrestrained excess in the ruling principle. T he decline o f Persia is due to a lack of self-control on the p a rt of its m onarchs, that



o f A thens is due to th e u n re stra in e d excesses of the masses. A satisfacto ry constitution m ust em body both m onarchic and d e m o c ra tic elem ents so th a t th ere is liberty com bined w ith restraint (6 9 3 d -7 0 2 b ). , T h e historical and political discussions o f Book I I I pick up the fu n d a m e n ta l them es o f Books I and II. T h e aim of legislation is virtue, w h ic h can be achieved only by self-control. Legislation m ust therefore p ro d u c e a society w hich eftsures this self-control on the part of its citiz e n s and w hich exhibits self-control in its forms of governm ent. M o s t existing states have failed to do this and the only way in w hich it can b e achieved is by setting u p a m ixed constitution. T h u s, in spite of th e ir ap p aren t irrelevance and m any digressions, Books I —111 provide a prolegom enon to legislation, setting out the basic principles to be fo llo w ed . W e should aim to p roduce a society characterized by selfc o n tro l and m ay expect to do this by ado p tin g a form of constitution in w h ic h different elem ents are com bined and act as restraints on one a n o th e r. A t th e end o f Book I I I th e C retan reveals th a t he is a m em ber o f a com m issio n established by the cities o f Crete to draw up the c o n s titu tio n for a new colony. T h e rest o f th e Laws in effect consists of th e A th en ian ’s suggestions as to w hat form this constitution should tak e an d w hat laws should be enacted for the new city. Books IV and V are co n cerned w ith th e prelim inaries to legislation. A description of th e site proposed for th e new colony gives the A thenian a chance to p ro n o u n c e on th e geographical conditions w hich are m ost suitable for a w ell-governed city (704a-707d). T h e n there is more general discu ssio n of th e problem s o f constitution making (708c-724b). Laws sh o u ld be given 'p relu d es’, i.e. pream bles designed to persuade citizen s to obey the law of th eir ow n free will rather than through fear o f th e penalty. T h e first part of Book V consists of a long prelude to the legal code as a whole (726a-734d). T h is is, in effect, a general e x h o rta tio n to virtue. T h e n the conversation turns to the size of the p o p u latio n , th e distrib u tio n of land and th e adm inistrative divisions of th e state (734e-747e). T h e arrangem ents are designed to ensure that e a c h household has an equal am ount of land w hich will be inalienable. T h e re will also be restrictions on the am ount ç f movable property w hich can be held. T h e legislative code p ro p er occupies Books VI to X I I. F irst there is w h a t we m ight call co n stitutional law — the arrangem ents for holding assem blies and electing the council and m agistrates and descriptions of th e ir responsibilities. T h ere follows a detailed exposition of the laws by w h ich th e city will be governed. I t begins w ith m arriage and the family



and moves through education and m ilitary training to agriculture and economic m atters. T h e central topics of crim inal law — theft, m urder, wounding and the like — are dealt w ith in Book IX , where there are substantial discussions of the purpose of p u n ishm ent and of responsibility. Book X is theoretically concerned with the law against impiety b u t is in fact a closely argued exposition of the central doctrines of Plato’s theology. A lthough this looks like a digression, it is clearly intended to provide a religious and m etaphysical underpinning for the political and moral ideas that occupy the other books o f the Laws. Book X I and the first part of Book X II are concerned with legislation on a variety of unconnected topics and appear to consist of fragm ents. T he dialogue ends with an account o f the nocturnal council. T he function of this curious body is to act as the reason o f the state by discerning the aims of legislation and the m eans of achieving them. To this end it will engage in philosophical research, concerning itself particularly w ith the unity o f virtue and the natu re and existence of God.


The character o f the constitution

It would appear that m ost features of the constitution proposed for the new C retan city derive from the practice of one or other of the G reek cities (M orrow , 1960). Plato is most strongly influenced by the law and legal practice of Athens, though his model is not so m uch the ‘extrem e’ democratic constitution of his own day as the ‘m oderate’ constitution supposed to have existed at an earlier period, when the Areopagus, a council of ex-m agistrates, was the dom inant influence. Some elem ents of the constitution are based on Sparta and others possibly on less wellknown cities. T h e A thenian often suggests that he is providing only an outline of w hat should be included in the legal code b u t on many matters he goes into considerable detail. T he Laws is, in fact, a systematic attem pt to codify and reform G reek practice in the light of a clear understanding of the ultim ate aims of law. In this respect it has, quite reasonably, been com pared w ith the work of B entham in Britain. T he new city is intended to be quite small — the ideal num ber of citizen households would be 5040. T he arrangem ents for the distribution of land and the restrictions on the am ount of movable property th at may be held are designed to ensure th at the citizen body will consist entirely of small farm ers, though Plato presum ably envisages that m ost, and perhaps all, of th e actual physical labour will be done by slaves. T rade and m anufacture -are forbidden to citizens



a n d will be carried o u t by resident aliens. It is arguable th at Plato has n o t seen, or has ignored, the full im plications of these m easures, since m any o f his p r o p o s a l seem to envisage a large com m unity w ith a more d eveloped econom ic life. In particu lar it is difficult to see how the citizens could find all th e tim e th at Plato expects them to spend on civic duties. T h e C onstitutional proposals are elaborate and in m any cases it is d ifficu lt to assess th eir full significance. T h e aim is clearly to produce a balance o f different elem ents along th e lines suggested in Book I II. T h e re is an assem bly atte n d e d by all citizens b u t it is not clear whether it has any real fu n ctio n ap art from th e election o f m agistrates. T he electoral arrangem ents are often com plex, som etim es com bining straig h tfo rw ard elections w ith selection by lot. It is clearly intended th a t th e ‘guardians of th e laws’ a body of thirty-seven very senior m agistrates should exercise th e greatest power. So, although ultim ate respo n sib ility m ay rest w ith th e citizen body as a w hole, in practice the m ain influence on policy will be exerted by a sm all group of elderly m en , d istinguished by th e ir long experience and th eir m oral and intellectual excellence. A dom inant them e is th e sovereignty of law. All inhabitants of the city will be subject to law in all th eir activities and the legal code will be alm ost unalterable. Even th e activities of the m ore im portant m agistracies will be strictly regulated — all will be subject to some k in d o f m easures to ensure th at they conduct them selves in accordance w ith law. T h u s law in some ways takes the place o f the philosopher kings who exercise sovereignly in the Republic. Since the ultim ate purpose o f law is, in P lato ’s view, to make the citizens virtuous, law and ed u catio n are so closely linked th at at tim es they becom e almost indistinguishable. T h e en tire life of th e com m unity is to be regulated so th a t the citizens develop th e rig h t kind of character. T his is the p u rp o se o f th e regulations th at strictly control the educational arrangem ents, literary and artistic activities and the religious life of the city. C ontacts w ith th e outside w orld will so far as possible be restricted to avoid subversive influences.


From the R epublic to the Laws

T h e re are m any differences, certainly of tone and em phasis, and pro b ab ly also of d octrine, betw een th e Republic and the Laws. T he m ost striking of these — betw een the two accounts of the ideal city — will be considered in the next chapter. Equally im p o rtan t is the way in



which the Laws differs from the dialogues of the early and m iddle periods in its account of virtue. In dialogues like the Charmides, Laches, Protagoras and Gorgias, Plato seems to adhere to the Socratic doctrine that virtue is knowledge, while the Republic suggests th at only the philosopher can be truly virtuous because only he has knowledge o f the good. In the Laws, on the other hand, th e suggestion seems to be that the citizens can become virtuous if only their desires and passions are disciplined in such a way that they obey the law as a m atter of habit. Discrepancies like this are particularly distu rbing to those who believe th a t the au th o r o f the Republic is the only tru e Plato. For this reason some G erm an scholars have seen th e Laws as a chaotic product of senile decline (see M uller, 1951). Those prepared to take a more positive attitude to the Laws can explain discrepancies betw een it and earlier dialogues in either of tw o ways. O ne is to suppose th at Plato’s philosophy had evolved over the years. A long line o f scholars has argued that Plato started from a Socratic intellectualism which stressed m an’s powers of reason to the virtual exclusion of the passions and desires and that he gradually m oved from this to a position in which more im portance is attached to these ‘low er’ elem ents in the soul. As D odds (1951, p. 212) puts it: ‘P lato’s grow ing recognition of the im portance of the affective elem ents led him beyond the limits of fifth century rationalism ’ (cf. O ’Brien, 1967, ch. 6). It is easy to see the Laws as the culm ination of this process. T h e second approach emphasizes th e fact that the Laws seems to have been w ritten for a different purpose and to be directed to a different readership from that envisaged for the Republic. This view is forcefully m aintained by G orgem anns (1960) who argues that the apparent discrepancies betw een th e Laws and earlier writings can generally be understood if one appreciates that it is supposed to be w ritten at a more ‘popular’ level. It is obvious, even on a superficial reading, that the Laws has a more practical and less distinctively philosophical orientation than most other dialogues. A part from anything else, we can point to the am ount of space devoted to detailed legislative proposals, something the Socrates of the Republic thought it unnecessary to consider (425c-e). T here are other indications th at the Laws may be intended for a different kind of audience. T h e A thenian’s two com panions have no enthusiasm for, and no expertise in, th e niceties of philosophy. They are represented as serious-m inded old m en, very interested in the problem s of legislation and devoted to their own cities b u t quite w ithout philosophical sophistication. Perhaps Plato envisages that his readers will share the same cast o f m ind. At one point (810e-812b) he



sug g ests th at th e Laws or som ething like it m ight serve as a textbook for schoolch ild ren , which presum ably im plies that parts of the dialogue, a t least, are in ten d ed to have a m orally im proving effect on a relatively u n so p h isticated audience. T h en there is the insistence th at laws should h av e pream bles designed to persuade th e citizens voluntarily to give th e ir obedience. Sam ples o f these pream bles occupy m uch space in the dialogue. A lthough they som etim es have a substantial philosophical c o n te n t they are prim arily in ten d ed to provide moral exhortation. S im ilarly, if we are concerned w ith th e practical problem s of the legislator, there is not m uch point in dw elling on the ‘tru e ’ virtue of the Republic w hich can be achieved only by philosophers. W e m ust c o n c e n tra te, ra th e r, on th e ‘p o p u la r’ virtues th at are w ithin the cap acities o f th e m an in th e street. T h u s, given w hat we can glean a b o u t th e different purpose of the Laws, as com pared w ith dialogues th a t have a m ore narrow ly philosophical aim , it is not surprising that m an y topics should be treated w ith a rath er different slant. I t m u st be em phasized th a t th e ‘developm ental’ and the ‘change of le v e l’ view are not totally incom patible alternatives·betw een w hich we hav e to make a o n ce-and-for-all choice. Both are likely to contain an e le m e n t o f tru th . We w ould naturally expect Plato’s thought to have u n d e rg o n e som e evolution in the period since he w rote the Republic a n d we w ould also expect him to ad ap t the co n ten t of the Laws to suit th e p articu lar purposes for w hich it was w ritten. It is com m on ground th a t th e Laws is a m ore practically orien ted work than the Republic and t h a t th is practical o rien tatio n affects the treatm en t o f alm ost all its to p ic s. T h e d ifficulty is th a t of d eterm in in g in w hat respects Plato believes th at his earlier position was m istaken and in w hat respects he is m erely suppressing earlier doctrines as irrelevant to his im m ediate p u rp o ses. Since Plato him self provides no explicit indications, one ca n n o t expect to answer such questions by considering the Laws in isolation. U ltim ately one has to decide w hat is plausible in the light of o n e ’s overall assessm ent o f Plato’s philosophical character. 6

English versions

T h e Law s is a particularly difficult w ork to translate. Its sentences are o fte n long and unw ieldy w ith an unclear gram m atical stru ctu re, its id io m is som etim es u nusual an d in m any passages there are obscurities w h ich m ight be due either to co rru p tio n o f the surviving m anuscripts o r to th e unrevised state in w hich Plato left the dialogue. M y ow n in te re st in th e Laws was p ro m p ted by the appearance of T . J. S au n d ers’ P enguin translation (1970a). Saunders makes it



abundantly clear that his main aim is to render the Laws easily accessible and intelligible to the m odern reader. H e divides the text into short sections for w hich he provides titles, he introduces b rief passages of com m entary and he distinguishes typographically the actual laws from discussions in w hich they are em bedded. His table of contents and list of crimes and punishm ents can be very helpful. T he translation itself is designed, above all, to be readable, and Saunders freely acknowledges th at he often has to w rite an in terp retatio n into the text, and som etim es even ‘o v er-in terp rets’ it. F or these reasons Saunders’ version is not the best to use w hen one needs to exam ine the detail of particular passages, b u t I have no doubt th at G reekless readers, and perhaps even some who can read G reek, are m ore likely to find the Laws accessible and to achieve an understanding o f its overall themes by reading Saunders, than by reading any other English translation. Among S aunders’ predecessors, Jow ett (1953) is readable b u t free. Bury’s Loeb translation (1926) is on th e whole reliable, though occasionally m arred by the adoption o f unlikely readings o f the Greek text. Taylor (1960a) is also in general accurate, b u t his use o f archaic English can be irritating. T he editions o f the G reek used by these earlier translators are inferior to the Bude edition (des Places and Dies, 1951-6) used by Saunders, b u t this seldom affects m atters of philosophical interpretation. T he best policy for G reekless readers embarking on a serious study o f the Laws is alm ost certainly to start with Saunders b u t to check the details of im portant passages against Bury, Taylor or the French version of des Places and Dies. A translation of the Laws by Thom as Pangle (1980) has appeared in the U nited States b u t is not generally available in the U nited Kingdom. Pangle, who is not him self a classical scholar, is strongly critical o f the ‘loose’ translations produced by traditional classical scholarship and he is particularly critical o f Saunders. H e aims to be as literal as possible and believes th at he can thereby com e closer to providing ‘direct undistorted access to Plato’s thought, exactly as Plato understood it’. T his aim , in my view, displays a m istaken philosophy of language and a related m isunderstanding of the purposes o f classical scholarship. If texts were com posed of w ords in the same kind o f way that walls are m ade o f bricks one could hope to produce a good translation sim ply by consulting a dictionary and su b stitu tin g English equivalents for each Greek word. B ut in reality the context in which a word occurs largely determ ines its m eaning; atran slatio n right for one appearance of a w ord can easily be wrong for another. T hus, as Karl Popper puts it, ‘every good translation is an interpretation of the



original te x t’ (1976, p. 23). A very literal translation may be m isleading, or m ore usually, devoid o f iiieaning. T his is why classical scholars have always left w ord-for-w ord tran slatio n to schoolboys and have concen trated instead on conveying th e th o u g h t of th e original as th ey u nderstood it. I t is an illusion to suppose th at a very literal translation can give us anything like direct access to P lato’s thought. I w ould not th erefore recom m end the use o f Pangle’s translation on its o w n , th o u g h it may have value as a check on Saunders or as a ‘crib ’ for those reading th e Laws in G reek.


Plato’s Political Philosophy Charmides, 162e-175d; Protagoras, 311b-320c; Gorgias, 447b-461b; Euthydemus, 288d-292e; Republic, 368c-369c, 471c-474c, 479a-502c, 540d-541b, 592a-b; Statesm an, 291d-303c; Epistles, V II, 323-337e, V III

Even from the brief sum m ary in the first chapter it will be obvious to most readers th at the city described in the Laws differs quite m arkedly from the city of th e Republic. For exam ple, in the Republic all power is entrusted to ‘guardians’ who have undergone a long training in philosophy, b u t in the Laws ultim ate sovereignty is supposed to rest with th e law rather th an w ith any particular group o f citizens, there is much less em phasis on the need for philosophical training than in the Republic, and all citizens, not just a select few, are expected to take some p art in governm ent. I® the Republic the rulers and the soldier class are forbidden to have property or even families o f their own, but in the Laws all citizens are to have th eir ow n farms and families. T he Republic is bitterly hostile to dem ocracy b u t the constitution of the Laws has some decidedly dem ocratic features. Broadly speaking, there are two ways of accounting for these discrepancies. T he first supposes that th e change from the Republic to the Laws represents a change o f heart on-Plato’s p art. M ost scholars who take this view would associate th e change w ith Plato’s own experience o f politics. In 367 to 366 he had been invited by his friend D ion to go to Sicily to assist with the education of D ionysius who had just succeeded his father as tyrant of Syracuse. T his scheme was totally unsuccessful: D ion was forced into exile and Plato retu rn ed to Athens. He was persuaded to go to Syracuse again in 361 b u t this tim e things, if anything, w ent even worse and Plato him self had difficulty in escaping. In 357 D ion led an expedition which seized control of Syracuse, b u t his attem pts at reform broke dow n amid factional disputes, and D ion was him self assassinated. It is tem pting to suppose th a t w hen Plato w ent to Syracuse he was



h o p in g to tu rn D ionysius into a philosopher king and to set up a state so m eth in g like th at described in the Republic. W hen the-Syracusan fiasco convinced him that these schem es were im practical, he devoted ' h im se lf to designing a m ere realistic blu ep rin t for politics — the city of th e Laws. So the Laws, on this reading, is a product of despair. An alternative view of the relationship betw een the Republic and the L a w s suggests th at these dialogues differ in th eir approach, not b ecau se P lato has had a change o f heart, b u t because they are w ritten w ith different intentions. O ne possibility is th at the Republic is in te n d e d to exhibit in a graphic way th e fundam ental principles of m orality and politics, while th e Laws seeks to show how these p rin cip les m ight be applied in practice. Saunders, for exam ple, w rites, ‘P lato could perfectly well have w ritten the Laws w hen he w rote the Republic and the Republic w hen he w rote the Laws, for they are the o p p o site sides o f th e same coin’. In S au n d ers’ view the Republic p re se n ts a ‘theoretical ideal’ w hile th e Laws ‘describes, in effect, the Republic m odified and realised in the conditions of this w ord’ (1970, p. 28). It has to be said at once th a t S aunders has overstated his case in sug g estin g th at th e Republic and th e Laws could have been w ritten at th e sam e tim e. O n any reading th ere are divergencies of doctrine b etw een th e tw o dialogues, p articularly in m atters of ethical theory. B u t it is still plausible to suppose th at there is a broad continuity b etw een th e political doctrines im plied in the Republic and those e x p o u n d ed in the Laws.


The philosopher king

M an y o f the dialogues argue explicitly or im plicitly th at excellence in p olitics is a m atter o f know ledge, so th at the ideal constitution would bestow u nim peded au th o rity on th e person(s) who genuinely know w h at to do. O n this view, dem ocracy, w ith its assum ption th at all are equ ally qualified to rule, is an absurdity. T he Charmides (162e-175d), Protagoras (esp. 311b-320c), Gorgias (esp. 447b-461b) and Euthydem us (288d-292e) com pare political knowledge to the skill o f a craftsm an and point to the conclusion th at this know ledge consists in a seco n d -o rd er skill enabling us to make proper use of all other skills. T h e Republic begins, like th e earlier dialogues, by com paring justice w ith everyday skills b u t goes on to give a m uch fuller account of the seco n d -o rd er know ledge req u ired by the ruler: it involves an in tellectual grasp o f the ‘F o rm s’, som ething achievable only by a tra in e d philosopher. T h e conception of politics as a skill reappears in



the Statesman (293c-302b), though this dialogue recognizes th at the unfettered rule of the philosopher king is not a practical proposition. It is im portant to see that the doctrine of the philosopher king in itself gives no direct guidance as to what constitutional form s should be adopted. T he philosopher king w ould rule w ithout restriction, but so too would the ty ran t, and Plato regards ty ranny as the worst degradation (Republic, 562a-580c; cf. Gorgias, 470d-472d; Statesman, 302c-303b). C onstitutional forms cannot by them selves guarantee that the ruler is enlightened. If we could find a suitable philosopher we m ight initiate a revolution to p u t him on the th rone b u t it w ould be impossible to ensure th at he continued to rule in an enlightened way or that a suitable successor was available w hen he died. Since it is im possible to provide any institutional guarantee th at the ruler will be wise, two different, b u t com patible, strategies appear to be available to those who hold politics to be a m atter o f knowledge. T he first is political education: if those likely to gain pow er are given an appropriate training there is a chance that the state will be wisely governed. T he second strategy is to adopt constitutional form s that encourage wise decisions or, at least, help to prevent foolish on es. This can be done in a variety of ways. I f the area of discretion granted to rulers is tightly circum scribed by law, they may be prevented from doing some of the m ost dreadful things; it may also be possible to ensure th at political decisions are taken only after prolonged discussion in which all points o f view can be heard; and the requirem ent th at candidates for election have appropriate experience or education may help to bar the m ost unsuitable characters from holding office. It is natural to suppose that one of P lato’s reasons for establishing the Academy was to create a training ground for those likely to have political influence — this was certainly one of the functions it served in practice. T o the extent that this was part of P lato’s intention, he was, in effect, following the first strategy, that of education. T he Laws im plicitly advocates a com bination of the two strategies. T he constitution proposed for the new C retan city consists of a set of interlocking measures designed to encourage wise governm ent, and this is accom panied w ith a constant em phasis on the need for education. So, although the. Laws does not advocate the direct rule of a philosopher king, it can be seen as an in terp retatio n at the practical level of the same underlying ideals.

• 16



The impracticality o f the R epublic

In th e Republic Plato shows no interest in law or in constitutional proced u res. T h e rule o f th e philosopher kings is m aintained firstly by m eans o f th e class system , w hich ensures th at those who come to political pow er have th e rig h t kind o f n atural endow m ent and have b een p ro tected from the tem ptations o f p ro p erty and fam ily ties, and secondly by a system of education th at is supposed to give the rulers th e rig h t kind o f philosophical u nderstanding. M ost readers of the Republic have ap preciated th a t this system could n o t be actualized am ong real m en. F or exam ple, even if hum an beings could be divided in to th ree d ifferen t intellectual categories, it is unbelievable th at the m ajo rity o f th em w ould meekly accept the im position of the Republic’s class system. T h e suggestion th at th e ideal state could be established a n d m aintained only by force does not help, since force could be effective only if a substantial body of the population could be p ersu ad ed to su p p o rt th e ideal. It is equally incredible th at the rulers, having gained pow er, should give u p all claims to property and family relationships. I f Plato d id n o t see these and sim ilar difficulties he w ould have been rem arkably naive. O f course, history has seen a good m any w ould-be reform ers conspicuous for their naivety, b u t Plato gives clear indications th a t he is not am ong them : several passages in th e Republic suggest th a t its ideal state is n o t to be seen as a practical possibility. In th e Republic, V, Socrates is m ade to em phasize the difficulty of setting u p th e state he is describing and to m aintain th at its v alid ity as an ideal does not d epend on its practicability. N evertheless he claim s th a t th ere could be such a state if philosophers becam e kings o r kings philosophers (471c-474b). L ater he says th at to establish the sta te it will be necessary to expel from th e city everyone over ten years o ld (540e-541a). He does n o t explain who would look after the city in th e absence o f th e adu lts or who w ould tra in the children. W hen, near th e en d o f th e discussion, G laucon d o u b ts w hether the state can ever be realized, Socrates replies th at it does not m atter w hether the state does or will exist. It-is a p a tte rn set up in heaven on w hich w hoever w ishes may m odel him self (592b). O fficially, at least, th e Republic is concerned, not w ith the state or the c o n stitu tio n , b u t w ith th e v irtue of justice. T h e state makes its ap p earan ce only by way o f analogy: Socrates suggests th at it will be easiest to look for justice first in the state and th e n to apply w hat has b een learn t to th e justice o f th e individual soul (368c-369c). T o this e n d he sets off to describe a perfect state w hich will perfectly in stan tiate th e virtues he is seeking. Som etim es he seems to lose sight



of the original purpose, b u t it is significant that w hen in the passage quoted above (592b) he finally draws his conclusions from the analogy, the ideal state is offered as a heavenly model for ourselves (not for our cities). T his m ethod of argum ent does not perm it Socrates to make any concessions to practical difficulties, since, if he did so, his state would be im perfect and could not offer a picture of perfect justice. B ut he is only too well aware th at this is n o t a perfect w orld. So, although his doctrine that the city of the Republic is the ideal presum ably com m its him to saying th at we should all aspire tow ards it, n o th in g com m its him to the belief that real m en will ever succeed in getting at all close to that ideal. T here is, therefore, no reason to suppose th a t w hen he went to Syracuse Plato took with him the Republic as a plan or b lu eprint for political action.


The Statesm an

Those who hold th at Plato changed his m ind betw een the Republic and the Laws are apt to believe th a t the Statesman shows him in the act of making this change. T he relevant section o f th e dialogue begins at 291c where the Eleatic Stranger, who leads the conversation, launches into criticism of the ways in which different kinds o f constitution are conventionally classified. People classify them firstly according to the num ber of rulers, w hether they are ruled by one m an, a few or the many, and secondly according to w hether they rely on enforced subjection or voluntary obedience, w hether they are rich or poor and w hether they follow law or are lawless. F or exam ple, the rule of one man according to law is called m onarchy, b u t w ithout law it is tyranny. T he S tranger’s com plaint is th at this classification ignores the fundam ental question — w hether the ru ler has the requisite political knowledge. For example, someone is said to possess the knowledge of medicine provided sim ply that he knows how to make people well; it is irrelevant w hether he exercises this skill over few or m any, qver willing or unw illing subjects. Sim ilarly, since tru e statesm en know w hat is best for th eir cities it does not m atter w hether they act with or w ithout law, w hether they are rich or poor or w hether th e ir subjects do or do not consent (292b-293e). A fter this apparent reassertion o f the philosopher king doctrine, attention shifts to the role o f law. Two parallel passages (294a-295b, 295b-302b) argue that laws are theoretically absurd b u t necessary in practice. T he first argum ent concentrates on the great differences betw een individual hum an beings. Because of those differences there



is n o reason to suppose th a t a general law laid dow n for all men will achiev e w hat is best for each individual. But, equally, the statesm an c a n n o t sit beside each person all the tim e telling him what to do. So th e re is no alternative to th e use of law. In this the statesm an is rather lik e a gym nastic in stru cto r w ho, being unable to specify advice to each m e m b e r o f his class, contents him self w ith telling them w hat is good fo r th e m ajority (295a-b). T h e second argum ent begins by re-em phasizing the theoretical a b s u rd ity o f law. A d o cto r w ho had to leave his patients for a while m ig h t leave th em w ritten in stru ctio n s ab out w hat to do in his absence, b u t we w ould not expect him to b in d him self by those instructions if he re tu r n e d early to find th a t his p atien ts’ condition had changed. S im ilarly it would be ab su rd for a dem ocratic or oligarchic g o v e rn m e n t to a tte m p t to regulate by law exactly how sailors were to sail th eir ships or doctors to heal their patients. In each case the point is t h a t we expect experts to use th eir ow n judgem ent in particular cases ra th e r th an to follow general rules laid dow n in advance. But, the S tra n g e r goes on, it w ould be even m ore absurd if, after establishing g e n e ra l laws, we allowed officials chosen by election or lot to set them asid e sim ply to cu rry favour. T h e conclusion of all this is th a t, since laws are the product of ex p erien ce and are b ro u g h t into being only after advice has been taken a n d th e people p ersu ad ed to accept th em , it is best to make sure that th e y are obeyed. A ru ler w ith tru e know ledge w ould have no need o f law s b u t, as things are now , there are no such rulers. So we m ust im ita te th e tru e co n stitu tio n by m eans of law. O n this basis the S tra n g e r constructs a new ranking o f constitutions. W here law is su p re m e , th e b est co n stitu tio n is th e rule o f one m an, the next best is th a t o f the few and th e w orst is th at o f the many. W here law is not fo llo w ed the o rd e r is reversed. T h e p o in t here is th at a single ru ler has m o st pow er b o th to do good and to do harm . D em ocracies are weakest in b o th respects. A lthough th e Eleatic S tran g er moves from the praise o f a p h ilo so p h e r king to th e advocacy o f a state ruled by law he does not a p p e a r to change his m ind on any point o f principle. He com bines the tw o ideas by showing th at they operate on different levels. T he aim of sta te c ra ft is th e good of th e city, and it is knowledge w hich norm ally en ab les us to achieve such goals. B ut a ruler who could direct every d e ta il of the com m on life in accordance w ith knowledge w ould have to b e m ore th a n hum an. In th e w orld as it actually is the best we can do is im ita te know ledge by m eans o f law. T h u s the philosopher king



remains valid as a theoretical ideal while the rule o f law is recom m ended in practice.


The letters

O f the letters th a t have com e dow n to us under P lato’s nam e, three, all of them widely accepted as genuine, cast some light on the Laws. Letter I I I addressed to Dionysius, purports to be w ritten soon after Plato retu rn ed to Athens in 366 or 365. L etters V II and V I I I , both addressed to the friends of D ion, p u rp o rt to be w ritten soon after D ion’s death in about 353 BC. O f these, letter V II is prim arily a defence of P lato’s own involvem ent in Sicily, though it does offer some practical advice. L etter V III, supposedly w ritten about the same tim e, is largely concerned to suggest the policies that the friends of D ion should follow. If genuine, these letters should cast a good deal o f light on the developm ent of P lato’s political ideas. In letter V II Plato tells how the experiences o f his early m anhood, in particular the death of Socrates, discouraged him from active politics. He believed that the laws and custom s of Athens and o f other cities had become hopelessly corrupt and th at philosophy alone could reveal the true nature of justice. So he concluded that the troubles o f the hum an race would not cease until either those who were true lovers o f wisdom achieved political office, or those who held political pow er becam e lovers of wisdom them selves. It was in this frame o f m ind th at Plato came to the court of D ionysius (325c-326b; cf. Republic, 473d, 487c489b, 50 le). He makes it clear that by using his influence on Dionysius he hoped to unite philosophers and rulers (328a). T h is letter is now seen as an im p o rtan t source for those who believe th a t Plato hoped to turn D ionysius into a philosopher king. But Plato does not say that he expected to actualize the city of th e Republic or anything like it. His main em phasis is on law. It was the corruption of the laws and customs of cities like A thens that distu rb ed him (325c-d, 326a), and in going to Syracuse he hoped to p u t into practice his ideas ab o u t ‘laws and the constitution’ (328c). H e taught that cities should be enslaved, not to hum an despots, b u t to laws (334c-d) and he eschew ed violent revolution (331c-d). In letter I I I , 316a, he m entions th a t he worked with D ionysius on pream bles to laws, w hich suggests th at he had already form ulated the idea that laws should have persuasive pream bles, one o f th e m ore distinctive ideas o f the dialogue, the Laws, as we have it.



T h e advice P lato gives to th e friends of D ion carries a similar e m p h a sis on th e im portance o f law. D ion, he believes, w ould have s o u g h t pow er, not for its ow n sake, b u t in order to create the best c o n s titu tio n and laws (351c). His followers should likewise seek to eq u alize the laws, i.e. im pose laws for the com m on good rather than for th e benefit o f th e ir ow n faction, and they should induce their o p p o n e n ts to obey these laws by exam ple (i.e. by obeying law th em selv es) as m uch as by fear (336c-337b). T o this end they should in v ite fifty wise m en from all over G reece to draw up laws fair to all (3 3 7 b -d ). T h e advice in le tte r V I I I is even m ore explicit. Plato urges th e followers o f D ion to establish th e suprem acy of law and to avoid the excess b o th of freedom and servitude by establishing a form of c o n stitu tio n a l m onarchy (354a-356b). H e recom m ends a constitution w ith th ree kings (D ionysius, H ip p arin u s, a half b ro th e r o f Dionysius w h o had joined D io n ’s faction, and D io n ’s own young son) whose fu n c tio n s seem largely cerem onial and religious. Real power would r e s t w ith th irty -fiv e g uardians o f th e laws ruling together w ith an assem b ly an d council. T h ere should be a special co u rt o f select judges to try capital cases. T h e triple kingship is apparently proposed simply to reconcile th e conflicting factions in Syracuse. All the other p rop o sals are to be found in th e Laws (though the num ber of guardians is th ere increased to thirty-seven). A ssum ing th a t the letters in question are genuine, it is clear that, at le a st by th e en d o f his involvem ent in th e affairs o f Syracuse, Plato was b eg in n in g to develop ideas characteristic o f the Law s and to p u t his p o litical faith in the suprem acy o f law. H e attributes sim ilar ideas both to his ow n younger self an d to D ion, and he does not appear to d istin g u ish these ideas at all sharply from th at o f the philosopher king. O n his own account w hen he first w ent to Syracuse he worked to e sta b lish good laws and th o u g h t that, as a prerequisite for this, the ty ra n t D ionysius w ould have to be trained in philosophy. W hen, on th e o th er h an d , he gives advice to th e followers o f D ion he does not su g g est th a t they should engage in philosophical studies. B ut there is n o evidence to show th a t he ever believed that philosophers could in p ractice rule u n en cu m b ered by law and no thing to suggest th at he was conscious o f any m arked change in his political principles. At m ost, th e re seems to be a gradual shift in em phasis from the distinctive w isd o m o f th e philosopher to a m ore h u m ble kind of w isdom based on ex p erience and enshrined in law.



5 Plato’s politics The picture th at emerges from this survey of P lato’s w ritings shows him starting w ith the doctrine th at politics is a m atter o f knowledge. This doctrine culm inates in the description o f the philosopher kings in the Republic. In view o f the fact th a t there is no practical way of guaranteeing that our rulers will possess the wisdom o f the philosopher, this ideal o f a philosopher king can w ork only at a theoretical level. But it does nevertheless have some practical implications in that it will lead us to prefer institutions which encourage rationality in the taking o f decisions. Sim ilarly, although the social institutions of the Republic could never be actualized one who sees them as an ideal will in practice favour institutions which encourage unity and harm ony rather than dissent and diversity. W herever Plato addresses him self to m atters of practical politics he advocates th e suprem acy of law, a doctrine worked o u t in great detail in the Laws. T his is entirely com patible with his doctrines at the theoretical level. If, as th e Statesman suggests, law em bodies decisions of thecom m unity taken after careful consideration, the rule o f law may be the closest ordinary m en can come to the rule of knowledge. So the Republic and the Laws can be seen as com plem entary to one another. A lthough there seems to be no change of principle betw een the Republic and the Laws it is probably no accident that th e works w hich treat politics at a practical level belong to the later period of Plato’s life. No doubt events in Syracuse helped to focus P lato’s attention on matters of practical politics, b u t other factors may be equally im portant. We know that Plato’s attitude to the fundam ental questions of philosophy changed betw een the Republic and later dialogues such as the Sophist and Statesman. In the Republic and other dialogues of the same period, Plato espouses the theory of Form s. As it appears in these dialogues the theory is concerned w ith the relationship betw een particulars and universals. Plato argues that the particular instances of, say, justice in this world may be called ‘ju st’ only in so far as they resemble or participate in the Form o f the just. T h e Form s are thus treated as paradigm s which the things of this w orld im perfectly resemble. T h e role of the philosopher is to turn his back on the everyday w orld and devote his atten tio n to the F orm s. In dialogues such as the Theaetetus the Sophist and the Statesm an, the Form s, if they appear at all, are no longer treated as paradigm s. Plato is more concerned w ith the relationships betw een concepts than with that between universals and particulars. He discusses how and to what extent very general concepts such as ‘being’, ‘sam eness’, ‘difference’,



‘m o tio n ’ and ‘re st’ may com bine w ith one another and devote a lot of atten tio n to the analysis o f m ore specific concepts such as those of the ‘so p h ist’ and th e ‘statesm an ’ or even of the ‘an g ler’ and the ‘w eaver’. T hese discussions are ch aracterized by a m uch greater concern for the w orld o f em pirical realities. T his change in P lato’s philosophical m ethod has been th e source o f m uch controversy (see F u rth e r R eading, section 8). I t m ay be connected w ith a new em phasis in his political thought. In particular, it may help to explain why he no longer talks o f a p aradigm city set u p in heaven b u t attends instead to cities in th e real world and why he shows so m uch more respect for experience. — H u m an beings are im perfect creatures living in an im perfect world. We are not fully rational or v irtuous and our needs and desires have an u n fo rtu n ate ten d en cy to o u tstrip the means available for their fulfilm ent. I f we are to have any decent kind o f life we need to co­ operate, yet we constantly com e into conflict w ith one another. T he task o f politics is, so far as possible, to recencile conflicting desires and interests and to create u n ity in place of discord. Itw o u ld have no place in a golden age w hen everyone was virtuous and w ant was unknow n. T h e Laws is largely concerned w ith these practical problem s o f making a good life possible in th e w orld as it actually is. T h e Republic, on the o th er hand, has n o th in g to say on these m atters and restricts itself to describing a com m unity in w hich hum an im perfections have been abolished. So if we are interested in Plato’s political theory — that is, in his answ ers to w hat we now und erstan d as political problem s — it makes sense to regard th e Laws as the prim ary source w hile treating the Republic as a kind o f prolegom enon designed to set out the underlying principles.


The Nature of Law Laws, 624a-632d, 644d-645c, 712b-716b, 769a-771a, 7 7 2 a-d , 8 7 4 e876e, 889a-890d, 960c-964a, Hippias M ajor, 284a-285b; Sta tesm a n , 29lc-30 3c; Minos (although this dialogue is probably sp u rio u s it makes an interesting com m entary on the Laws)

1 Law in the Greek city For m ost English speakers, the paradigm case o f law is a statute enacted by Parliam ent or Congress. B ut there is, of course, m uch law that does not fit this p attern — the com m on law, early custom ary law, international law and so on. T he G reek word nomos (.plural: nornoi) has, if anything, an even w ider extension. It may refer to any form of order which is actually accepted or which ought to be accepted by a social group (Ostwald, 1969, pp. 20-54). T h u s it can apply to w hat we would regard as rules of morality, or etiquette, to religious beliefs and practices, to the general way of life of a particular com m unity and even to the habits of m ankind as a whole w hen com pared, for exam ple, to those of the beasts. O ften the best translation is not ‘law’ b u t ‘convention’ or ‘custom ’. Ostw ald (1.969) has argued that the application o f nomos to what we would call ‘statute law ’ was a com paratively late developm ent, associated in A thens w ith the rise of democracy. The dem ocrats preferred nomos to the older term thesmos because it suggested a norm accepted by the people, rather than a measure imposed on them from above. T h e wide scope of nomos is reflected in the content o f the Laws. Plato is concerned with all kinds of social norm , including the fundam ental principles of th e constitution, laws prohibiting crim es such as m urder and theft, the regulations that will govern agricultural and com m ercial activities, standards of religious observance, th e ways in which children will be b ro u g h t up and patterns o f sexual behaviour. He is aware th a t some aspects o f hum an behaviour cannot be governed by w ritten statutes enforced by the courts, but he does not consider that



they th erefore fall outside th e scope o f nomos. In stead he distinguishes betw een w ritten and u n w ritten laws. H e uses this distinction in at least th ree different contexts. (1) Som etim es th e point is th at it is, as a m atter o f practice, in ap p ro p riate to lay dow n form al rules w ith penalties to regulate the details of, say, fam ily life or behaviour in the h u nting field (788b, 7 9 0 a-b , 822d-823a). (2) At 7 9 3 a-d he seems to be m aking the im p o rtan t p o in t th at a legal system requires a background of accepted m oral beliefs, custom s an d conventions. T h ese may not be p a rt of the form al legal code, b u t w ith o u t them it w ould collapse. (3) At 8 3 8 a-b and 841b the u n w ritten laws are apparently sexual taboos, w hich m ay influence b ehaviour m ore pow erfully than any form al legislation. In none o f these cases does P lato suggest th a t there is a fundam ental d ifference 6f kind betw een th e w ritten and th e u n w ritten laws. All areas o f life fall w ithin th e leg islator’s concern and it is his task to establish appro p riate norm s o f behaviour. W here formal penalties and w ritten laws are inappropriate, he has to rely on exhortation and p ersuasion, b u t there is no difference o f principle betw een the two. O ne consequence of th is is th a t Plato does not recognize a distinction betw een law and m orality. N o r is there an area o f private life with _w hich law may not, in principle, interfere.


L aw and purpose

T h e Laajs opens w ith a question ab o u t the source of law: ‘D o Cleinias an d M egillus a ttrib u te the C retan and Spartan systems to a god or to som e m an?’ Both piously a ttrib u te th eir ow n systems to gods..T his suggests th a t they accept some kind o f divine com m and theory — law dem ands our obedience because it em bodies the edicts o f the gods. O ne im plication o f th a t theory w ould be th at established law is beyond criticism because it rests on divine authority. L ater the A thenian m akes it clear th at he too gives law some kind o f religious basis, b u t for th e m om ent he does n o t dwell on this point. Instead, he alters direction an d asks his com panions ab o u t th e purposes o f th eir own cities’ m ost distin ctiv e in stitu tio n s (625c). In this he sets the tone for the whole dialogue. A ttention is focussed on the end or purpose th a t law is supposed to serve m uch m ore th an on its source. Unlike the divine com m and theory, this perspective inevitably lays existing systems of law open to criticism , either as failing to accom plish their purposes or



as being directed at ends that are altogether m isguided. In assum ing th a t the institutions o f the D orian states have a purpose, the A thenian does not com m it him self to any particular theory of law. O n any view, the individual provisions o f a legal code may be expected to serve some purp o se — after all, the legislators presum ably thought th a t they were achieving som ething by their enactm ents. B ut a stronger conception o f law as purposive may be implicit in the C retan ’s reply (626a-c). H e claims that the lawgiver of Crete directed his regulations concerning both public and private life tow ards the goal of victory in war. T h e language im plies th a t victory is the unique end of all the C retan legislation, not solely o f those institutions about which the A thenian had inquired (a point b lu rred in Saunders’ translation). A few pages later the A thenian attacks piecemeal legislation and insists quite explicitly that all the legislator’s activities be directed to the one ultim ate goal (630d-632d). T his single goal is described in a variety of ways, b u t the A thenian clearly believes that it will involve m aking citizens com pletely v irtuous, th at this will bring about the well-being and happiness o f the citizens and th at it can thus be identified w ith the com m on good of all the citizens (631b, 697d, 715b, 875a-c, 923b, 925e). At the very end o f the Laws the dem and for single-m inded legislation provides the m ain basis for the activities of the nocturnal council (962b-c). This assertion th a t there is an unique goal at w hich all legislation should aim is capable of two varieties o f interpretation: (1) It may sim ply m ean that laws ought to aim at this goal. One can m aintain this while agreeing th at, in the world as it is, m any (perhaps most) laws do not have this aim . Such m easures will be bad laws b u t they will nonetheless be laws. (2) O n the alternative interpretation it is seen as p art o f the intrinsic nature of law th at it should seek a particular end. A m easure not directed to this end is not in the full sense law. M any of th e legal theorists know n as positivists would be happy to accept th a t law should be directed to a particular goal provided that this is understood in sense (1). B entham and A ustin, for exam ple, both believed that law depends on th e will of the sovereign b u t coupled this with the belief th at all laws ought to be directed to the greatest happiness of the greatest num ber. If the enactm ents of a wicked or misguided sovereign were not directed to this end, the resulting provisions would, on their view, be bad laws b u t nevertheless valid as laws. T he second view, that law by its very nature has to serve a particular end, m ust be denied by any legal positivist b u t is characteristic of



authors in the natural-law tradition. T hus A quinas argues that every law is directed to the com m on good and th at this involves m aking the citizens virtuous. A tyrannical law, therefore, is not properly a law, b u t ra th e r a perversion o f law {Summa Theologiae, la 2ae, 90, 2; 92, 1; 96, 4). P u shed to its logical conclusion, this m ight seem to imply th at a bad law (i.e. one not conducive to th e right end) cannot really be called law, in o th er words th a t there is no such thing as an unjust law (Cicero, De Legibus, II. v. 11-12; A ugustine, De Libero Arbitrio, 1.5; A quinas, Sum m a Theologiae, lae 2ae, 95, 2 & 4). W here does P lato stand on this question? In the M inos, 314e, Socrates is m ade to m aintain th a t there cannot be a bad law and thus to com m it him self to th e natural-law position. T h a t dialogue is probably spurious, b u t the same d octrine is affirm ed in the Hippias M ajor, 284a285b, w hich m ost scholars now accept as genuine. N ow here in the Law s is Plato so ex plicit, b u t th ere are a n um ber of passages w here the A thenian talks as th o u g h law m ust necessarily be good and in accordance w ith reason and as th o ugh the legislator m ust necessarily be rig h t (e.g. 6 4 4 c-d , 645a, 659d, 714a, 715d, 728a). Elsew here he speaks o f laws and legislation as correct or incorrect (e.g. 627d, 705e, 707b, 715b) or im plies th a t law has natural standards of rightness (e.g. 63 1 c-d , 690c, 853a). T h e fullest tre a tm e n t of this problem is betw een 712b and 715d. T h e re th e A thenian, w hile officially discussing w hat form of constitu tio n should be im posed on the new C retan city, seizes the o p p o rtu n ity to distinguish betw een genuine and p seu do-constitu­ tions. Crete and S parta have genuine constitutions because th eir laws aim at the com m on good o f the whole state. O ther so-called co n stitu tio n s are sim ply m echanism s by which one faction in the city im poses its will on th e rest. T h e section is in te rru p te d by a digression on th e age of C ronos w hich leads to the conclusion th at in obeying law we are obeying reason, th e im m ortal elem ent w ithin us (713c-714b). Som e people, it is tru e , say th a t th ere are as m any kinds o f law as there are form s o f constitu tio n . T hey believe th at laws ought to prom ote the in terests o f th e rulers and th a t justice is w hatever suits the stronger party (714b-c). T h e A thenian, how ever, denies that these are genuine constitutions. In his view laws are n o t correct unless they serve the good o f the w hole city (7 15a-b). H e goes on to contrast the fortunate states whose rulers are o b ed ien t to law w ith the w retched ones where law is trodden u n d er foot. I t is natural to in te rp re t this passage as asserting that a genuine law m ust seek the good of the state and th a t measures which do not seek the com m on good are not, strictly speaking, to be called laws. In fact the



Athenian is not quite as explicit as this. He does tell us th at c o rru p t states lack genuine constitutions b u t he does not say in so m any w ords that they have no laws — merely th at their laws are not correct (orthos). Other passages suggest the same kind of ambivalence. T h e A thenian never seems to doubt the validity o f existing legislation even though he recognizes that laws vary from city to city and sees it as his task to discover which of these laws is good and which bad ( 634c, 635a, 637d, 683b, 708c, 770e, 962b-c). T he same attitude is evident in his treatm ent o f the C retan and Spartan systems. These fall short of the ideal in some very im portant respects (631a, 634c, 705d) b u t the Athenian does not doubt their legality. T he inference is that a misguided measure may still be genuine law. This may seem to suggest th at the A thenian is hopelessly confused — that he is sim ultaneously trying to sit on both the positivist and natural-law sides of th e fence. But there is in fact an interpretation which makes very good sense of his position. In dialogues such as the Phaedo (97c ff.) and the.Republic (508b-509b) Plato argues th a t we cannot really understand anything until we know how it is related to the good. In other w ords we need to know what purpose it serves. This is obviously th e case w ith artefacts such as knives. We cannot know what a knife is w ithout knowing the purpose or function for w hich it is designed — in this case cutting. A knife may therefore be said to be defined by its function. It is characteristic of such item s th at the line between being a bad specim en of its kind and not being a specim en at all is blurred. For exam ple, we can say that a good knife cuts well and that a bad knife cuts badly. But w hat do we say if we come across a knife-like object th at is quite hopeless at cutting? D o we call it a very bad knife or say th a t it is not really a knife? T he answ er may not be clear. We can apply these ideas to our problem about th e object o f law. Plato obviously believes that it is part o f the nature of law that it should seek th e good of the whole com m unity. In other w ords law is defined by its function of prom oting the com m on good. It follows th a t we may be uncertain w hether to regard a very misguided provision as a bad law or as not a genuine law at all. T here will be a continuous gradation from good laws, at one end, to measures at the other end th at are so bad as not to m erit the title o f ‘law’. M easures that come betw een the two may deserve some respect while falling well short o f the ideal (see A quinas, Summa Theologiae, la 2ae, 96, 4). On this view the A thenian has no need to make it clear w hether misguided measures are or are not law. Law, on his conception of it, has its own intrinsic standards of rightness and w rongness. He can



th erefo re talk ab o u t particu lar laws as correct or incorrect. T h e nearer a m easure com es to fulfilling these standards, the m ore truly can it be called ‘law ’. But th ere is not a black and w hite division betw een g en u in e and spurious system s. E ven the m isguided codes of existing states have some claim to validity.


The role o f reason

T h e A thenian an d his com panions always appear confident that the good w hich is the aim of law can be objectively discerned and th at it can th erefore be known w hether a particular m easure is or is not good law. T h ey see them selves as discussing the natu ral correctness and e rro r o f law (627d) and believe th a t in doing so they are seeking w hat is ‘tru e ’ an d ‘b e st’ (634c). T h e au th o r o f the M inos, w hoever he was, accurately re fle a s this doctrine w hen he makes Socrates claim th at law is a good (i.e. true) opinion of the city and so m ust aim to be the discovery o f w hat is th e case (314d-315a). Law , in th e A th en ian ’s view, is an em bodim ent or expression of reason and an object o f knowledge. T h e judgem ents of reason, w hen em bodied in a decree of th e city, becom e law (644c-d; cf. 645a). R eason strives to becom e law (835e) and in obeying law we are obeying th e edicts o f reason (713e-714a). T h e assim ilation of law to reason is helped by an etym ological connection w hich Plato (wrongly) discerns betw een nomos (law) and nous (reason or intelligence) 713e-714a, 957c; cf. 836e. T h is view o f law as an object o f rational knowledge underlies the account o f the n octurnal council. T h e senior m em bers are the reason of th e state and th e ir young assistants its senses; b o th are needed to ensure its survival (961d). T h e ir m ain task is to discern the ends and m eans o f legislation (962d). In this they are com pared to experts such as navigators, generals and doctors who have knowledge o f th e end to be p u rsu ed and are able to apply this in practice (96 le-9 6 2 b , 963a-b). T h e A th en ian ’s theology su p p o rts his view o f the objectivity of law: ‘G o d holds th e beginning m iddle and end o f all things and justice follows him to p u n ish those who break his divine law’ (715e-716a); ‘G o d , not m an, is th e m easure o f all th in g s’ (716c). T he connection of these theological pronouncem ents w ith the rationality of law becomes clear in Book X. T h ere the A thenian argues at length th a t the order of th e universe shows it to be u n d er th e control of reason; we may therefo re be sure th a t there are suprem ely rational deities who love the ju st and hate th e u n ju st and th a t the natural order is so arranged th at



the wicked will autom atically suffer for their m isdeeds. T he idea th a t there are irregular m ovem ents in the heavens is treated as blasphemous because it contradicts this picture (820e-822c). T h u s law is the reflection w ithin the hum an sphere of the rational order th at governs the universe as a whole. In keeping w ith this there is a vehem ent attack on those who see law, together w ith m oral and religious beliefs, as products of hum an invention rath er than of n atu re (888e-890b). T he great heresy is the distinction, prevalent am ong Plato’s intellectual predecessors, betw een nomos (law or convention) and phusis (nature), w ith its im plication that nomos is relative to particular hum an societies and lacks any fundam ental im portance. The A thenian, by contrast, seeks to show th at genuine law is a p a rt of nature. Indeed one way of interpreting, not only Book X, b u t the whole of the Laws w ould be to see it as a system atic attack on the nomos-phusis distinction. Plato would replace this w ith a view that unites law and nature by showing that they are both products of the same divine reason and therefore involve the same kind of order. I t is the work of hum an reason to discern the tru th about law and it may be aided in this by observation of the orderly movem ents o f the heavens. D espite its aesthetic appeal th e parallel betw een the order created by­ law in hum an societies and the order apparent in the visible heavens does not solve any fundam ental problems in the theory of law. T he trouble is that the order of the heavens tells us nothing whatever about what we, as hum an beings, ought or ought n o t to do. T h ere is no sense in w hich we can literally em ulate th e m ovem ents of the heavenly bodies. So, even if we were to grant that the universe is u n d er the control o f divine reason, we still would not know w hat forms o f law ought to be adopted by hum an com m unities. Like other ethical rationalists, Plato can make good his position only if he can show that his moral recom m endations are in some way derivable from selfevident principles of reason. In practice he cannot do this. T h e measures he recom m ends rest largely on two assum ptions, (a) that whatever makes for the stability and security o f the state is good, and (b) th a t conventional standards of m orality are to be upheld. T he general tendency of his work is to give the dictates o f conventional morality the status of divine truths. Associated w ith this tendency to identify reason and convention is an unclarity about how political wisdom is to be attained. T he general position seems to be th at it is a kind o f scientific knowledge requiring the awareness of first principles and the ability to put them into practice. T he parallel is usually with some kind of trained expert (e.g. 709a-e, 961e-962b, 963a-b). But in practice the A thenian seems to



rely on th e accum ulated experience of m ankind rather than on any fo rm of specialist know ledge.T his is evident in his ready acceptance o f co nventional m oral beliefs, in his ap p aren t use o f existing codes oi law, in th e praise he gives to C rete and S p arta, and, above all, in the respect h e shows for old age. E ducation, for exam ple, is defined as ‘the process o f d raw ing and guid in g ch ild ren tow ards th at principle w hich is p ro n o u n c e d rig h t by th e law and confirm ed as tru ly right by the ex p erien ce o f th e oldest and m ost ju st’ (659d, trans. B ury), and there are m any o th er trib u te s to th e supposed w isdom of the old (e.g. 634de, 66 5 d -e, 715d-e). In th e constitution of the C retan city the im portant m agistrates m ust all be aged at least fifty (755a, 765d, 946a; cf. 951e, 961a). I f we take such passages at th e ir face value there are obvious p ro b lem s. T h ere is no reason w hy the judgm ents of the old should coincide w ith those o f rational insight. M oreover the old, though they have th e benefit of experience, are com m only supposed to be preju d iced and inflexible. T h u s th e assum ption th at they are the rep o sito ry o f tru th m ight be a recipe for sim ple conservatism . W e may how ever be able to make a b etter case for Plato on the hypothesis that he is w riting at a ‘pop u lar level’. T h e m em bers of the nocturnal council w ho will be the experts on legislation are to follow a program m e of philosophical study. P resum ably o u t of deference to Cleinias and M egillus the A thenian does n o t explain this in any detail. I t m ight be th e kind o f study laid dow n for th e philosopher kings of the Republic, th o u g h it is perhaps m ore probable th a t it w ould resem ble the kind of dialectic described in the Sophist (see below, chapter 12, section 5). T h e m em bers o f th e n octurnal council will be the experts on legislation. T h ey will also be q u ite old. I t m u st be these th a t the A thenian has in m ind w hen he refers, in passages like 659d, to the w isdom o f the old. Cleinias and M egillus and the popular audience th ey supposedly rep resen t could not be expected to u n d erstand the philosophical work o f the n o ctu rn al council, b u t they do respect the w isdom o f the old and they do adm ire the order of the universe. It w ould make sense therefore th at the A thenian should choose to em phasise these factors and play dow n m ore distinctively philosophical elem ents. T h is in terp retatio n w ould absolve the A thenian from the charge of confusing reason w ith convention, b u t it raises form idable problem s at an o th er level. We have to accept (a) th a t the practice of some form of Platonic dialectic can lead to an insight into the fundam ental principles o f law and m orality, and (b) th a t it is possible for at least some hum an legislators to link these insights w ith the practical business o f law



making. Plato him self denies that philosophical insight can be conveyed in a book (Letter V II 341d-e). T hus we are all to some extent in the position of Cleinias and M egillus. We are forced either to take on trust the possibility o f a genuinely philosophical legislation or to fall back on experience as the best available substitute.


The role o f the legislator

In the real world, systems o f law have to govern the daily lives of peoples whose circum stances vary enorm ously from place to place and time to tim e. So it is not enough for a legislator to be aware o f eternal moral tru th s, if there are any. H e needs to translate any such tru th s into detailed regulations appropriate to the particular circum stances of his own com m unity. Plato is well aware o f this — he em phasizes what we m ight call the ‘contingent factors’ in legislation and w hat he calls the ‘o pportunities given by G od and chance’ (709a-c). But he is by no means consistent in his handling o f such problem s. In the Statesman it is argued that laws, being general, cannot take account of individual circum stances. T he ideal would therefore be the unim peded ru le o f a wise man who could deal appropriately w ith each case. A state governed by law is a second-best alternative, needed only because wise enough rulers are not generally available (see above, chapter 2, section 3). T he Laws also concedes th at theoretically speaking, the absolute rule of a wise m an m ight be ideal: ‘know ledge is unsurpassed by any law or regulation; reason, if it is genuine and really enjoys its natural freedom , should have universal pow er’ (875c). T he difficulty is that, in practice,no hum an being is capable o f exercising untram m elled power: not only do m ost m en lack w isdom b u t there is also the danger th at their desires will get the b etter of their judgem ent (713c-d, 714a, 875c-d; cf. A ristotle, Politics, III . xv, 1386a 7-35). Thus reason m ust be em bodied in law. B ut, even then, the legislator should, if possible, leave the courts wide discretion to deal with individual cases. So, where good courts can be established, the legislator will content him self w ith setting out guidelines or examples of what is to be done (876c-d). D espite its superficial attractions, this view seems to rest on a m isunderstanding. T h e function o f law is not simply to rem edy the defects of hum an rulers b u t to bring benefits that could be secured in no other way. One of its chief functions is to enable us to co-ordinate our activities by m eans of rules: if I know th at m ost people will obey the law, I can, w ithin certain lim its, predict the behaviour of my fellow



m en and plan my ow n life accordingly. To plan my life in a rational w ay, I also need to know w hich kinds of act will be penalized by the a u th o rities. T h is is possible only w here there is a settled system of law. B u t it is not only by regulating p re-ex isten t form s o f activity th at law enriches o u r lives; it also makes new form s of activity possible. Only w here th ere is a law of prop erty can we buy and sell things; only where th ere is a law o f inheritance can we bequeath our property to others, an d so on. N one o f this w ould be possible if an absolute ruler settled all questions on a purely ad hoc basis. In practice Plato n o t only recognizes the need for a settled system of law b u t carries it to extrem es. H e expects the legislator to establish a detailed code th a t will rem ain valid for all tim e w ith little possibility of change (see below , c h a p te r 8). In p art this presum ably reflects the general G reek ad m iratio n for stability in law and P lato ’s own dislike of political change, b u t it m ust also be connected w ith his view o f law as a p ro d u c t o f reason. I f we em phasize the role of reason in providing know ledge of etern al tru th s an d regard hu m an laws as the co u n terp art am ong m en of th e divine o rd er th a t governs the universe, th en we will n o t expect an ad eq u ate legal system to be produced haphazardly by piecem eal legislation. I t m ust be the result o f a single unified vision, an d it therefo re req u ires a wise legislator. O nce he has done his work th ere will be no need for im provem ents or m odifications, because the p rinciples o f reason em bodied in th e law are as perm anent and u n changing as th e laws o f m athem atics or of celestial m otions. T hus P la to ’s belief in th e pow er o f reason leads him in directions quite opposite to one another. W hen he em phasizes the variability of hum an affairs he is led to th e ideal of a wise ru ler u n im p ed ed by law; w hen he em phasizes reaso n ’s function in the discovery o f eternal tru th s he expects the legal system to be equally perm anent and inflexible. A solution to this problem requires a m ore com plicated account of th e relation betw een law as established by hum an legislators and the etern al tru th s supposedly discovered by reason. T he point is well h an d led by A quinas who distinguishes the roles of natural and hum an law. N atu ral law, w hich rem ains the same at all tim es and places, lays dow n th e u ltim ate ends th at are to be p ursued, while it is the task of hu m an law to determ ine the m eans by w hich these ends are to be sought in particular circum stances. So hum an law may vary from place to place and change as conditions change (Sum m a Theologiae, la 2ae, 95.2; cf. A ristotle, Nicomachean Ethics, V. vii). I t follows from this that th ere are two ways in which hu m an law may be derived from the natu ral law. Som etim es it is deduced from the principles o f natural law in m u ch the sam e way th a t a conclusion in geom etry is deduced from



premises; for example, the precept ‘D o no m u rd e r’ is deduced from the more general natural-law principle ‘D o harm to no m an’. In other cases the hum an legislator is m ore like a creative artist: he has to produce som ething w hich will m eet the ends of law b u t has considerable freedom in deciding how to do this. In practice Plato also sees the lawgiver as a creator of institutions. His legislator is a craftsm an rath er than a scientist. He know s th at a good life can be secured for the citizens only by the cunning design of laws. But he seems curiously unconscious of this when he theorizes about legislation. In spite of a few passages where he com pares the work of the lawgiver with that of a craftsm an (803 a, 858a-b), he is more apt to present it as being like science or philosophy — a search after pre-existing truth. This is especially apparent in Book X w ith its insistence that law is a product o f nature rather than art (888e-890a, 890d).


Plato and natural law

It is not accidental that in this chapter a num ber of points o f contact have em erged betw een P lato’s theory of law and th at of Aquinas. Scholars have rightly argued that Plato is one o f the founders of the natural law tradition of which A quinas is the m ost im portant exponent (M aguire, 1947; Hall, 1956). Plato shares with the natural-law philosophers the belief that there are objective moral principles valid for all time and for all m en and capable of being discerned by reason; th at there is a fundam ental unity betw een these moral principles and the natural laws that govern the behaviour of the lower anim als and of inanim ate objects; and th a t th e object o f any genuine law is the com m on good w hich involves m aking men virtuous. But there are two qualifications to be added here. F irstly, Plato influenced the naturallaw trad itio n indirectly th ro u g h Aristotle rather than at first hand. Secondly, the natural lawyers also owed m uch to Stoicism and, of course, C hristianity. Among the ideas derived from these sources is the argum ent that the natural law is accessible to all men by the use of reason and that, since all thus share a com m on law, they are all fellow citizens (M arcus A urelius, 4.4; cf. Rom ans, 2.14). Plato apparently believes that all m en have reason (644d, 713e) but couples this with the view that only certain specially qualified people can discern the law that is ‘tru est’ and ‘best’. His theory thus lacks the hum anitarian bias of later natural law: he has little sym pathy for slaves and aliens. N atural-law theories have been used to influence practical politics



in th e direction of b o th reform and reaction. One who holds th a t there are objective stan d ard s of law may criticize and reject as invalid system s of positive law th a t do not m eet these standards. T h u s, in this c e n tu ry , th e N azi regim es indirectly caused a revival of natural-law theo ries because this way o f th o u g h t enabled jurists to say th at the u n ju st ‘laws’ o f these regim es were not genuine laws and so had no claim to be obeyed. At o th er tim es it has sim ply been assum ed that existing system s o f law em body the true natural law and are therefore b ey o n d m oral criticism . T h is is th e fault B entham detected in B lackstone. Plato shows traces of b o th kinds of argum ent. His belief in an objective law enables him to criticize the C retan and Spartan system s and, m ore virulently, the pseudo-constitutions of factionrid d e n states (714b-715d, 832c). B ut w hen he comes to propose legislation for th e new C retan city, he generally assum es th at existing system s o f law and m orality reflect the divine law of reason and thus raises them to status o f eternal truths. W hatever th e practical influence of natural-law theories, they all share a weakness in th eir theoretical foundations. We can all no doubt agree on certain very general prem ises; for exam ple, that good is to be p u rsu ed or th a t som e kind o f o rd er is essential if hum an beings are to live a decent kind o f life. T h e difficulty is in seeing how we can argue from these tru istic prem ises to substantive conclusions about what p articu lar kinds o f acts oug h t to be req uired or forbidden by law. Plato does not succeed in bridging this gap: now here does he make any really serious a tte m p t to show th a t the p articular provisions he recom m ends follow from his general conception o f law.


The Aims of Legislation Laws, 625c-632c, 688a-b, 693b-c, 697b-c, 705d-706a, 707d, 718c724b, 81 lc-812a, 829a, 857c-859b, 961c-964a; Republic, 426c-435a, 441c-445b; Gorgias 503d-505c We saw in th e last chapter that law, as Plato views it, m ust aim at the overall good o f the city. As it stands, this doctrine lacks any real content, since it gives no account o f the specific form s o f good to be sought by legislation. Plato does have plenty to say about this, though it is not clear w hether his different accounts of the specific aims of legislation are consistent w ith one another.

1 Critique o f the Dorian legislation The A thenian’s enquiry about the purposes of the m ost distinctive Cretan and Spartan institutions (625c) prom pts Cleinias to a general account of th e aims of the C retan laws. T h e legislator, he claim s, has arranged everything with a view to w arfare, and in doing so has implicitly criticized the generality of m ankind who do not realize that, in reality, they are always at war. W hat ordinarily passes for peace is in fact a state of undeclared conflict in w hich all cities are ranged against each other. It is right therefore to concentrate on military preparedness because a state defeated in w ar loses everything worth having (625c-626b). T h e A thenian replies to this by extending the idea th at we should continually be p repared to defend ourselves: he applies it, not only to relations betw een cities, b u t also to those betw een villages, households and individuals. H e then suggests that perhaps we can even be enemies of ourselves. T h e Cretan seizes eagerly on this idea: everyone is at war with everyone else and each m an is at war w ith himself. So the most im portant victory is the victory over self; in other words we all need self-control (626c-e). T he A thenian then gets Cleinias to agree th at a



city too can be vanquished by itself. T his would happen if the worse elem en ts in the population seized control over the better. So the p olitical eq uivalent o f th e virtue of self-control is a situation in which th e b est elem ents dom inate th e state (an obviously questionable analogy) (627a-c). At this point th e A thenian changes direction by a rg u in g th a t th e b est situation is n o t one in w hich th e b etter elem ents v a n q u ish and destroy th e worse, b u t one where the weaker elem ents w illingly su b m it to th e b etter. T h e ideal therefore is to secure peace a n d reconciliation and to create frien d ship by m eans of law. So victory in w ar, like m edical treatm en t, is good only as a m eans. T h e highest good and tru e aim o f legislation should be peace (627d-628d). A t one level b o th the A thenian and the C retan seem to be making g o o d points here. It is true th at defeat in war can render m ost other th in g s valueless and it is equally rig h t to direct attention to the dangers o f civil war and to th e value o f internal peace and friendship. W ho c o u ld deny these points? B ut it does not therefore follow th at either of th e se objects should be treated as th e sole end of legislation. T h ere are m an y different goods w hich can be realised in society and if the leg islator pursues any o f these to th e exclusion o f all others he is bound to sacrifice m uch th a t is valuable. O ne m ight therefore argue th at it is w ro n g to suppose th a t th ere could be a single goal of legislation. T he legislator m ust seek so far as possible to realise m any different goals. A n alternative w ould be to characterize the ultim ate goal in such a way th a t it can be seen to em brace all th e particular goals th a t are held v aluable. I t looks as th o u g h this is th e direction in w hich the A thenian w ishes to move. His objection to the supposed goal o f the C retan and S p a rta n system s is not th at it is w rong b u t th at it is too narrowly d efined. In 629a-630d th e A thenian uses contrasting quotations from the p oets T yrtaeu s and T heognis to point o u t the inadequacy o f the D o ria n ideal. T y rtaeu s, th e favourite poet o f the Spartans, praises the m an who shows courage in w ar, b u t Theognis values m ost the man w ho can be tru sted in civil strife. T his trustw orthiness, in the A th e n ia n ’s view, requires not only courage (andria) b u t also justice (,dikaiosune), self-control or tem perance (sophrosune) and wisdom (phronesis). So a good legislator will design all his laws to inculcate this to tal virtue. F rom this, the A thenian infers th at it was a mistake to suppose th at great legislators aim ed chiefly at im parting courage, for courage is only one p art o f virtue. O ne should rather say th at these legislators aim to im part v irtue or goodness as a whole. I t is im p o rtan t th erefo re not to legislate piecem eal, as is generally the custom , b u t system atically, taking virtue as th e starting point (630d-631a).



An im portant speech at 6 3 Ib -6 3 2 d expands this view of the legislator’s activity. Correct legislation secures the happiness o f those who live under it by securing for them both the hum an and the divine goods. T h e hum an goods are health, physical strength, beauty and wealth (if wisely used), b u t these depend on the divine goods w hich are first wisdom, secondly ‘a tem perate condition of the soul involving reason’, thirdly justice, which is a product of the other divine goods, and fourthly courage. So the hum an goods look to the divine and the latter to reason as their leader. T he A thenian then explains at some length how the legislator should supervise the whole of the citizen’s lives, from conception to death w ith a view to im parting these virtues. He lays particular em phasis on the need for the legislator to understand the passions and em otions. Finally, he suggests that guardians should be set over the laws. Some of these will be guided by wisdom and some by tru e opinion. In this way reason will bind the system together, dem onstrating that each elem ent is subordinate to tem perance and justice rather than to wealth and am bition. T he idea th a t virtue is the goal o f legislation rem ains the A thenian’s official position throughout the dialogue, and, in theory at least, shapes most of his practical proposals. I t is reasserted w hen he first tu rn s to consider the constitution o f the new state (705c-706d, 707d) and receives its strongest reaffirm ation in the final account of the nocturnal council. Just as a doctor m ust understand th e aim of his craft (health) and the general of his (victory) so the state m ust contain some elem ent that understands its object. This is the function o f the nocturnal council which m ust fix its gaze on the single object to w hich all institutions of the state should be directed, namely virtue. It will therefore need to understan d how virtue is both one and many, and be able to instruct the citizens in it (961e-964d).


Virtue and peace as objects o f legislation

The A thenian does not retract his suggestion (628c-d) that legislation should aim at peace and reconciliation. In his handling of the quotation from Theognis (630a-d) he implies that com plete virtue is to be sought because it makes people reliable in times of civil strife. T he obvious inference is that virtue is to be valued prim arily because it prom otes peace and security. Yet, w ithin a page or two, the A thenian announces that the virtues are divine goods whose possession is the only guarantee of happiness (631b-d). T h e im plication here is that



v irtu e is the h ig h est form of good — to be sought for itself alone. T hus th ere appears to be a serious am biguity. Is virtue itself the ultim ate goal or is it to be sought m erely as a means to peace? A sim ilar p ro b lem arises elsew here in the Law s, especially tow ards th e end of the historical excursus in book I I I . Tw ice in this section the A thenian reiterates th e doctrin e o f 631c-632d th at the end o f legislation is com plete virtue (6 8 8 a-b , 697b-c). T he dom inant them e in this p art o f th e dialogue is the need for the virtue of tem perance or self-control. Since the A thenian regards this as an essential p reco n d itio n o f th e o th e r v irtu es, his doctrine here is consistent with th e idea that legislation should seek to make the citizens virtuous. But he also m aintains th a t the aim of legislation is freedom , friendship and w isdom (693b-c; cf. 694b, 695d, 701d). L ater the ideal of friendship is p u sh ed to an extrem e. In th e tru ly best state, all citizens w ould share com m on feelings about everything — a situation w hich could come a b o u t only if pro p erty and fam ily relationships were abolished (739c; cf. 807b). At 693c th e A thenian apologizes for characterizing the ends o f legislation in so m any different ways, b u t claims th at it does not m a tte r w hether the goal is d escribed as self-control, friendship or w isdom , since these are all in reality the same. In these passages th e A thenian appears to be w orking w ith two basic conceptions o f th e end to be sought. A ccording to the first conception th e en d is v irtu e , w hich is in itself the greatest good of the soul. A ccording to the second conception the end of legislation is to make th e city as peaceful and unified as possible. T o be consistent, the A thenian m ust argue th a t these tw o conceptions of the end coincide — th a t in p ursuing v irtue we will prom ote unity and peace in the city and th a t in p ursuing this peace an d unity we inevitably prom ote virtue. He attem pts to show this at 828e-829a. [T he city,] like the individual m an m ust live well. N ecessary conditions for living well are not doing w rong to others and not being w ronged by them . O f these the first is not very difficult. B ut it is extrem ely difficult to acquire the power of not being w ronged. O ne cannot have this power to the full extent w ithout becom ing good to th e full extent. T he same goes for the city. I f it is good it will have a peaceful life, b u t if it is bad it will be subject to strife from w ithin and w ithout. T h e argum ent here seems to be th a t the virtuous individual and the virtu o u s city can som ehow escape suffering w rong at the hands of o thers. V irtue and peace th u s go together and can be seen as different



aspects of the same ultim ate goal. But this, on the face o f it, seems absurd. It is obvious to everyone that good m en are som etim es the victims of injustice. How, then, can the A thenian suggest th at virtue protects us from being wronged? So far as the relationships betw een individuals are concerned, the Athenian could defend him self by arguing that, w here the city as a whole is virtuous, each of the inhabitants will be good. N one of them will therefore be willing to w rong his fellow m en. T h u s those who live in the virtuous city will be certain not to suffer injustice. Each of them would, then, be protected, not so m uch by his own virtue, as by the virtue of his fellow citizens. Even if this argum ent is acceptable, the A thenian w ould still have to explain how the virtuous city is protected from w rongs done to it by non-vtrtuous foreigners. H is position on this point seems to be that virtue will make the city strong politically and m ilitarily. It will therefore be pro o f against external attack as well as from internal subversion. T his idea has some plausibility so long as one thinks of virtue as a m atter of diligently perform ing one’s allotted social role. The virtuous city would then be strong because each citizen would loyally and courageously do w hatever the wise rulers told him . M uch of the A thenian’s legislation does indeed seem designed to encourage this kind of conform ist attitude. (See especially 700a-701d, 729d, 793b-c, 942a-d.) ■ T he trouble is th a t we now seem com m itted to a quite different view of virtue from th a t espoused by the A thenian w hen he speaks of the virtues as ‘divine goods’ (631b-c). His position there is that the virtues, as divine goods, are the greatest possible blessing on those who possess them. If virtue was simply a m atter o f conform ity to social norm s, it would, no doubt, be a blessing to the com m unity b u t not necessarily to the virtuous m an himself. After all, it may not always be in his interests to do what society requires. T he A thenian is not very explicit about why he thinks virtue is always a blessing to its possessor, b u t, so far as one can see, he would adhere to the doctrine o fth e Republic th at virtue is good for us because it is an orderly state of the soul in w hich the passions are properly subordinated to the reason. (See chapter 5 below.) But there is no obvious reason why a person who is virtuous in this sense should necessarily be an obedient citizen, prepared to act as a reliable cog in the machinery of state. T here is a significant tension here betw een two different conceptions o f virtue. If virtue is a good state of th e individual, there is no obvious reason why the city with virutous citizens should be strong; if virtue means conform ity to social



n o rm s th ere is no obvious reason why the v irtuous m an should be happy. T h e difference betw een these tw o conceptions could be im portant. S o long as one concentrates on the idea th at the object of legislation is v irtu e one may conclude th a t th e A thenian’s fundam ental concern is w ith the happiness and m oral fulfilm ent of the citizens as individuals, a n d th a t the value of the individual soul has priority over th at of the s ta te (H all, 1981, p. 97). B ut this in d ividualist interp retatio n would be o u t of th e questio n if the v irtue th a t is to be inculcated consists sim ply in a disposition to do w hatever is req u ired o f one in the interests o f the s ta te . T h is p ro b lem can be resolved only if P lato can develop his acc o u n ts o f virtue and of the state in such a way as to show th at the two co n cep tio n s o f virtue could in practice coincide.


P la to ’s legal moralism

In so fa r as they treat virtue as the object o f legislation the A thenian and h is com panions are logically com m itted to a form of legal m oralism , i.e. to th e belief th a t the law should seek to im prove the moral character o f th e citizens. T h is view may be d isturbing to a m odern reader b ecau se m ost o f us have been b ro u g h t up, u n der the influence of lib eral thinkers like J.S . M ill, to regard any kind of legal m oralism with su sp icio n . T h e m ain objection raised by these thinkers is th a t if the sta te seeks to make people good it pre-em pts choices that should p ro p e rly be left to th e individual and th u s stifles individuality. These p o in ts w ould npt have w orried th e D orians b u t, by im plication at least, th e y oug h t to have w orried some o f P lato’s contem poraries. For exam p le Pericles, as reported in T hucydides I I , 35 ff., praises Athens fo r th e freedom it bestow s on th e individual citizens. T hey are free to c o n d u c t their private lives and to educate th eir children as they wish. P lato’s m oralism is, no d o u b t, largely due to the absence from Greek th o u g h t and language o f th e kind of d istinction th at we w ould draw b etw een law on th e one hand and m atters o f m orality or o f convention o n th e other. B ut it is also closely connected w ith some of the m ost fu n d a m e n ta l features of P lato ’s philosophy. One o f these features is th e Socratic d octrine, fam iliar to readers of dialogues like the Gorgias a n d the Republic, th at the individual can be happy if and only if he has a v irtu o u s soul. F rom this, together w ith the innocuous-looking prem ise th a t th e legislator should seek th e welfare of the citizen body as a



whole, there follows the conclusion that the legislator should, so far as possible, seek to make the citizens virtuous. So, although Socrates is sometimes represented as a protagonist of the liberal conscience, the doctrines that Plato derived from him lead directly to the doctrine of legal moralism that is anathem a to the liberal tradition. T he idea th at it is the task of th e legislator to make people virtuous implies that the legislator can know w hat form s of life are required by virtue. So the A thenian’s legal m oralism rests also on his objectivist view of m orality — his view of it as a m atter of knowledge. W hat is more, it involves a particularly strong form o f objectivism . T he Athenian believes that there is a reasonable prospect of finding a legislator w ith reliable knowledge of right and w rong and th a t this legislator will be able to lay dow n detailed rules of right living. H e will regulate every aspect of the citizen’s lives w ith a view to the ‘divine goods’ (631b-632c). Someone who wishes seriously to defend a doctrine of individual liberty will alm ost certainly reject this strong form of objectivism . He may argue th at moral values are not objects of know ledge b u t sim ply express the preferences of the individual or the group. (I t is likely that some of Plato’s contem poraries w ould have developed the distinction betw een nature and convention in the same way, though the evidence is not very clear.) An alternative W’ould be to replace Plato’s strong objectivism with a more limited form . One could then argue (a) that even if moral values are knowable in principle, no one can in fact be trusted to have an infallible knowledge of these m atters, and (b) that, while ultim ate values (such as peace, freedom and justice) are objectively discernible, they may be instantiated in many different ways so that there is room for each person to determ ine his own form of the good life. Even if we cannot accept P lato’s account o f moral knowledge, it does not, I think, follow that we m ust reject the whole of his moral and political philosophy out of hand. L iberal thinkers have argued their case not just by attacking the idea th at rulers can know what is right for their subjects b u t also by depicting the attractions of societies that allow a high degree of individual freedom (m uch as Pericles praises Athens in the funeral speech). Sim ilarly it is open to Plato to paint the attractions of a society com m itted to the kind of moral order th at he admires. He can try to show that only in such a society can hum an beings satisfy their most fundam ental needs. One can read the Laws as an attem pt to do this, to make the reader feel the attractions of a life com m itted to Plato’s ideal of law. Even if we do not share Plato’s



epistem ological prem ises a reading o f the Laws can therefore provoke u s into re-exam ining some o f th e assum ptions that underlie our own political attitudes.


Lazo as education

T h e A thenian’s dem and th a t th e lawgiver should supervise all the activities of th e citizens, and should instruct them as to w hat is right an d w rong by using praise and blam e, honour and dishonour as well as b y penalizing m isbehaviour (631b-632d), implies that the legislator h as an educational role. T h is conception is worked o u t in two ways. F irstly , there are laws to govern the upbringing of children and to establish appropriate educational institutions. T he second, and more distinctive, idea is th at legislation itself m ust involve a kind of in stru c tio n , th a t it should seek to shape the character by persuasion as well as by coercion. T his distinctive view is developed through a medical analogy. T here are two kinds of doctor. T h e first kind are slaves them selves and mostly have slaves as patients. T hey ru sh from p atien t to patient and, on the basis o f experience alone, issue orders w ithout explanations. T he second kind are free doctors who treat free m en. T hey make careful enquiries about th e nature o f th e illness and explain to their patients w hat is happening. T hey give in structions only w hen they have secured the p atien t’s consent. T h u s they heal prim arily by persuasion (719e-720e). Sim ilarly th ere are tw o m ethods of legislation, the ‘single’ m ethod w hich sim ply lays dow n penalties for certain kinds of act, and the ‘d o u b le’ m ethod in w hich the prescription of the penalty is preceded by an attem p t to show why the acts in question are wrong and to persuade people not to do them . T h e ‘d o u b le’ m ethod is taken to be tw ice as valuable as th e single (720e-722c). T h e im plication of this is th a t the lawgiver m ust in stru c t and educate the citizens (857c-e). He will do this by m eans of pream bles which use persuasion rather than th reats to produce the req u ired behaviour (722c-723d). T h e A thenian follows the ‘d ouble’ m ethod b o th by setting out a general pream ble or prelude to the whole legal system (726a-734e) and by providing pream bles for individual laws. (In practice it is often difficult to distinguish the pream ble from (a) the general discussion which introduces a p articular section of legislation and (b) the law proper w hich lays dow n the penalty.) T he A thenian is so proud of his pream bles th a t he attrib u tes them to divine inspiration. H e also



suggests that the Laws itself should be the m ajor textbook used in the education of the young. Schoolm asters m ust be com pelled to learn it, and any similar works, by heart and to teach them to their pupils (81 lc-812a, 957c; cf. 858c-859c). The idea th at the lawgiver should use exhortation and instruction so that the citizens behave rightly out of a sense of moral d u ty , rather than through fear of penalties, sounds attractive. Surely, persuading people voluntarily to do what is required is preferable to coercing them by sanctions? But the A thenian’s proposals deserve to be treated with scepticism. T h e difficulty is not that they would be im practical — it is all too clear that governm ent propaganda can be very effective. Rather, the proposals are questionable on moral grounds. One point raised here concerns P lato’s concept of persuasion. In the Gorgias (454b-455d) he distinguishes the kind of persuasion effected by rhetoric, which results merely in belief, from the kind which produces genuine knowledge. One would like to think that the preambles to the laws exercise persuasion o f the latter kind (H all, 1981, pp. 93-6), b u t this is clearly not the case; they are exhortations rather than argum ents (see, e.g., 726a-734d, 741a-e, 772e-773e, 823d-824a) and it is their literary qualities in which the A thenian feels so proud (81 lc -e ). T his is inevitable, given (a) the view of virtue in the Laws which em phasizes the need for correct training of the em otions and (b) the belief th at only a lim ited num ber of the citizens can come to a fully rational understanding. In any case Plato never pretends that understanding can be conveyed in a few w ords. On the other hand, there is nothing in the Laws, apart from the suggestion that poets should say that justice and happiness coincide even if this were not the case (663d-664a), to support Popper’s allegation (1966, p. 270) that by ‘persuasion’ Plato means largely ‘lying propaganda’. T he preambles mostly have the character of rather conventional sermons. Although the preambles are in themselves p retty innocuous, the reliance which the A thenian would lay on them and on other means of persuasion could be very dangerous. T here are two points here: (1) As Plato recognizes, propaganda can extend to areas of life where penalties would be ineffective and th e social pressures it creates can be as pow erful as the fear of sanctions. So the kind of persuasion advocated by the Athenian could restrict the area w ithin which the individual is free to choose his owrn way o f life. (2) Persuasion is unlikely to be effectiv e'if people are free to propagate views opposed to the official line. So if a governm ent wishes to use persuasion as a major means of achieving social conform ity it cannot tolerate freedom o f speech. Thus



th e idea of law as education is inextricably linked to one of the features o f th e Lazos th a t a m o d ern reader is likely to find m ost disturbing: its failure to allow room for self-expression on the part o f the citizens.


Virtue Law s, 626b-636e, 644b-645c, 653a-c, 660e-664a. 688a-689e, 6 9 6 b 697c, 710a-b, 726a-734d, 782d-783a, 835c-836a, 963a-964a, 965c-e; Protagoras, 352a-360e; Gorgias, 466a-468e, 491c-508c; Republic, 4 2 7 d 444e; Timaeus, 41d-44d, 69c-70d, 86b-87b

The divine goods which are the chief object of legislation correspond to the four virtues listed in the Republic, 427e ff. O ther features o f the opening pages of the Laws will also seem familiar to those acquainted with earlier dialogues. For exam ple, th e idea th at the legislator should aim at virtue as a whole may recall the Socratic doctrine th a t virtue is one because it consists in knowing the good. Sim ilarly, the suggestion that laws will need guardians, some using know ledge and some true belief, may rem ind us of the Republic, where the philosopher rulers alone have knowledge, though the lower classes may have true belief. But, as well as these points o f sim ilarity, there are also features o f these early pages that are inconsistent w ith P lato’s earlier m oral philosophy, or at least suggest some confusion o f thought. C onsider, for exam ple, the following: (1) At 630b the A thenian suggests th at even m ercenary soldiers, notoriously reckless and b ru tal characters, can exhibit courage. This implies that courage is separable from the other virtues and that virtue cannot therefore be one. (2) T he same conclusion seems to follow from the A thenian’s proposal at 632d-e to consider different institutions each designed to inculcate a different part o f virtue. (3) T he Athenian gives the four virtues different places in an order of priority, with wisdom first and courage em phatically last. This would be nonsense if the virtues were all one and therefore identical with one another. Even w ithout the doctrine of the unity of virtue, this ranking of the particular virtues would require explanation. (4) Sim ilarly, the assertion that justice results from a m ingling of wisdom, tem perance and courage is both obscure in itself and inconsistent w ith the claim that virtue is one.



(5) W hen he first lists the four virtues (630a-b), the A thenian in clu d es sophrosune, tem p eran ce, b u t w hen he repeats the list at 631c h e refers instead to a ‘tem perate condition o f the soul involving re a so n ’. O ne w ould expect th e choice of this phrase to have some significance, b u t it is by no m eans clear what that significance could be. Som e scholars have concluded th at this whole section is radically in co h eren t, for, although P lato seems to be reaffirm ing some of the key d o c trin e s of his earlier m oral philosophy, m uch o f w hat he says looks p a te n tly inconsistent with those earlier views (G igon, 1954; cf. M üller, 1951). In this ch ap ter, I shall ad o p t a m ore positive approach and shall tr y to make sense o f these early pages in the light of the account of v irtu e as it develops th ro u g h the dialogue.


M oral psychology

A t 632e, after establishing th at virtue is the m ain object o f legislation, th e A thenian tu rn s to consider the D orian institutions designed to inculcate courage. As th e C retan and S partan u n d erstand it, courage consists in doing b attle w ith fears and pains, b u t they readily agree w ith the A thenian w hen he suggests th a t it is equally im p o rtan t to struggle against desires and pleasures. In d eed it is the m an who gives w ay to these who has really lost m astery over him self (632e-633e). T he A th en ian proposes th e in stitu tio n o f drinking parties as a means of train in g th e young to avoid this danger. Im p licit in this passage is a fairly sim ple kind o f m oral psychology. A m a n ’s ‘tru e ’ or ‘b e tte r’ self is distinguished from his pleasures and p ain s, his desires and fears (‘his passions’ one m ight say). V irtue is that state o f th e soul in w hich th e tru e self exercises its proper dom inion over these ‘low er’ elem ents. We have already m et this contrast b etw een a b e tte r self w hich deserves to rule and a worse self which sh o u ld obey (6 2 6 d -e, see above, ch ap ter 4, section 1) and sim ilar m otifs occur elsew here (e.g. 689a-c, 726-728a). A passage at 863a864c, where th e A thenian distinguishes three kinds of injustice, is p articu larly im p o rta n t in this respect. T h e first kind of injustice is due to th e d o m ination o f th e soul by anger, the second to its dom ination by pleasure and the th ird to ignorance. T h e A thenian points o u t th at we talk of people ‘overcom ing’ or ‘being overcom e by’ pleasure and anger b u t not o f th eir ‘overcom ing’ or ‘being overcom e’ by their ignorance (863d). G iven the A thenian’s conception of the hum an personality, the reaso n for this is p retty obvious. A m an ’s tru e self is identified with his cognitive capacities (i.e. th a t p art o f him w hich knows or believes).



Ignorance is a defect in this part of the personality, so the man who goes wrong through ignorance, unlike the man w ho goes wrong through anger or pleasure, has not been overw helm ed by som ething external to his ‘tru e ’ self. The suggestion (863b) that anger (and presum ably pleasure) might be regarded as distinct parts of the soul is an obvious reference to the doctrine that the hum an soul has distinct reasoning, spirited and appetitive parts. This doctrine will be familiar to m ost readers o f the Republic (435a-44lc) b u t the parallel w ith the T im aeu s may be more significant. In th at dialogue Tim aeus explains how h um an beings are each created with a small portion of divine ‘s o u l-s tu ff. T he natural movement of this divine soul is the same uniform circular m otion as is revealed by the heavens, and Tim aeus identifies this m ovem ent with the activity of reason. But when the soul is im planted in the body it becomes subject to sensations and passions, giving rise to all kinds of irrational m ovem ents. T he hum an being will have no rest until it brings these irrational forces into conform ity w ith the uniform morion of reason (42e-44d). L ater the passions are treated as form ing a ‘inortal’ soul attached to the im m ortal elem ent of reason. Tim aeus distinguishes a spirited and an appetitive p art w ithin this ‘m ortal’ or ‘lower’ p art o f the soul and assigns them to separate parts o f the body (69d-70d). His doctrine thus corresponds to that of th e R epu blic in that he treats the rational p art of the soul as the true self and believes that virtue consists in dom ination by this rational elem ent o f the lower parts. The L a w s follows the Tim aeus and th t R epublic in treating the reason as the only divine or im m ortal elem ent w ithin us (713e-714a). T his is why the virtues which involve following th e dictates of reason are divine goods, while goods o f the body are merely hum an (631b-c, 697b, 728d-e, 743e). Similarly, in the prelude to the L a w s (726a734d), the first part, in which the legislator praises virtues for their own sake, is said to be spoken ‘from the divine point o f view’, while the second part, in which he praises them as leading to a pleasant life, is spoken ‘from the hum an view point’ (732d-e) — pleasures and pains are, of course, p art of the ‘m ortal’ soul. In the T im aeus and the R epublic the separation o f reason from the lower elem ents of the soul is bound up w ith the distinction between know ledge and belief. T he function o f reason is to know the eternal Forms, so in obeying reason we are acting on our knowledge of the Forms. The-Lazy*, in keeping with its m ore practical focus, treats right belief as an acceptable alternative to know ledge (632c, 689a-e, 864a). It is assum ed in practice th at the virtuous m an will act, not on his own



in sig h t into the foundations o f good and evil, but in accordance w ith law , w hich rests on the insight and w isdom of the old (see above, c h a p te r 3). T h is is particularly evident in the prelude to the laws, w h ere the legislator begins by insisting th at the b etter elem ents w ithin us m ust rule over th e worse (726a) b u t then assum es that this is eq u iv alen t to saying th a t we m u st obey the law (728a). T h e same kind o f consideration may explain why th e w ord for ‘w isdom ’ in the lists of v irtu e s (6 3 0 a-b , 631c) was p h ro n esis, n o t sophia (the w ord used in the R e p u b lic ). I f A risto tle’s usage (N ico m a ch ea n E th ics , V I, v & vii) is any g u id e , phronesis refers rath er to the practical w isdom w hich expresses itse lf in the p ru d e n t conduct of one’s public and private business, w hereas soph ia w ould suggest th e theoretical w isdom o f the philosopher.


M o r a l respon sibility

T h e account of th e soul and its way of functioning im plicit in the L a w s raises problem s ab o u t responsibility. In so far as these problem s c o n cern the penal code they will be discussed in chapter 14, b u t they also have im p o rtan t im plications for the general account of virtue in th e L a w s. T h e difficulty is m ost a p p aren t in the ‘prelu d e’ to the legal code (726a-734d) w hich takes th e form o f an exhortation to ‘honour th e soul’. In this prelude th e legislator for the m ost part addresses the citizens as responsible beings — it is up to them to be virtuous and they m u st not evade th e ir resp onsibility by blam ing others (727b). But he also suggests th a t, since no one is w illingly unjust, crim inals should be reg ard ed w ith pity rather th an anger, the im plication here apparently b ein g th a t they are n o t really responsible for their m isdeeds. So the legislator appears to be adop tin g in consistent attitudes to the question o f individual responsibility. M ore generally one m ight argue th a t the em phasis th ro u g h o u t the L a w s on th e im portance o f education and law in m aking the citizens virtuous precludes any attem p t to regard people as responsible for th eir ow n shortcom ings. B ut later, when he discusses the divine g overnm ent of th e universe,.the A thenian insists q u ite em phatically that o u r state of v irtue or vice is the product o f our ow n will (904c-d). T h e re is a parallel p ro b lem in th e T im aeu s. In th at dialogue the d em iu rg e or creator explicitly suggests th at hum an beings can be a cause o f evil to them selves, i.e. they are responsible for at least some of th e ir m istakes (42e), b u t it is also said th a t we can bring the disorderly m ov em en ts u n d er the control o f reason only if we have the right



education and upbringing (44b-c). At 86b T im aeus gives a purely physiological account of certain ‘diseases of the soul th a t are due to the disposition of the body’. Among these are alm ost all the cases commonly called ‘incontinence with regard to pleasures’. N o one voluntarily becom es bad — one becomes bad th ro u g h a disorder o f the body or bad upbringing, b o th of which are hateful to everyone and involuntary (86d-e). A few lines later Tim aeus sum s up his position. Those of us who are bad become so ‘through two causes that are quite against our wills’: through bad conditions of the soul w hich in tu rn result from bodily disorders or th ro u g h being b ro u g h t up in badly organized cities. T h e blam e for these things m ust, he suggests, fall on parents and those who bring up children rather than on the children themselves. But T im aeus still seems to hold th at it is up to us to look after th e condition o f o u r own souls .‘a man m ust try in every way possible, by education, by the activities he pursues and by study to escape from evil and seize hold o f its opposite’ (S7a-b). On a superficial reading, it may look as though T im aeus is here represented as a determ inist, that, in his view, all h u m an acts are the result of physiology or environm ent and no one may therefore be held responsible for his m isdeeds. But this conflicts not only w ith the claim at 42e that we may be the cause of evil to ourselves b u t also w ith the willingness Tim aeus displays in this very passage to blam e parents and educators and w ith his belief th at we ourselves should try to escape from wickedness. So if Tim aeus was a determ inist he w ould be a very inconsistent one. T h e solution to the problem lies in the recognition that Tim aeus does not suggest that all hum an acts are causally determ ined in all respects. His position seems to be th a t one’s virtue or vice is th e product of two factors (a) the strength o f the disorderly movements of the lower soul and (b) the degree of effort the rational soul puts into controlling them . No d o u b t there are cases, as in children and those who have been very badly b rought up, where the disorderly m ovem ents are so strong that no one could control them . In more norm al cases they are controllable providing that we try hard enough. But even if our vice is due to a failure of effort on the p art of the rational soul it is still not voluntary. T h e rational soul never w an ts vice even though it may be so feeble that it puts up very little stuggle. The same solution can be applied to the problem o f responsibility in the L aw s. No one ever really wants to be vicious b u t we may become so if we are badly brought up or if we make insufficient effort to struggle against vice. In many cities the education given to the young may be so bad th a t it is practically impossible for people to becom e virtuous. T he city of the L a w s on the other hand will be well constituted with a



p ro p er system of education. One may therefore assume that any vice displayed by the citizens is largely their own fault. S om ething like this position has to be adopted by anyone who wishes to m aintain a belief in individual responsibility while acknow ledging the im portance of environm ental causes: one has to argue that, w hile such causes may always influence action, only in u n u su al circum stances is this influence so pow erful as to determ ine w hat the agent does (i.e. to give him no chance o f behaving rightly).


S e lf-co n tro l a n d the p o ssib ility o /ak rasia‘

I f virtue requires th at o u r passions be guided by the judgem ents of reason as en sh rin ed in law it follows th at three kinds o f vice are possible. T h e first w ould occur where our judgem ent is m isguided as a resu lt o f intellectual erro r so th at we honestly hold false opinions about w hat we oug h t to do. T h e A thenian takes this possibility seriously in his attack on atheism in Book X. T h e second kind o f vice would be exhib ited by a m an who struggles unsuccessfully to overcome tem p tatio n . H ere th e judgem ent is correct b u t the passions push us too strongly in th e opposite direction. T his is w hat we call ‘weakness of w ill’ and w hat A ristotle calls a k ra sia . T h e th ird kind of vice is that displayed by a m an whose passions are so dom inant th at they distort his judgem ent. His desire determ ines his conception of w hat is good an d right. T h is is probably w hat A ristotle would call akolasia (N icom ach ean E th ics, 1152a 4 -6 ). C orresponding to the last two forms o f vice are tw o form s of v irtue. O ne may behave rightly eith er because one has w rongful desires b u t overcom es them or because one’s desires have been bro u g h t into th e harm ony w ith the judgem ents of reason. Broadly speaking, Book I of th e L a w s concentrates on the first kind of v irtu e, th at w hich consists in overcom ing desires, while Book II concen trates on th e second form , the harm onization of desire with rig h t judgem ent. As we have already seen, a m ain them e of Book I is the need to provide institutions w hich will train the young in resisting pleasures an d desires as well as pains and fears. T his point is developed at great length. Som etim es the speakers use m ilitary m etaphors: we m ust fight against pleasure and pain; v irtue consists in being victorious over th e m , b u t to be defeated is vice (633c-634b, 647c-d). In another passage the h um an being is rep resen ted as a p u p p et pulled in different directions by different strings; his task is to co-operate w ith the ‘golden strin g ’ of reason or law while resisting pleasure and pain (644d-645c).



D rink makes us like bad men because it reduces our self-control (645e646a), b u t drinking parties held under the right conditions are to be recom m ended because they can heip train people to resist pleasures (646d-650b). If people need training to help them exercise self-control it follows that the state of being unable to control o n e’s passions m ust be relatively common. In other words people really m ust be subject to the condition called a k ra sia , weakness of will. This is m ade clear in Book III where the A thenian describes what he calls the ‘greatest folly’ or ‘greatest stupidity’: I t is what we see in a m an who hates, rath er than loves, w hat he judges to be noble and good, while he loves and welcomes what seems unjust. T his disharm ony betw een o n e’s feelings of pain and pleasure is, I m aintain, the extrem e of stu p id ity . . . . W hen the soul opposes form s of knowledge, opinion or reason that are naturally fitted to rule I call it folly; likewise w hen in a state the masses disobey the m agistrates or laws and in an individual when rational judgem ents in the soul achieve nothing, or even worse than nothing. (689a-b) The word am a th ia , which I have rendered ‘stu p id ity ’ could also bear the translation ‘ignorance’. But the point is not that those referred to in. the passage lack knowledge: they know w hat is right, b u t, because their passions are not in harm ony with their reason they fail to do it. It is fair, therefore, to take this passage as a description of a k ra sia . This acceptance of ak ra sia may come as a surprise. Socrates notoriously had denied the possibility.of a k ra sia . H e had also argued that no one willingly does wrong. These two claims are so closely associated with one another that they are often treated as equivalents. Plato in the L a w s still claims that no one willingly does wrong. How then can he allow the possibility of a k ra sia ? One m ight seek to explain this by the hypothesis that m uch of the argum ent of the L a w s is directed to an unsophisticated audience. The idea of akrasia was apparently so well entrenched in ordinary thought that one can see why the A thenian would not wish to perplex the two D orians by denying it. But this suggestion may not be needed. There are independent grounds for believing th at Plato at this stage in his career would him self have accepted the possibility o f akrasia. To un derstand how he could do this one has to see th a t the doctrines (a) that akrasia is impossible and (b) that no one willingly does wrong, although closely associated, are not in fact equivalent. T his is clear if



we consider th e argum ent by w hich Socrates, in the P ro ta g o ra s 352a360e, attem p ts to show th at no one willingly does wrong. T he denial of a k r a s ia play an essential p a rt in th at arg u m en t b u t only because S o crates apparently assum es th a t a w rongful act done as a result of a k r a s ia w ould be voluntary. T h u s in o rd er to m aintain that no one w illingly does w rong he has to deny the reality o f ak ra sia . B ut this is n o t th e only way in which one could su p p o rt the claim th at all w ro n g d o in g is involuntary. An alternative strategy w ould be to argue th a t acts done through a k ra sia are n o t v oluntary in the full sense. A g en u in ely voluntary act, one m ight say, m ust accord with the wishes of th e rational soul. In a k ra sia th e rational soul is overw helm ed by the lo w er elem ents. So acts done th ro u g h weakness of will are not g e n u in e ly voluntary. T h u s we becom e u n ju st against our will. T his is th e position adopted by T im aeus at 86e w here he argues th at acts a ttrib u te d to incontinence ( a k ra s ia ) are really involuntary and by the A th e n ia n in L a w s 734b w here he m aintains th a t intem perance or self in d u lg en ce (a k o la sia ) is involuntary because it results either from ig n o ran ce or ak ra sia . T h ese passages can be seen as th e conclusion of a developm ent w h ic h may have been going on th ro u g h m ost o f Plato’s philosophical c a re e r. Socrates, as rep resen ted in the P ro ta g o r a s , plays dow n the im p o rta n c e of the passions and insists th at virtue is knowledge. T he G o rg ia s, R epu blic and P h aedru s have m uch more to say about the ‘lo w e r’ soul. T he R epu blic even describes w hat seems to be a state of a k r a s ia (439e-440a). It is n o t altogether surprising therefore that the T im a eu s and the L a w s should ap parently accept the possibility of a k r a s ia and should hold the correct train in g o f the lower soul to be as im p o rta n t for virtue as is rig h t judgm ent. B ut so long as Plato treats p leasu res and pains as separate from th e tru e soul he can continue to tr e a t vice, a state in w hich th e tru e soul is overw helm ed by external forces, as involuntary.


T h e harm ony o f passion a n d ju d g em en t

T h o s e w ho do w rong th ro u g h a k ra sia have correct beliefs about w hat is rig h t b u t are overcom e by pleasures or pains. B ut, of course, not all w ro n g d o in g is of this kind. W e often indulge our desires w ithout seriously acknowledging th a t we ought to do otherw ise. T his could be d u e to intellectual error b u t it is m ore com m only the result of allowing o n e ’s desires to d isto rt o n e’s b etter judgem ent. O ur desire to do so m e th in g stops us atten d in g to th e m ore objectionable features o f the



proposed act. Or we manage to convince ourselves that w hat we desire to do is, after all, right. Conversely, we may behave rightly because we actually want to do what is m orally required. V irtue, th en , may consist in having desires that are habituated to accord w ith o n e ’s better judgement. Plato’s recognition of this point is clear from the beginning o f Book II. The A thenian argues there that virtue and vice first en ter a child’s soul through the feelings o f pleasure and pain. These feelings m ust be correctly trained in childhood, though only w hen the child reaches the age of reason will he be able to see for him self th at the dispositions thus acquired harm onize with the judgem ents o f his understanding. This harm ony is virtue (653b-c). A few pages later the A thenian m aintains that, since education aims to lead the children to accept the principles of law as confirm ed by the wisdom o f the old, the im p o rtan t th in g is to accustom the children to having appropriate feelings of pleasure and pain (659d-e). T h e A thenian attaches great im portance to m usic and dance as devices for reinforcing this educational process. It is therefore more im portant that the perform er should have the right opinions and feelings that agree with them , than that he should have technical expertise (654b-d). T h e criterion of excellence in art is not pleasure as such b u t the pleasure felt by the good (658e-659c}. In general the aim of education is not merely to encourage us to control our wayward passions, b u t to make us feel pleasure in virtue and find vice painful. O f course it may not in practice be possible to draw a clear line between those who are virtuous through the exercise o f self-control and those who have correctly trained passions. W e may begin by having to struggle against our desires, inhibiting them through shame or fear of punishm ent (646e-647c), b u t come eventually to love the good life for its own sake. Conversely our first steps in vice may be due to weakness o f will, b u t as we go on we may lose our awareness that what we are doing is w rong. T here is a gentle slide from weakness into wickedness. T his is presum ably why th e A thenian says (734b) that licentiousness, a ko la sia , may result from ak ra sia . H e m ay have a similar point in m ind when he describes those who, through a discrepancy betw een their natural character and th eir upbringing, enjoy m usic which they really know to be bad. They are like men who fall into bad com pany and find a certain pleasure in their evil acquaintances. T hey may rebuke their com panions b u t they do not really mean w hat they say. T hey have only a dream like awareness of their own vice and, even if they are asham ed actually to praise those whose com pany they enjoy, they inevitably becom e like them (656a-b).



Self-indulgence can easily becom e self-deception. T his seems to be an underlying them e o f the prelude to th e L a w s in Book V, 727a-734e. T h e legislator exhorts the citizens to h o n o u r th eir souls, b u t one o f his main points is th a t we deceive ourselves w hen we th ink th at we honour o u r soul by indulging it o r by trying to evade responsibility for our m isdeeds (727a-b). A m ain cause o f vice is self-love w hich, like all form s of love, blinds us to th e tru e character o f its object. So self-love disto rts o u r perception of th e just, the good and the noble, makes us take o u r ow n folly for w isdom and leads us to th in k we know everything, w hen, in fact, we know practically nothing (731e-732a). T h e language here recalls the Socratic idea th a t the greatest evil is to think one has know ledge w hen one is, in reality, ignorant. N o doubt the rem edy for ‘double ignorance’, as it is called in Book IX , 863c, consists p artly in the acquisition of a sounder know ledge o f right and w rong. B ut since it is usually o u r own desire th at blinds us to our ignorance, the correct h ab itu atio n o f the lower soul m ust also play its part. So the L a w s ’ recipe for virtue com bines the right instruction of th e judgem ent w ith th e correct training of the em otions.


Sophrosune: the m ain sprin g o f the Laws?

A ccording to Barker (1960, p. 343) self-control is the ‘p rin cip le’ or ‘m ain sp rin g ’ o f th e L a w s . In one way this could be m isleading. ‘Selfc o n tro l’ may suggest th e k in d of virtue th a t consists in fighting against and overcom ing th e passions, w hereas Plato is m ore concerned to p roduce a harm ony betw een the passions and the judgem ents of reason. B u t, in fact, B arker seems to be using ‘self-control’ as the equivalent o f G reek virtue o f sophrosune. T his m eant m uch more to the G reeks th an ‘self-control’ does to us. In A thens it was recognized as the characteristic virtue of th e citizen in tim e of peace, expressing the ‘all-em bracing order and m orality of restraint and lim itation’ required by th e city-state (N o rth , 1966, pp. 150-2). It could also describe the rath er unim aginative self-discipline o f the Spartans (N o rth , 1966, pp. 102-4; cf. T hu cy d id es, I, 68-9). In the C h arm ides it is successively defined as quietness (159b), respect (160e), doing o n e ’s own work (161b), self-know ledge (164d). All these definitions have a basis in popular thought, though in th e C h arm ides they are rejected in favour of th e Socratic claim th at sophrosune is know ledge o f good and evil. A nother popular idea considered by Plato is th at it consists in heeding th e m axim ‘nothing in excess’, i.e. th at sophrosune is m oderation (P hilebu s 45d-e). A ccording to the G orgias (4 9 Id) the m any regard



sophrosune as a m atter of controlling the appetites b u t in the same

dialogue Socrates describes it as a state of order in the soul (506c507a). T h e R epublic cites the popular view of sophrosune as obedience to rulers and the control o f one’s appetities (389d-e) and then subsumes this definition in an account o f it as a harm ony of the different elements in the state (430d) and in the soul (442c-d). The word sophrosune itself occurs infrequently in the Laras. W hen its meaning is explicit it generally refers to the capacity to restrain one’s appetite, the popular virtue which the A thenian says is valueless in itself (696b-e). In these contexts ‘self-control’ is an adequate translation, b u t they do not justify the claim that self-control is the mainspring of the L a w s. W hat does deserve this description is the wider ideal o f sophrosune which em braces, not only self-control, b u t also order, harm ony, m oderation and self-knowledge. T hese ideas dominate m uch of the dialogue: (1) T h e idea th a t the aim o f legislation should be the victory of the better elem ents in the state and in the soul is considered, b u t replaced by the ideal of harm ony through law (626b-628b). (2) T h e institution of drinking parties is advocated on the grounds that they develop and test the young m an’s capacity to resist pleasures (633d-634c, 636e-650b, 671a-674b). (3) T h e long digression on education and the arts in Book I I , 652670e, develops the conception of virtue as a harm ony of reason and emotion. (4) T he failures of Argos and M essene are a ttrib u te d to a discord between rational judgem ent and emotion. T h e suggested rem edy is the acknowledgment of due measure or proportion in the distribution of power (686a-687e, 691b-d). (5) T he successes of the Persian m onarchy and of the Athenian democracy are attrib u ted to their m aintenance o f a m ean between freedom and despotism . T heir failures are attrib u ted to the abandonm ent of this m oderation (693d-701e). (6) T he prelude to the legal code begins w ithin an attack on those who think they honour the soul when they are in fact yielding to their desires (726a-728c). (7) A few pages later self-love is said to blind us to what is really good and make us take our wisdom for folly (731d-732b). (8) T h e prelude ends w ith the praise of the tem perate life with moderate pleasures and m oderate pains (732b-734d). Between them these passages touch explicitly or im plicitly on almost all aspects of sophrosune in the w idest sense. F undam ental to them all is the idea that virtue requires a right relation betw een reason



an d passion. L eft to them selves th e passions w ould lead us into all m an n er o f excess and wickedness. We possess sophrosune if reason governs our passions and keeps us in the way of m oderation. U n d e rsto o d in this wide sense, sophrosune is the raison d ’etre of the social an d educational in stitutions d escribed in the later books. It is as fu n d am en tal to the L a w s as justice, dikaiosu n e, is to the R epublic.


T he u n ity o f virtu e

I n Book X II the key task assigned to th e nocturnal council is th at of d iscern in g th e one end o f all legislation, and for this they m ust u n d e rs ta n d how v irtu e is b o th one and m any. It is easy enough, the A thenian there suggests, to see why we give particular virtues different n am es as though they were d istinct from one another. F or example, c o u rag e, w hich is concerned w ith fears, develops naturally in children a n d anim als w ith o u t any elem ent o f reason. But obviously there cannot be w isdom w ith o u t reason (963e). T h e difficult task, w hich the A th en ian leaves unexplained, is to see in w hat respects these virtues are one. T h is passage im m ediately explains some o f th e peculiarities noticed a t th e b eginning o f this chapter. T h e courage th at springs up naturally in ch ild ren an d anim als is presum ably the same as that evinced by the m ercenaries o f 630b. I t is sim ply th e capacity to resist im m ediate fears fo r th e sake of som e longer-term goal. One can see how the D orian in stitu tio n s m ight produce this form of courage in isolation from other v irtu e s. T h e re are o th er passages th at seem to refer to a corresponding fo rm o f tem pérance (696b-e, 71 Oa-b). B ut these unreasoned capacities c a n n o t be w hat th e A thenian generally understands by virtue, for v irtu e in his view requires that th e passions be subordinate to right ju d g em en t. O ne cannot therefore be properly virtuous w ithout know ledge or tru e belief about th e good. T his is presum ably why the divine goods are said at 6 3 Id to look to reason as their leader and why th e A thenian speaks in th e same passage o f ‘a tem perate condition of th e soul involving reason’. Sim ilarly at 6 96b-c he says that tem perance is valueless in itself b u t th a t no oth er v irtue is w orthw hile w ithout it. T h e point m ust be th a t both wisdom and the capacity to control one’s passions are necessary to virtue. Self-control w ithout w isdom could be u se d for th e w orst ends, b u t w isdom w ould be equally valueless if one lacked th e self-control to p u t it into practice. T h e re is no explicit reference in the L a w s to the kind of virtue which, according to th e R epu blic, can be achieved only by the philosopher who



apprehends the Form of the good. O n the other hand the kind of virtue described in the L a w s does bear a close resem blance to the kind of virtue that m ight be expected of the soldier or auxiliary class in the Republic. T his virtue consists in the capacity to m aintain, th ro u g h all tem ptations, the principles w hich one has been taught in youth (412c414a). It does not require knowledge, as such, b u t it does presuppose true belief which is acquired through right education. It is possible that Plato w ithholds m ention o f the ‘philosophical’ form of virtue because he judges it to be. out of place in a practically oriented dialogue like the L a w s. We shall see in chapter 12 th at there are similarities betw een the courses of study prescribed for the nocturnal council in the L a w s and the philosop h er-guardians in the Republic. We m ight take this as a h in t th at the R ep u b lic’s account of virtue is being presupposed w ithout being stated. On the other hand, we have seen in this chapter a num ber of reasons for supposing that Plato m ight have changed his view. His various treatm ents of the Socratic paradoxes do seem to show an increasing d o u b t about the power o f the rational soul to control its irrational elem ents. T he underlying reasons for these doubts come to the surface in the moral psychology o f the T im aeus, w ith w hich the L aw s seems to be in agreem ent. A ccording to this psychology, we have w ithin us at least two elements, one divine and rational, the other mortal and irrational. Since the m ortal elem ent is by nature intractable, the rational elem ent can never guarantee to control it. T here cannot therefore be a form of knowledge that guarantees virtuous behaviour. In the Socratic dialogues, virtue is held to be a unity because all virtues ultim ately consist in a knowledge o f what is good. In the last few pages of the L a w s, th e A thenian is made to insist th at virtue is both one and many. It may therefore seem th at the doctrine of virtue as knowledge is still lurking in the background. However the account of virtue just given suggests an o th er way in which the thesis th at virtue is both one and m any m ight be supported. It might be one, not because it is identified w ith some kind o f knowledge, but because it always involves the proper subordination of the passions to right judgem ent. The names given to the particular virtues draw attention to the different aspects or com ponents of this harm ony. ‘W isdom ’ points to the leading role of right judgem ent, ‘tem perance’ and 'courage’ to the need to control, respectively, pleasures and pains, ‘justice’ to the fact that this inner harm ony leads to right action. One may even be able to make some sense o f the ranking o f the virtues. W isdom comes first because rational judgem ent m ust be in control, tem perance comes second because the m ain danger is that we will succum b to pleasure.



T h e difficulty is to see why justice comes betw een tem perance and courage. If th e account just suggested is correct it should really come in to a different category since it is a result rather than a com ponent of th e soul’s harm ony. Perhaps th e A thenian has been led astray by his desire to em phasize th a t physical courage is of com paratively m inor im p o rtan ce. O f course this account is largely speculative b u t it is su fficien t to show th a t th e doctrin e o f the unity o f virtue need not be inconsistent w ith th e general account o f virtue in the L a w s. T h e charge o f radical inconsistency in the opening section of the dialogue cannot, therefore, be upheld. T h e tru th probably is th at Plato w ould not want u s to press the details of these passages. Unlike the Socrates o f the early dialogues who w ould doubtless have pounced on the assum ption that courage can be p u rsu ed in isolation from the other virtues, the A thenian treats his com panions’ views w ith a cautious respect. He leaves their p resuppositions unchallenged w hile allowing his own acco u n t o f v irtu e to reveal itself gradually as the dialogue progresses. O ne consequence of this tech n iq u e is th at one cannot treat the early p arts of th e dialogues as fully articulated expressions o f Plato’s considered doctrine. T hey have to be taken in the context of the dialogue as a whole.


Pleasure and the Good Life

L a w s , 636d-e, 644c-645c, 653a-c, 660d-664a, 732d-734e; P ro t a g o r a s , 351d-358d; Gorgia s, 492d-500d; P h a e d o , 68d-69c; R e pu blic, 580c-

588a; Philebus, especially 59c-67b

A central doctrine of the L a w s, as of m any other Platonic dialogues, is that it is in each individual’s own interests to be virtuous. In the R epublic , Plato had supported this thesis with three m ain argum ents: (a) that since virtue is a healthy and harm onious state o f the soul it is to be desired even more th an bodily health; (b) that the virtuous life is more truly pleasant than any other; (c) that after death the virtuous will be rew arded and th e wicked punished. All three argum ents are present, at least by im plication, in the L aw s. It shares w ith the R epublic the view of virtue as a harm onious state o f the soul u n d er the direction of reason. (See above, chapter 5.) T h e A thenian also m aintains that the wicked will fare badly after death by being sent to a lower plane of existence along w ith their fellow sinners (903b-907c; cf. 728b). But, in so far as the A thenian argues explicitly for v irtue, he pays most attention to the idea that it is pleasurable. This has puzzled many readers because it could be taken as a com m itm ent to the hedonist view that the good after which we should all strive is sim ply pleasure. P lato’s attitude to hedonism seems to have varied over the years. In the P ro ta g o ra s he apparently identifies pleasure with the good, b u t the G orgias and the P haedo are diametrically opposed to this. T h e R epublic, too, m aintains th at pleasure is n o t the good, even though it argues that in fact the good life will be more pleasurable than any other. T he Philebus , a dialogue that may be close in date to the L a w s , treats pleasure as a constituent o f the good life b u t argues that it is not the most im portant elem ent nor the one that makes the life truly good. T h e position in the L a w s is very obscure. In Book I, pleasure appears prim arily as som ething to be resisted and overcom e. In Book II, the A thenian stresses the need to educate the young, so that they feel pleasure in the right things, and argues that the just life m ust be



th e pleasantest. A passage in Book V has been taken to claim that we can and should do only w hat seems pleasantest to us. N one of these passages is clear in itself and jointly they can present an appearance of total confusion. O ne may well feel sym pathy w ith C rom bie’s com m ent (1962-3, vol. I, p. 271) th at Plato m anages ‘to sit on all the fences’.


B ook I: pleasu re a n d self-restra in t

T h e purpose o f th e d rinking parties w hich th e A thenian advocates in Book I is to tra in th e young to overcom e pleasure and desire (633d634c, 636a-650b, 671a-674c). O n the face o f it this suggests a strongly an ti-h e d o n ist doctrin e. O f course, a h ed onist can advocate a kind of self-control — he can advise us to resist im m ediate tem ptations in o rd er to m axim ize o u r pleasures in th e long te rm — b u t the A thenian’s sw eeping condem nations of pleasure (e.g. in 633d-634c) seem to apply to all alike. T hese passages have an alm ost ascetic ring. T w o passages in this book discuss at some length the role o f pleasure and pain in h um an choice: (1) T h e ‘tw o springs’ passage, 636d-e: here the A thenian m aintains th a t any discussion o f legislation m ust be alm ost entirely concerned w ith th e roles o f pleasure and pain. ‘T hese flow like two springs sent fo rth by nature; a city, a private individual or any other creatu re th a t draw s from th em th e right am ount, in the rig h t place at th e rig h t tim e, is happy; he th at draw s foolishly and at the wrong m om ent is m iserable.’ (2) T h e ‘p u p p e t o f th e gods’ passage, 644c-645c: according to this pictu resq u e account o f h u m an choice, th ere are in each of us two m utually antagonistic advisers, pleasure and pain. We also have opinions ab o u t th e future. T h e expectation o f pleasure is hope; the expectation o f pain is fear. In addition, there is w ithin each of us a pow er of reasoning or calculation th a t tells which am ong these pleasures, pains, hopes and fears is b e tte r and w hich worse. W hen em b o d ied in a decree of the city this reasoning is called ‘law ’. T h en , in one o f his m ore striking im ages, th e A thenian invites his listeners to look on m an as a p u p p et of th e gods. W e are each of us suspended, as it w ere, from th ree strings. T h e strings o f pleasure and pain are hard as iron, b u t the golden string o f reason and law is pliant and flexible; it needs help to overcom e the o th er forces. O u r task, then, is to co­ operate w ith this golden string in resisting th e others. I f one was determ ined at all costs to im pose a hedonist reading of the L a w s , one m ight conceivably be able to force the ‘two springs’ passage



into th at m ould. T he point w ould th en have to be that we need to choose our pleasures and pains carefully in order to get the greatest possible pleasure in the long run. But, on the face of it, this interpretation is highly im plausible. If a medical m an argued, as Greek medical m en often did, that the road to health lay in eating the right am ount at the right tim e and place, one would not conclude th at good health consisted in eating the greatest possible am ount in the long run. Rather one would suppose that there was some independent criterion of health which might require one to curtail the intake of food. Similarly the natural reading of th e ‘two springs’ passage is th at there is some independent standard of good in the light o f w hich one m ight need to restrict one’s pleasures. Pleasure itself cannot, therefore, be the good. T h e ‘p u p p et of the gods’ passage is more puzzling. T here are at least three points that m ight prom pt a hedonist in terpretation, (a) I f the expectation o f pleasure is hope and the expectation o f pain is fear, that suggests th at pleasure is the only good and pain the only evil, (b) T he image of the p u p p et m ight lead us to expect a determ inistic m odel in which all hum an behaviour is regarded as the product of pleasures and pains. It would follow that we have no choice but to seek pleasure and avoid pain, (c) T h e description of the th ird string as ‘logism os calculation, recalls the P ro ta g o ra s, where the role of reason is to calculate the balance of pleasures and pains. Against this we can set two other points, (a) T he p u ppet, oddly enough, has a certain freedom : it can choose w hich o f the th ree strings to co-operate with. If we take this seriously it suggests that the choosing self is distinct both from reason and from the passions, though Plato usually identifies the self w ith the reason. It is difficult to see how this m odel could be worked out coherently, b u t the im plication appears to be th a t we can act in ways that do not simply maximize our pleasures and th a t we sometimes o ught to do so. (b) T he A thenian says both that the golden string of calculation judges pleasures and pains, hopes and fears, deciding which of them is better and w hich worse, and th at, in co-operating with calculation we have to resist the pull of the other strings. T he im plication is surely that we judge which pleasures are to be pursued and which avoided in the light of some independent criterion of goodness and that this criterion is known to reason. Overall it is difficult to avoid the conclusion th at the A thenian has not thought through his account o f choice in these passages. It looks as though he is trying to com bine two possibly incom patible doctrines,



th a t pleasure and pain are the m ain determ inants of conduct, and that we ought to follow reason in resisting these feelings.


B ook II: pleasu re a n d h appiness

T h e m ain them e o f Book 11 is th a t th e young m u st be educated to take pleasure in the rig h t things. T h is, in itself, im plies a more positive evaluation of pleasure th an one m ight gather from Book I. P a rt o f th e A th en ian ’s educational doctrine is th at poetry m ust be co ntrolled. T h e leading them e o f p oetry m ust be th at the virtuous are happy. Those things norm ally held good (e.g. w ealth, fine physique, pow er, long life, etc.) are blessings to the good m an b u t are positively b a d for the u n ju st (660b-661d). T h e C retan and the Spartan find this h ard to accept: like som e o f S ocrates’ respondents in earlier dialogues, they can agree th a t the life o f th e u n ju st m an is dishonourable, b u t not th a t he is u n happy or th at his life is unpleasant or unprofitable to him (661d-662a, cf. C rito , 49b, G orgias, 474c). T h e A thenian replies by asserting his own com plete faith that justice, happiness and pleasure coincide. As a lawgiver, he would im pose severe penalties on anyone who said th at a bad m an could live a p leasan t life or th a t there is a difference betw een w hat is just and what is p ro fitab le. H e su p p o rts this view w ith a com plex argum ent designed to show th a t anyone who advocates justice, be he a god, a law giver or a p a re n t, m ust agree th a t the m ost just life is also the m ost pleasant (662c-663a). T h e argum ent is a form o f redu ctio a d absu rdu m : the A thenian claims th a t anyone who advocates justice while denying th at the just life is the pleasantest will be com m itted to the obviously false conclusion th at the h ap p iest life is n o t th e pleasantest. T h e key prem ise is at 662e. A ccording to the A thenian those who urge the young to live a just life claim to w ant the happiness o f th eir charges. T hey are thus com m itted to saying th at the just life is the happiest. T his means that if the just life was n o t th e pleasantest these people w ould have to agree th at the h ap p iest life was not the pleasantest. B ut the A thenian thinks th at this is ab su rd , for, he asks, ‘W hat good could come to us unaccom panied by pleasure (663a)?’ T h u s th e only sensible conclusion is that the just life is also the pleasantest. T h is argum ent rests on two questionable assum ptions. T h e first is th at anyone who advocates justice is com m itted to m aintaining th at the just life is happiest. T his is one o f P lato ’s standing beliefs, b u t it is certainly not self-evident. In the p resent context he supports it only by



showing that hum an legislators and parents urge the young to be just while claim ing to want their happiness. T he trouble w ith this argum ent is th at, at most, it proves only that people often think and talk as though justice and happiness coincide, not that they coincide in reality. T he second dubious assum ption is th a t the happiest life is pleasantest. T he A thenian does not su p p o rt this directly; instead he suggests in a series of rhetorical questions that the ju s t life m ust be pleasantest. ‘Is a good reputation am ong men and gods a fine thing but not pleasant?’ ‘Is it not pleasant to com m it no acts o f injustice and to have none done against oneself?’ But, even if we were to agree w ith the Athenian in answering ‘N o ’ to these questions, we would be com m itted only to saying that the just life is p le a sa n t, not th at it is the pleasantest. M oreover the argum ent relies on the supposed consequences of justice. Plato rightly rejected such argum ents in the R epublic, since, even if justice generally produces pleasant consequences, it need not always do so. T here is, regrettably, no guarantee th at the just man will escape injury by others or that he will get a good reputation. So the A thenian has proved neither th at the just life m ust be happy nor that it m ust be pleasantest. In the passage that follows (663b-664c), the A thenian com bines a disreputable argum ent with one that is suggestive and interesting, if not wholly convincing. T he disreputable argum ent is th at, w hether true or not, it will be useful to propagate the idea th at the just life is pleasantest, since no one would allow him self to be talked into an act unless it produces more pleasure than pain (663b). T he interesting argum ent is th at our perceptions may be clouded by our point of view. From the standpoint o f the u n ju st m an, u n ju st acts appear pleasant and just acts appear unpleasant. From th e just m an ’s standpoint the opposite is the case. But the good m an’s view is authoritative for truth. So it is the task of the legislator to correct these tricks of perspective. T here is an im portant point here. D ifferent people take pleasure in different things. Someone wholly com m itted to justice (i.e. who wanted above all to be just) would presum ably find his greatest pleasure in justice. So for those w ith just souls the just life will be pleasantest. B u t one cannot conclude from this th at the life o f the just will necessarily be pleasanter than that of the unjust. As Crom bie (1962, p. 271) points out, it could be simply that |ustice is pleasant to those who have a taste for that kind of thing. T he really unjust man could well find his greatest pleasure in behaving unjustly. T here is, moreover, no reason why the just m an’s verdict should be regarded as peculiarly authoritative on this issue — that w ould prejudge the



q uestion. So, although th e A thenian’s argum ents go some way towards show ing th at th e just life can be p le a s a n t , they certainly do not prove th a t it is th e pleasantest kind o f life. Leaving aside th e q uestion w hether the A thenian has succeeded in pro v in g th at the just life is pleasantest, we may com plain th at he has still not taken u p a clear position on the fundam ental issues. His insistence th at justice, happiness and pleasure m ust all coincide may suggest that pleasure is th e ultim ate good and th e claim at 663b th at no one w ould w illingly do an act th at did not result in a balance of pleasure o v er pain m ight indicate a com m itm ent to the psychological hedonist p o sitio n th at hu m an beings can only seek pleasure. But his point could still be th a t, alth o u g h th e good life always happens to be pleasant, pleasure is not itself th e good. 3

P leasu re a n d the choice o f lives

Before we pass to Book V, it m ay be helpful to distinguish two conceptions of m oral choice. T h e first conception treats m orality as prim arily concerned with the choice o f particu lar acts on particular occasions. T he kind o f question we have to answ er is ‘W hat shall I do here and now ?’ or ‘W hat shall I do next T uesday evening?’ Since m ost o f us know which kinds of th in g s give us pleasure and w hich do not, it may make good sense to answ er a question o f this kind by saying, ‘D o w hat will give you m ost pleasu re.’ H ere th e hedonist m odel o f choice has definite attractions. Plato, how ever, concerns him self not so m uch w ith the choice o f acts as w ith the choice o f character. T h e R ep u b lic, for exam ple, sets o u t to show th a t we will be better off if we have just, rath er than unjust, souls, an d th a t a life o f justice is therefore preferable to one o f injustice. Just a n d u n ju st acts e n te r th e p ictu re only because they ten d to shape the character — we becom e just by behaving justly and unjust by behaving un ju stly (444c-e). T h e focus o f th e L a w s is sim ilar. T he legislator sets o u t to make th e citizens virtuous, not m erely to secure external com pliance w ith his edicts. T h is view of m orality does not necessarily im ply the absurd idea th at som eone w ith an as yet unform ed character m ight seriously sit dow n to decide w hat kind o f p erson to be. O n e’s character is, o f course, to some e x te n t form ed before one is capable of taking conscious decisions of th is kind. P lato ’s concern is to convince the ordinary m an th at it is w orth struggling to retain one’s m oral integrity and to persuade p aren ts and legislators that it is in their ch ildren’s own interest to be b ro u g h t up in th e paths of virtue.



Clearly a choice oflives would be m ore com plicated th a n a choice of acts. In particular one could n o t tell someone w ho was genuinely asking him self w hat kind of life to live th a t he should choose th e life he would find m ost pleasurable. W hat gives us pleasure depends largely on our desires, wishes and interests. In choosing a life one w ould be choosing, in ter a lia what desires, wishes and interests to adopt. T hus an appeal to the distinctive pleasures o f any life prejudges the issue in its favour. T h e A thenian’s claim th a t the pleasures o f the just life are more genuine because the decision of the just m an is authoritative (663b-c, cf. R epu blic , 582a-583a) begs the question in precisely this way. One could reto rt th at the just m an finds pleasure in justice only because he has been conditioned to do so by his conventional upbringing and has therefore had his capacity to enjoy the pleasures of injustice b lunted. So, while the A thenian is en titled to point o u t that justice has its own pleasures, he is not entitled to say th at they are necessarily greater or more genuine than the pleasures involved in other kinds of life. T h ere are, however, other ways in w hich pleasure could be relevant to the decision which of two different kinds o f life is to be preferred. Firstly there are some experiences, especially certain physical sensations, that almost everybody, whatever his interests, finds pleasant or painful. A life that involved a great deal o f physical pain would, to that extent, seem undesirable to virtually everyone. T hen there is the m atter of external circum stances. One may desire things that are in short supply or difficult to come by, or one may w ant to do som ething that is possible only when conditions are particularly favourable. A life th a t includes m any such desires is inherently more likely to result in frustration and disappointm ent. Lastly, we may notice th at different lives contain different mixes of pleasure and pain. One life may contain periods o f exhilaration m ingled with tim es of despair; another may furnish an even-tenored contentm ent. One could have a preference for one pattern rath er than another w ithout being com m itted to any particular way of life. G iven these points, one can see th a t, although Plato is not entitled to m aintain that the pleasures of justice are intrinsically preferable to those of injustice, he can legitim ately refer to pleasure in his argum ent for the just life. H e could claim th at the pleasures of justice, unlike many others, carry no risk of physical pain, and th at, because they do not depend on external circum stances, they are less likely to be fru strated or to be mixed w ith great disappointm ents. T his is the strategy he follows in Book V.



PLEASURE AND T HE GOOD LIFE B ook V: d ivin e goods an d hum an pleasures

T h e A thenian returns to th e subject o f pleasure in his general ‘p relu d e’ o r ‘p ream ble’ to the legal code (726-734e). T he first six pages o f this p relu d e consist of an exhortation to live virtuously and, in particular, to h o n o u r th e soul as o n e’s m ost divine possession. T h e n a change o f d irection is announced: T h u s, as regards th e rig h t character of institutions and of individuals, we have now laid dow n practically all the rules that are o f divine provenance. T hose th a t are o f hum an origin we have not, as yet, stated. B ut state th em we m ust, for we are talking to m en, not gods. By n a tu re , pleasures, pains and desires are especially hum an; and from these, o f necessity, every m ortal creature, so to speak, depends and hangs by the strongest cords of influence. T h u s one should com m end the noblest life, not m erely because of superiority in reputation, b u t also because, if a m an consents to taste it and does not shun it in his youth, it is likewise superior in th at w hich all m en seek — an excess of joy and a deficiency of pain th ro u g h o u t o ne’s whole life. (732d733a, translation adapted from Bury) T h e co n trast here betw een th e rules of divine provenance and those of h u m an origin is of th e greatest im portance since it reveals the relationship betw een a n u m b er o f distinct and superficially conflicting strands in the moral theory o f th e L a w s. F ro m a very early stage in the dialogue th e A thenian has argued that, since the soul is our m ost divine possession, virtue (the good of the soul) is to be p referred to all hum an goods. (See especially 631b-c, 697b, 726-728c). W ith in th e soul it is th e im m ortal elem ent of reason, n ot the m ortal elem ents (the passions and desires), that is supposed to be divine (644c-645c, 713e; cf. T im aeu s, 42c-44d, 69d-e). T hus w isdom has first place am ong th e divine goods and they look to reason as th e ir leader (631c-d). O ne h o nours the soul by resisting the b lan d ish m en ts of pleasure and by en d u rin g pain and fear (727c). T hus w hen he talks from th e divine point o f view the A thenian is anything b u t a hedonist. H is rules o f divine provenance bid us to struggle against pleasures in order to h o nour th e reason and to subordinate ourselves to it. If we were fully rational beings, like th e gods, we would presum ably conform w ithout question to the dictates of reason. We would see what was good and do it w ithout hesitation. But, as em bodied beings, we are



subject to irrational forces, to desires, to pleasures and pains, and to m otions like anger and fear (644c-645c, 863b-c; cf. T im a eu s , 42a-b). We are thus likely to be virtuous only if we can see th a t the virtuous life will be pleasant (663b). Fortunately, the A thenian believes, the virtuous life is pleasant, so the case for virtue can be m ade on the hum an, as well as on the divine, level. A nd, since the feelings of pleasure and pain have the strongest influence on h u m a n beings, it is this aspect th at is most relevant to the practical work o f the legislator (636d). T h at is why it is im portant to train the young to feel pleasures and pains in the right kind o f way (635a-c, 659d-e '1. By right education the disorderly motions of the lower, m ortal soul can be su bordinated to those of reason ( T im a eu s, 44c, 87a-b). So in setting o u t th e rules of hum an origin th e A thenian is seeking to dem onstrate not th at pleasure is the good but th at a creature subject to pleasures and pains can still be advised to pursue the same kind o f virtuous life th a t would be recom m ended by the reason. 5

Pleasure an d choice

The A thenian now proceeds to describe the rules of hum an origin (733a-d). H e m aintains that anyone who tastes th e good life in the correct way will see th a t it is superior in respect o f pleasure and pain. But what, he asks, is correctness here? To u n d erstand th at we have to com pare the pleasanter and more painful lives and see w hich is genuinely in accord with hum an nature. We will th e n find that: (1) We desire pleasure b u t do not choose or desire pain. (2) We do not desire a neutral state in place o f pleasure, b u t we do desire it in place of pain. (3) We desire m ore pleasure with less pain; we do not desire less pleasure with more pain. (4) W h en tw o states contain equal proportions o f pleasure and pain we cannot say which we would prefer. (An alternative translation would be: ‘We cannot say w hether we would desire a state in which pleasure and pain are equal’.) In assessing these feelings we have to take into account their num ber, size, intensity and equality (or the opposite). Hence: (5) In the case of a life that contains feelings w hich are num erous, great and intense, we want it if pleasure predom inates, bu t not if pain is predom inant. (6) In the case of lives containing feelings th a t are few, small and weak, we do not want that in which pain predom inates, b u t we do want that in which pleasure predominates.



(7) A life w ith an equal balance of pleasure and pain is to be reg a rd e d in th e same light, i.e. we desire it in so far as it predom inates in w hat we like, b u t we do not w ant it in so far as it predom inates in w h a t we do not like. T h e conclusion is th at ou r whole lives are bound up in the feelings of p leasu re and pain and th at we m ust therefore distinguish what kinds of life we by n ature desire. Anyone who suggests th at we may desire som eth in g ap art from these is talking th ro u g h ignorance or inexperience. It is n atu ral for a m odern reader to see this passage as an anticipation o f the B entham ite felicific calculus. T he underlying assum ption of su c h a calculus is that, w henever we have a choice betw een two or m ore co urses of action, the various possible outcom es can, in principle at least, be ranked on a scale o f pleasure and pain. I f we are rational, we choose the course th a t will bring m ost pleasure and least pain. T he B entham ite calculus th u s carries an explicit com m itm ent to hedonism . B u t before jum ping to any hasty conclusions we should notice some im p o rta n t differences betw een P lato’s calculus and B entham ’s. In th e first place, Plato is concerned w ith a choice of lives rather than a choice o f acts. As we saw above, such choices cannot be m ade sim ply o n the principle o f m axim izing pleasures and m inim izing pains, th o u g h pleasure and pain are not altogether irrelevant. T h e second p o in t is th a t the p resen t passage does not provide a m ethod of ranking all lives in relation to one another. It explains how we will choose b etw een certain pairs o f alternatives b u t these by no m eans cover all possibilities o f choice. In p articular, nothing has been said about how we w ould choose betw een lives containing a predom inance of difference kinds of pleasure. T h e u p sh o t o f this is th at 733a-d lists a series o f conditions w hich govern the choice of lives b u t w hich are in sufficient to determ in e th e choice in every respect. T o p u t it another way, pleasures and pains may be constraints upon choice b u t they need n o t be th e sole objects o f choice. In 733d-734e these considerations are applied to the choice of lives. T h e lives of tem perance or self-restraint, wisdom , courage and health are opposed to those th a t are foolish, cow ardly, licentious and diseased. It is argued th a t th e m an who really knows w hat he is doing will prefer the tem perate life to the licentious. It is gentle both in its pleasu res and in its pains and has an abundance o f the form er. T he licentious life, on the o th er hand, is extrem e in b o th pleasures and pains and has a balance o f th e latter. N o one therefore will voluntarily be licentious. People becom e licentious th ro u g h ignorance or inability



to control themselves. Parallel argum ents are used to recom m end the brave, wise and healthy lives. T he conclusion is th a t the life o f moral and physical excellence is superior to that of vice in being pleasanter, as well as in beauty, rightness, virtue and reputation. I t makes its possessor happier in every respect. The argum ent here does not justify the A thenian’s conclusions as they stand. We may grant th at the tem perate life probably is, on balance, pleasanter than one of unlim ited self-indulgence, though this is not strictly proved. B ut one cannot conclude from this that virtue is pleasanter than vice in general. N othing has been done, for exam ple, to prove that a virtuous life is pleasanter than one o f cool calculating villainy. But the A thenian does not need to prove such a strong conclusion. If, as he m aintains, there are rational non-hedonistic grounds for pursuing virtue, all he needs to show is th at it is w ithin the capacity of hum an nature to choose the life of v irtu e, i.e. that the virtuous life can satisfy the constraints on choice described in 733a-d. For this he has to show only that the virtuous life has its own pleasures, that these generally exceed the pains, and that th e vicious do not normally do better in this respect. We are now in a position to answer m ost, th o u g h not all, of the problem s raised earlier in this chapter. In p articular, one can see how the A thenian m ight hope to reconcile the hedonistic and the nonhedonistic strands in his thought. H e believes, it seem s, that virtue is good in itself because it is the ordering o f the soul in accordance with reason. In so far as we are rational, we will therefore require no further argum ent for virtue. B ut, since we are em bodied beings, we shall be able to pursue the life o f virtue only if it is pleasurable. T h u s, although pleasure is not itself the good, it is a necessary condition of the good life for m ankind. In this way the A thenian can, w ith o u t inconsistency, m aintain the propositions that are m ost central to the ethical doctrine of the L aw s: (a) that the virtues are intrinsically of suprem e value; (b) th a t virtue can require us to resist the tem ptations o f pleasure; (c) that we can become truly virtuous only by learning to take pleasure in the right things; (d) th at the good life is pleasurable. V irtue is valuable in itself because it is the life according to reason, b u t the virtuous life is possible for hum an beings only to -th e extent th a t they can take pleasure in virtue. T his interpretation aligns the L a w s quite closely w ith the R epu blic and the Philebus. In the R epublic Socrates is first m ade to argue that justice is intrinsically preferable to injustice (444e-445b) and later that the life o f the just man is more truly pleasant (580c-587c). In the Philebus he concedes that pleasure is an essential ingredient of the good



life b u t argues th a t it is by no m eans th e m ost im portant ingredient nor th e one w hich really makes th e life good (20c-23a, 64b-67b). Both d ialogues apparen tly accept th a t no one w ould deliberately choose a life know n to be less pleasant, b u t th a t does not com m it them to the view that th e good consists in pleasure.


Political Analysis

676a-702a; 302b-303d


Re public,

543a- 580a;

Statesm an,


At first sight the long historical digression w hich occupies alm ost the whole of Book I I I looks irrelevant to th e m ain them es of the L a w s. But its purpose becomes obvious on a closer view: it sets out the principles which, according to the A thenian, should govern both o u r assessm ent of existing states and th e construction of the new C retan city. So, in many ways, it parallels th e passages in th s R epu blic and the S ta te sm a n in which Plato classifies and analyses existing constitutional forms.


The origin o f states

The book begins w ith an account of the developm ent o f hum an society from an im aginary tim e in the rem ote past when natural disasters had destroyed all previous civilizations (676a-682e). T his account is notable for its sense o f th e enorm ous lengths o f tim e w hich have preceded the present epoch and for its reliance on rational reconstruction rather than m yth. It establishes a n u m b er o f points significant for political theory: (1) A lthough it is not autom atic or innate for m an to live in society, states develop naturally over a long period of tim e. A t first there are only family groups, these coalesce into larger units until, eventually, cities are established. Plato im plicitly rejects doctrines like those he put into the m ouths of Protagoras (P ro ta g o ra s , 320d-322d) and Glaucon (R epu blic , 358b-359b) which represent political life as unnatural, the result of divine intervention or of hum an agreem ent. (2) T he prim itive shepherds who lived scattered apart on the hills were lacking equally in virtue and vice (677b, 678b). In other words the L a w s assumes that life in a civilized com m unity is the norm al goal of hum an developm ent. (In the R epublic , 372e, Socrates at least



preten d s that the prim itive life is ideal.) (3) P aternal g overnm ent was originally the norm . W hen family g ro u p s joined together, lawgivers had to be appointed to choose among th e custom s o f th e different groups and to select rulers (68 lc -d ). T hus law giving is a deliberate hu m an activity, though it rests on a natural developm ent.


P eloponnesian h isto ry

A tten tio n next shifts to the history of th e three D orian states of the Peloponnese: S parta, M essene, and Argos (682e-689e). T h is, largely fictitious, account establishes two points o f fundam ental im portance to the political philosophy o f the L a w s : (a) the political im portance o f sophrosune, and (b) the need for a mixed constitution. A ccording to th e A thenian, the D orians, on their arrival in the P eloponnese, established th ree linked states. W ithin each of these the kings and the people bou n d them selves by oaths not to infringe each o th e r’s rights. Each state also swore to come to the aid o f the others if eith er o f them was th reaten ed w ith d estruction. T his arrangem ent should have m eant th at the D orians would obey their laws, b u t in fact it came to nothing (682e-686b). T h e reason, according to the A thenian, is th at, although hum an institutions, used correctly, can give great pow er to those who have them , they can be positively h arm fu l to those who do not know how to use them correctly. If we judge w rongly in determ ining th e ends for w hich powerful in stru m en ts are used, they will lead to disaster, like th at encountered by the foolish m an whose foolish prayers are answ ered. T h e trouble w ith the D orians is th at they were subject to the ‘greatest folly’, i.e. to a disagreem ent betw een th eir feelings o f pleasure and pain and their rational judgem ents (see above chapter 5, section 3). T his ‘folly’ affects th e state w hen the people are unw illing to obey their rulers or the laws. In th e case of an individual, it m eans that the rational principles a man possesses, so far from doing good, may actually be harm ful. N o one in th is condition should hold any office. T h e lawgiver m ust therefore strive to drive out this folly and im plant the wisdom w hich consists in inn er harm ony (686b-689e). T his brings us back to the them es o f Books I and I I , the fu n d am en tal im portance o f sophrosune and the failure of the D orian institutions to foster it. T h e earlier books showed th at sophrosune cannot exist w ith o u t th e rig h t kind of political institutions. N ow we learn that even th e best in stitutions are doom ed to failure if the citizens



lack sophrcsune. N either can exist w ithout the other. So the legislator’s work is incom plete unless he inculcates a proper self-discipline. U nderlying this there is a more general point: no co nstitution can survive unless the people accept it and are willing to follow it, even when they find this uncongenial. I t is n o t enough sim ply to suggest good constitutional forms; one m ust also secure a m atch betw een the constitution and the character and aspiratons o f those who live under it.


S even claim s to rule

At 690a-691a there is a digression in which the A thenian considers seven ‘claims to rule’ in cities and in households. (S aunders’ translation, ‘seven titles to au thority’, may convey too strong a suggestion of legitimacy.) T h e claims are those (a) of parents over their children, (b) of noble b irth , (c) of age, (d) of m asters over slaves, (e) of the stronger to rule th e weak, (f) of the wise to rule the ignorant and, (g) of those selected by lot (which is taken as a sign o f divine favour). It is not at all clear on w hat principle this classification is based. Any interpretation produces seemingly insoluble difficulties: (1) Are the claims to rule claims which are actually m ade, w hether justified or not, or are they claims w hich the A thenian endorses as justified? If th e list is intended to cover all those claims actually made by P lato’s contem poraries, why is the claim of wealth not m entioned? If, on the other hand, it is a list o f claims endorsed by the A thenian himself, it ought not to include the claim of strength w hich he later explicitly rejects (714b-e). (2) G rote (1888, p. 309) and Strauss (1975, p. 47) note that claims (e) and (f) receive most em phasis. Claim (e), the claim o f strength, prevails among the anim als and is called natural by P indar but the Athenian him self prefers (f) the claim of wisdom. Significantly he calls it ‘the natural rule of law over willing subjects’. Strauss, recalling the G orgias, 488d, where the many are said to be strong, identifies the rule of the strong w ith democracy and that of the wise w ith kingship. This would link the present passage to one a few pages later where democracy and m onarchy are nam ed as the two m other constitutions (693d-e) b u t it has no other su p p o rt in the text and probably rests on a m isunderstanding of the kind of classification th at is now being offered. T he Athenian insists that m a n y state there m u st be rulers and ruled (689e) and that the claims to rule apply to households as well as to states (690a). It seems, therefore, that he is concerned not so m uch



w ith constitutional form s as w ith th e claims of individuals to be rulers. T h e re is no need to m ention a claim of the m any because the many can n o t all be m agistrates, b u t th e A thenian does m ention the claim of th o se selected by lot, the dem ocratic states’ way of choosing officials. Sim ilarly nothing is said about m onarchy as such, presum ably because th e claim of any individual to rule as m onarch m ight be based on a n u m b e r o f d ifferen t grounds. M onarchy has in fact been defenced on g ro u n d s of w isdom , strength, noble b irth and paternal authority. (3) T h e claim s of th e m aster to rule his slave and the paren t to rule his child apply in th e first place to households. T he others all apply in th e first place to states. T o list th em side by side, is, to say the least, confusing (cf. A ristotle, P o litic s , I.i, 1252a 9-17). (4) At least one o f th e claim s, th at o f lot, indicates a definite m echanism by which rulers can be selected. T he claims o f noble birth an d age also suggest procedures for deciding who should rule and they correspond to know n co n stitutional form s. T he claim of wisdom , on th e o th er h an d , suggests no p articular procedures. H ow are we to decide who is wisest? I t is difficult to avoid th e conclusion th at Plato has not th o ught out at all clearly w hat he is try in g to do w ith the classification of claims to rule. A t 690d the A thenian suggests that conflict betw een the different claim s is an im p o rta n t cause o f dissension, b u t in fact he attrib u tes the failure of th e D o rian states n o t to this b u t to failures o f character. O th erw ise, a p a rt from ra th e r casual references at 692a and 714e, he ignores the classification altogether. T his is a pity, because his fu n d am en tal idea may be a good one. In P lato’s day, at least, su p p o rters of different constitutional form s differed fundam entally on th e q uestion, ‘W hat considerations are relevant in deciding who should exercise pow er?’ As long as such disagreem ents persisted there w ould always be a danger o f civil strife. So it m ade sense to devise a co n stitu tio n w hich took account of as m any o f the different claims as possible. T o a considerable extent, Plato achieves this in the co n stitu tio n he proposes for the C retan city.


S p a r ta a n d the m ixed con stitu tion

A rgos and M essene failed because th eir kings, corru p ted by luxurious living, transgressed the very laws they had sworn to obey. T he legislator should see th at in all things one m u st respect due m easure (otherw ise ‘p ro p o rtio n ’ or ‘m oderation’). Ju st as ships cannot bear sails th a t are too large or bodies too m uch food, so a m an cannot bear



more than a due measure of power. A young man who wields great power unchecked is bound to be overcome by folly (691c-d). O f the three D orian states, S parta aione survived in good order. This happened because it developed a system in which there were (a) two kings, rather than one, (b) a council o f elders and i c ; pow erful officials called ephors, chosen by a m ethod close to that of lot. All o f these could act as a check on one another. T hus it survived by becom ing a m ixture of different elem ents and acquiring ‘due m easure' (691d-692a). T he original founders of th e Peloponnesian states sought in vain to impose moderation by means o f oaths. Now, in the light of th e Spartan experience, we can see what really should have been done (692b-c). To found a state which will be free, wise and in friendship w ith itself, the legislator m ust avoid establishing any office with great or unm ixed power (693b). This introduces another of Plato’s key principles, that of the moderate or mixed constitution. (The word m etron means ‘m easure’ in the sense of ‘th at by which som ething is m easured' b u t it also has the sense ‘due m easure’, ‘lim it’ or ‘proportion’. T he corresponding adjective, m etrios , means literally ‘m easured’ but is usually translated ‘m oderate’.) T here is, of course, nothing particularly original in the praise of m oderation. Sayings like ‘N othing in excess’ or ‘T he half is b etter than the w hole’ were com m onplace. Even the application of these ideas to constitutional theory is not new. T h e constitutions of Solon and Theram enes, for exam ple, had both been called ‘m oderate’ or ‘m easured’. B ut an appeal to ‘m oderation’ or ‘due m easure’ could mean many different things. As applied to Solon’s constitution it m eant a balance betw een the claims o f the rich and the poor. As applied to that of T heram enes it referred to a m oderate oligarchy which was supposed to come midway betw een extrem e oligarchy and democracy. In the present context Plato is em phasizing the need to restrict the power of individuals. T he power of the Spartan m onarchy is restrained by dividing it between two kings and by com bining it with other kinds of authority. T his idea th at m oderation can be secured through a constitution in which different elem ents act as a check on one another is of immense historical importance. It is, perhaps, the. m ajor contribution of the Lauis to political theory. But there is more than this to P lato’s conception o f the mixed or m oderate constitution, as becomes apparent in the pages that follow.



POLITICAL ANALYSIS P ersia a n d A th em

At 693d th e A thenian proposes to describe two ‘m other constitutions’ (T aylor: ‘m atrices o f c o n stitu tio n s’), m onarchy and dem ocracy. He regards all other constitu tio n s as com binations o f these. T he extrem e o f m onarchy is rep resen ted by th e Persians, th at o f dem ocracy by A thens. T he success o f the Spartans and C retans is attrib u ted to their avoidance of these extrem es. O n the A thenian’s account the Persians have experienced altern atin g periods o f success and disaster. At first, un d er Cyrus, they kept a balance betw een slavery and freedom . As a resu lt they were not only free them selves b u t becam e m asters of others. Because the rulers g ranted their subjects a degree of freedom and equality th e soldiers were well disposed to their com m anders and willing to en co u n ter dangers. Because there was freedom o f speech, any m an was able to add his w isdom to the com m on stock. T h u s everything w ent well because of their freedom , their friendly feelings, and their ability to share their pow ers o f reason (694b-c). A fter th e d eath o f C yrus disaster overtook the Persians, because C y ru s’ sons who had been b ro u g h t up by the wom enfolk w hen their fath er was away on cam paign were, as we would say, ‘hopelessly spoilt’ (694c-695b). A fter D arius seized th e throne the process repeated itself. D arius enacted laws w hich gave a degree o f equality and lim ited his financial exactions. T h u s, by m oney and gifts, he won over the Persians, aroused feelings o f friendship and com m unity and gained the good will o f his arm y (695c-d). A fter his death disaster struck again because his successor, X erxes, had had the same bad upbringing as the sons of Cyrus (695d-e). T h e Persians failed through an excess of slavery and despotism . T he A thenian experience, by contrast, shows that liberty w ithout authority is vastly inferior to a m oderate form o f control. T he success o f Athens at the tim e o f the Persian wars is attributed to fear of the laws which they were accustom ed to serve as slaves do a m aster. T his fear o f the laws is called aidos: ‘resp ect’, ‘reverence’ or ‘aw e’ (S aunders’ ‘m odesty’ is too weak). Such attitudes together with their fear of the invaders created feelings of frien d sh ip and com m unity (698b-c, 699c). This voluntary slavery to th e laws was especially evident in the case o f laws governing m usic (700a). People used to subm it willingly to discipline in these m atters, b u t now they consider only w hat gives pleasure; the applause of th e m u ltitude has replaced the judgem ent o f those who really know w hat is good or bad. F ro m m usic this disorder can spread to affect the whole life o f the com m unity. People begin to think they know everything and to behave lawlessly. T h ro u g h confidence in their



own wisdom they become fearless, and this fearlessness breeds a kind of shamelessness — a disregard, through excess o f liberty, of the opinion of one’s betters. T here follows an unw illingness to serve the authorities and a contem pt for parents, for law and even for the sanctity o f oaths (700a-701d).


The politics o f due measure

T here is no doubt th at Plato is unfair to Athenian dem ocracy. Law and order did not break down even though.there was a substantial degree of personal freedom. Plato’s account of Persian history may be equally inadequate. B ut Persia and Athens are introduced into the argum ent not for their own interest but as examples of extrem e m onarchy and extreme democracy. They function as ideal types. So w hat really matters is not w hether the picture of these two states is historically accurate b u t w hether Plato is right in suggesting th a t a well ordered state m ust be a m ean betw een the two extremes as he describes them. It is natural to suppose that in discussing m onarchy and dem ocracy Plato is concerned w ith constitutional form s, th at m onarchy represents the unchecked rule of a single individual and dem ocracy the rule of the many. T h u s Taylor suggests that the A thenian is dem anding a balance betw een personal rule and popular control (1960a, p. xxvii). But in fact th e A thenian says nothing about institutions as such; he concentrates entirely on the need for a balance between freedom and despotism . A balance o f this kind was apparently achieved both by th e Persians und er Cyrus and D arius and by the A thenians at the time of M arathon and Salam is, even though the Persians had no dem ocratic institutions and the A thenians no m onarch. T his is what one would e x p e a in the light of the S ta tesm a n , where Plato argues that it does not m atter w h ether there are few or m any rulers. W hat really m atters is w hether the ruler(s) possess the science of ruling (292c). If such an exp ert ru ler is n o t available, what matters is not prim arily the num ber o f rulers b u t w hether they rule in accordance w ith law (302b-303b). W hat is m eant by freedom and despotism in this context? T he freedom Plato attacks is an absence of restrictions, a situation in which individuals may do as they please. One would therefore expect its converse, despotism , to be a situation in w hich this personal liberty is very restricted. But, if this were so, then the A thenians when they were ‘slaves to the law’ would probably have been living under a greater despotism than the Persians ever experienced. As M ill saw,



dem ocracies may th reaten individual liberties m ore th an m onarchies. I n fact, the trouble w ith the Persians in their more decadent periods seem s to be th a t they w ere subject to arb itrary exploitation by their ru lers. T h e kings used their subjects for th eir own purposes w ithout an y consideration of w hat was rig h t or good for the subjects them selves. T h e b etter m onarchs, on the other hand, sought the good o f th e ir subjects an d listened to w hat they had to say. So the subjects w illingly accepted th eir rule in th e same way as the A thenians willingly enslaved them selves to the laws. T h e balance to be sought betw een m onarchy and dem ocracy is not a m idw ay point betw een having m any restrictio n s on personal liberty and having none. It is a condition in w hich restrictions are accepted w illingly because they are recognized to be good. T h is is b ro u g h t o u t clearly in Book IV , 712e ff.: m ost existing form s o f rule do not deserve the title ‘co n stitu tio n ’ because th ey req u ire some citizens to live in subjection to others like slaves (712 e - 7 13a). T h e tru e constitu tio n will involve subjection to laws that aim at the good o f th e whole com m unity (715a-d). P la to ’s theory o f th e m oderate or m easured constitution is the political co u n terp art of a doctrine p ro m inent in his later metaphysics. In the S ta te sm a n (283c-284b), he distinguishes two kinds of m easurem ent. One kind concerns itself w ith the greatness or smallness o f its objects relative to one another; th e other is concerned w ith their greatness or sm allness relative to ‘due m easure’ or ‘the right stan d ard ’. Plato suggests th a t th e arts and crafts produce good and beautiful results by m aintaining due m easure. In the P hilebus (23c-26d), all reality is divided into th e u n lim ited, the lim it, the m ixture of these two, an d the cause o f this m ixture. T h e unlim ited includes qualities capable o f being m ore or less, for exam ple, hotter and colder; the lim iting factors are m easure, p ro p o rtio n and ratio. T h e im position o f the limit on th e u n lim ited produces order and harm ony, so health, m usic and good w eather, for exam ple, are all th o u g h t o f as m ixtures o f th e lim it an d th e unlim ited . T h e good life involves the im position o f law and o rd er, w hich involve th e lim it, on o u r pleasures and appetites. In the universe at large these orderly m ixtures are created by the divine reason o f w hich our hum an reason is a small fragm ent. T h e parallels b etw een these passages and th e doctrine o f the m oderate constitution are obvious. D espotism and freedom form an indefinite continuum like th e hot and the cold. In a well ordered state these extrem es are reconciled by law. Law in its tu rn is a p ro d u ct o f reason which governs th e universe. G od is th e m easure o f all things (716c). T h e theory o f the lim it and th e u n lim ited presum ably has its origin in Pythagorean m usical theory. I t fits m usic well. If one stops a



vibrating string so that the ratio of th e two parts is, say, 1:1, there will be a concord; if one stops it so that the ratio of the p arts is, say, 5:2, there will be a discord: it will sound as though the two parts are in conflict w ith one another. T hus one can think o f m usical harm onies both as m idpoints betw een extrem es, and as reconciling conflicting elements. One may be able to apply such notions to health, since health does depend on m aintaining the right tem perature and the rig h t ratios among the chemical constituents of one’s body. But they offer no guidance w hen applied to the soul or to political systems. N o d o u b t the virtue and the well-being o f a state depends on the acceptance of some form of law and order, bu t form s of political order cannot be expressed in num erical ratios, and in any case there is no one form of order which is objectively right, laid down in the nature o f things. Thus in politics and ethics, the doctrine of m easure lacks the scientific status it has elsewhere. In practice it boils down to the claim th at a good constitution m ust avoid extrem es, impose checks on individual power and reconcile conflicting claims.


The Sovereignty of Law L a w s , 627e-628a, 656c-657b, 691c-692c, 698b-c, 699c, 700a, 712b-

715d, 769a-771b, 772a-d, 822e-823a; S t a t e s m a n , 300a-301a


In w h a t sense can la w be sovereign?

T h e fine schem e of th e Peloponnesian states collapsed because their ru lers, lacking self-control and subject to no external constraints, failed to obey th e laws. T h e A thenians, likewise, prospered w hen they w ere slaves to their laws and declined w hen they abandoned them . T he assu m p tio n w hich underlies these accounts, th at in a well run state law w ill be suprem e, is b ro u g h t to the surface in Book IV, 712b—715d. At 7 1 2 d -e, M egillus professes him self unable to classify the Spartan co n stitu tio n according to the conventional categories of political th o u g h t. It is not exactly ty ran n y , aristocracy, dem ocracy or kingship, th o u g h it has elem ents of all these kinds. T he A thenian attributes M eg illu s’ difficulty to the fact th at S parta has a genuine constitution: w h a t pass for co n stitutions elsew here are sim ply arrangem ents by w hich one section of th e com m unity enslaves another (712e-713a). T h e re follows a digression on the age of C ronos, the im aginary tim e in th e rem ote past w hen divine beings ruled directly over m en (713ae). T h e conclusion is th a t we should ru n households and cities in accordance w ith th e sm all elem ent we possess o f im m ortality, i.e. in accordance w ith reason. W e call the form o f order created by reason ‘law ’ (713e-714a). T h e A thenian th en attacks existing states whose socalled laws sim ply enforce th e interest of the stronger party: genuine laws m ust be directed to the good o f th e city as a whole (715b). Rulers m u st be chosen, not for their w ealth, strength, physique or any such th in g b u t for their w illingness to be slaves to the laws. I see destruction th reaten in g any city where the laws are o verruled and lack authority; b u t where law is m aster o f the rulers and the rulers are its slaves, I see salvation and the prom ise o f every blessing th at the gods bestow on cities. (715d)



This claim th at law should have ultim ate authority can be interpreted in a num ber o f different ways: (1) T h ere is a sense in w hich any state in w hich there is a recognizable governm ent m ust acknowledge the au thority of law. There m ust, at very least, be a rule which says who is in charge. An absolute m onarchy, for example, m ight be ru n according to one basic law, ‘Obey th e king’. But in a state o f this kind one could not say that the ru ler was the servant or slave of the laws, since th e law gives him unlim ited power. (2) A state may be said to be governed in accordance w ith law if the governm ent recognizes the so-called ‘rule of law’, i.e. the governm ent acts in accordance with general rules; these rules apply to all alike and are m ade public in some way; disputes are settled by some nonarbitrary procedure. Observance of th e rule of law, understood in this sense, certainly restrains governm ents and may prev en t some of the worst injustices. But if the governm ent can change th e law at will, it can hardly be said to be the servant of the laws. (3) To make sense of the notion that rulers may be servants of the law we need to add at least two further conditions, (a) T he laws m ust be sufficiently specific to act as genuine restraints on governm ent action, (b) T he governm ent m ust not be able to alter the law at will,


H ow can the sovereign ly o f law be ensured?

A good m any of the A thenian’s proposals could be seen as prom oting the sovereignty of law: in particular every official in the proposed Cretan city w ould be subject to some kind o f judicial control. T he A thenian w ould th u s p u t into practice the lesson he drew from his discussion o f Spartan history — th at no one should be given unrestrained pow er (6 9 lc -d , 693b). His doctrine here is the ancestor of m odern theories of constitutional checks and balances, including M ontesquieu’s celebrated separation of powers. B ut Plato does not distinguish judicial and executive power. M o st of the officials in his state would possess both. Instead th e constitution is so arranged that each body of officials is responsible to some oth er organ of state. T hus no one should be able to get away with any illegality. Plato sees clearly that talk o f the suprem acy of law becomes meaningless if th e dom inant parties in th e state can change the law at will. At 713c-715d he contrasts the sovereignty of law w ith political systems in w hich so-called ‘laws’ are used by factions for their own purposes. G enuine laws are designed to prom ote the good o f the whole



c o m m u n ity , not to satisfy the greed o f the dom inant party. T o guard a g a in st this kind o f abuse th e A thenian envisages th at a detailed legal code will be laid dow n in advance by a single legislator or legislative com m ission. T h is code will be very difficult to change and the citizens will be educated so th at they becom e slaves to the law, not only in the sense that they respect the legislator’s explicit provisions, b u t also in th e sense th a t they strive to em body his ideal of virtue (822e-823a). Som e scholars m ight challenge this claim that law, in the A thenian’s view , should be very difficult to change (H all, 1981, pp. 98-100). T h e re are certainly a n u m b er o f passages where he suggests that the legislation proposed for the C retan city is incom plete. It will be left to th e officials o f the new state (principally the guardians o f the laws) to fill in th e details 772a-c, 818e, 828b, 835a, 843e, 846b-c, 957a-b, 968c). In at least one case he adm its th at his m easures may need alteratio n (not m erely supplem entation) in the light of exprience (840e). In 7 6 9 a-d he com pares the task o f legislation w ith that o f an a rtis t who constantly retouches his pictu re, always im proving it, b u t n e v e r reaching perfection. O ne m ight draw from this the im pression th a t th e A thenian sees legislation as a continuing process. T his w ould, I believe, be a m istake. In m any o f th e passages w here the A thenian m e n tio n s th e need for legislative supplem entation he stresses th at this is to be concerned w ith m atters of detail, the kind o f th in g th at any c o m p e te n t person could handle or for w hich existing codes o f law p ro v id e adequate guidance (772a-c, 843e, 846b, 957a). He does not, it seem s, anticipate significant innovations. M oreover he expects the perio d durin g which supplem entation will be needed to be strictly lim ited . In the case o f th e regulations for religious festivals there sh o u ld be a ten, year period durin g w hich the various boards of officials (acting w ith th e original legislator if he is still alive) should report om issions in th eir ow n d ep artm en ts to the guardians of the laws and sh o u ld suggest im provem ents. T h e regulations will then be declared u n alterab le and enforced along w ith the measures established by the legislator. T h ereafter th e details of th e festivals may be altered only of necessity w ith th e consent of th e whole people, all the officials and all th e oracles (772b-d). T h e re is a sim ilar, though less explicit, treatm ent fo r th e details o f legal procedure (957a-b). T he im plication o f this is th a t once th e state is properly established, the law will be very difficult to change — precisely w hat one w ould expect from the A thenian’s reveren ce for the long-established codes of C rete and Sparta, his g en eral hostility to change, and his belief th at law em bodies principles o f th e eternal reason.




Is the sovereignty o f law realistic?

One possible objection to the A thenian’s conception of th e sovereignty of law is that his ideal is not a practical possibility. In the work of any hum an legislator th ere are bound to be im perfections th at need to be rem edied in the light of experience. M ore im portantly he fails to recognize that law m ust change itself to m eet changing social and economic conditions. A system as rigid as that described in the L a w s would be doom ed to collapse; so the ideal of the sovereignty of law, it might be said, is no more realistic th an the ideal of the philosopher king. Against this it could be argued th at a degree of fixity in law is obviously desirable. Otherw ise there will be nothing to prevent constitutional rulers from turning themselves into tyrants by (e.g.) prolonging their own term s o f office or proclaim ing th a t all opposition is henceforth illegal. A nd, since most people obey the law out of force of habit, it is unlikely that a code w hich constantly changed would com m and the respect of the population at large — it m ight even be difficult for them to rem em ber what the law is. Any civilized way m ust therefore presuppose some kind of com prom ise betw een Plato’s ideal of an unchanging code and a system in which rulers can alter the law at will. In practice m ost of the states we now regard as having constitutional governm ents have effected such a com prom ise by distinguishing different kinds o f law. Besides the ordinary rules by w hich the citizens are supposed to govern their daily lives, there are certain rules which determine what is to count as a valid law and how legal disputes are to be adjudicated. Rules o f this latter kind, or at least the m ost im portant of them , may be thought of as form ing the constitution. T he constitution may also include other elem ents, particularly rules to render void any legislation which w ould infringe basic rights of the citizens. One of th e essential features of constitutional governm ent is that these fundam ental rules should not be subject to frequent or arbitrary change. This does not m ean that the constitution m ust be unchangeable: merely that in practice it m ust be difficult to effect constitutional change. W here, as in the U nited States, there is a w ritten constitution, this stability is ensured by the requirem ent that special procedures m ust be adopted before constitutional changes can be b ro u g h t into effect. M any believe th t the absence of this kind of constraint is the m ain weakness o f parliam entary dem ocracy as it has evolved in G reat Britain. But even here there are some constraints on legislation. T h e m em bers of Parliam ent cannot sim ply change the law



a t will. T h e ir decisions are law only if certain rath er com plex p ro ced u res have been followed. T o some extent P lato is aware o f the distinction betw een c o n stitu tio n a l an d oth er laws. Book V I, 75 la -7 6 8 e , describes w hat we w o u ld call the constitution o f the proposed C retan state, and the tra n s itio n to th e substantive legislation is noticed (768e). B ut Plato sh o w s no aw areness th a t constitutional law could be given any special s ta tu s . In general it is assum ed th at talk o f a state’s laws, nom oi, and of its co n stitu tio n , p o lite ia , are one and th e same thing. E qually he makes legislative change so difficult as to be virtually im possible. In doing so h e pitches the requirem ents o f the sovereignty of law im possibly high. S tab ility need not m ean total rigidity. A second objection to th e ideal o f th e sovereignty of law is th at any legal system is bou n d to ensh rin e th e interests of the dom inant classes. T h u s those who appeal to the sovereignty o f law are, it is alleged, in fact seeking to perp etu ate class rule. T his p a tte rn of criticism , w hich has resu rfaced in M arxist theory, was fam iliar to Plato (712e-713a, 714b715a; cf. R e p u b lic , 338c-339a). Plato ’s reply to this objection is clear en o u g h : genuine law seeks the good of th e whole state (715b) and is the p ro d u c t o f reason (713e-714a). So w hile it may be tru e that so-called c o n stitu tio n s o f existing states reflect class interests this would not be tr u e of a state w ith a genuine constitution. As it stands this reply is unconvincing because it assum es that there a re absolute standards o f rig h t and th at hum an legislators may know w h a t these are — points th a t would be denied by P lato’s critics. But P la to could fall back on the doctrine o f the S ta tesm a n , 297e-303b. E v e n w hen legislators have been im perfect, a state governed by law is b e tte r off th a n o n e subject to the arbitrary rule of an individual tyrant o r o f a faction. One does n o t need to believe that there ever could be a sta te w ith a com pletely rational code o f legislation in order to m aintain th a t the rule of law offers th e best hope for m ankind.


Sophrosune a n d the so vereig n ty o f law

T h e doctrine o f the sovereignty o f law em braces m ost o f the keyp o litical ideas expounded in the L a w s. F ro m the beginning, one view o f th e legislator’s role is th a t he has to produce internal peace as well as secu rity from external dangers. T his is best done through reconciliation by means of laws (627e-628a). In 715a-d the A thenian a rg u es, in effect, that this can only come ab o u t if the rulers are servants o f th e law. Sim ilarly the discussion of A thenian and Persian history



shows that a state can only prosper where the constitution is m easured, i.e. where voluntarily accepted laws im pose an order o r stru ctu re upon it. And the experience of the Peloponnesians shows th a t this lawabidingness can be achieved only where political pow er is divided among different office-holders. But the sovereignty of law is not simply a political slogan. It is the counterpart o f the oth er main them e developed in these early books, the them e o f sophrosune , self-control. At the beginning, it is argued that the greatest victory is that over oneself, i.e. the victory o f the better elements in the soul over the worse. B ut the political parallel suggests that even such a victory is inferior to a genuine harm ony, a condition where our conflicting passions and motives have been reconciled by the internalization of law. M ost of Books I and II are devoted to showing how th e institutions of the state can be designed to prom ote this inner harm ony. Book I I I , on the other hand, argues th at the institutions o f the state them selves depend on an underlying sophrosune , the ability to abide willingly by the rules. Since, in practice, no hum an being possesses this virtue to a sufficient degree, the constitution m ust be arranged so that external constraints supplem ent our self-control. Plato sees a two-way relationship betw een the sovereignty of law and sophrosune. Law can be truly sovereign only w here the people possess self-control, b u t they can acquire this virtue only by being brought up under the right institutions. T here is obviously a good deal in this view. Laws by their very nature are co n straints u p o n w hat we do. Obedience to law requires us to act against our ow n desires, at least on occasion; so voluntary obedience requires a m easure o f self-control. Of course, a state can survive if some of its m em bers obey the law only to th e extent that they are coerced. B ut if no one obeyed the law voluntarily there would be no one to exercise this coercion; we cannot all coerce each other all the time. T hus self-control is essential to life in a com m unity. On the other hand, it seems that hum an beings cannot acquire self-control if left to Themselves. We acquire it by internalizing rules laid dow n by parents, teachers and society at large. So sophrosune and th e sovereignty of law are m utually interd ependent; we cannot have one w ithout the other. In this respect th e argum ent of the L aw s is m uch more powerful than that of the R epublic. T he key virtue in the R ep u b lic is justice both in the state and in the individual. One major difficulty in the. R epu blic is that the relationship betw een the just individual and the just state is purely analogical. T he state is just when the three classes which com prise it are properly related to one another. T he soul o f the



individual is likewise just when its three parts exhibit the correct kind o f ord er. O n this account th ere is no obvious reason why the just state, ra th e r than states o f other kinds, should be inhabited by just individuals. In th e L a w s there is also a parallel betw een the state and th e soul, b u t the connection betw een the good state and the good soul is m uch m ore obvious. One can see why the sovereignty o f law should flourish only where the citizens possess sophrosune and also how recognition of th e sovereignty o f law m ight itself prom ote that virtue.


Towards the Ideal State Lazes, 702a-e, 704a-712a, 734e, 739a-e, 745e-746d. 747d-e; R e pu b lic,

especially 473c-474bj 499a-502c, 592b

The L a w s, like th e R epu blic, has often been described as utopian (e.g. Saunders, 1970a, p. 13). If this description means sim ply th at the L a w s depicts an im aginary society, no one could take exception to it. But the term ‘utopian’ has m ore than one sense. According to the dictionaries, it can be applied to anything which w ould be ideally perfect but is not, as a m atter o f fact, a practical proposition. In political philosophy it is generally applied to schemes of reform which envisage the abolition of existing form s of society and their replacem ent by b rand new models. Thus there are two questions which need to be answ ered before the Laics is characterized as utopian: (1) In what sense is the city described in the Lazos an ideal? (2) W hat part does the description of the ideal state play in Plato’s political theory?

1 Id ea l states and utopias Aristotle (P o litics , IV.i) rightly observes that, although political theory is always concerned with the ideal or best constitution, what constitutes the ideal will vary according to the precise circum stances we envisage. T h e point here is really a logical one. 'B e st’ means ‘best possible’ and possibility is always relative to assum ed background conditions. The best constitution achievable by us in the next five years will not be th e same as the best constitution that could be achieved by people who were more rational or more benevolent than ourselves, by ones who lived in a kinder natural environm ent or by ones who had inherited a different set of custom s and institutions. Confusion on this point is the source of many criticism s of utopian theory. In discussing a proposed ideal society we have to be clear



w h e th e r it is envisaged as having an im m ediate application or merely as w h at w ould be best if circum stances were m ore favourable than they a c tu a lly are. J u s t as th ere are different varieties of ideal so the description o f an id e a l can play d ifferen t p arts in political theory. Talk of utopianism g en erally brings to m ind schem es such as those o f the levellers or of the e a rly socialists who were attacked by M arx and Engels. T hey aim ed ‘to d isc o v e r a new and m ore perfect social order and to impose this on so ciety from w ith o u t by p ropaganda and w herever possible by model e x p e rim e n ts ’ (E ngels, 1972, p. 609). T h e ir schem es were intended as b lu e p rin ts for im m ediate action. It is entirely appropriate, therefore, to o b ject to such schem es on th e grounds th at they could not possibly b e p u t into practice in the social and econom ic conditions of the time. T h is does n o t m ean th a t such utopian schem es are entirely valueless. T h e y could still help to focus th e discontents o f the underprivileged a n d to define th e principles by w hich a b etter life could be sought. A t th e o th e r extrem e there are descriptions o f ideal societies whose a u th o rs do not claim that they could, even in principle, be p u t into p ra c tic e by real m en in the real w orld. P lato’s descriptions of th e age of C ro n o s (L a w s, 7I3a-714a; S ta te sm a n , 269c-275a) would come into th is category. At th at tim e th e gods ruled directly over m en, and tro u b le and m isfortune were unknow n. P lato, like H um e after him ( T r e a tis e , I I I , I I , ii), m entions such stories, n o t because he believes th a t th e golden age could ever becom e a reality, b u t to contrast it with th e real w orld in w hich we live. In ord er to survive in these more d ifficu lt conditions we have to create peace and harm ony by means of law rath er th a n by relying on divine rulers. B etw een these two extrem es th ere are m any different possibilities. M o re ’s U to p ia , for exam ple, describes a com m unity whose inhabitants are fully hu m an and w hose n a tu ra l en v iro n m en t is not very different fro m our own. T h e difference is th at they operate consistently at a level o f rationality w hich we achieve only rarely. N o one w ith any sense w o u ld expect to make this U topia a reality, b u t the contrast betw een U to p ia and our own society throw s into relief our own folly and greed. Speaking m ore generally, th e description o f an ideal society w hich is n o t a practical proposition can give a m ore concrete expression to a b stra c t political and m oral principles. In ste ad of merely asserting the p rin cip les o f (e.g.) liberty, equality and fraternity, one can attem p t to describe a society in w hich these principles are fully instantiated. A lth o u g h such a society will never come about, it may be som ething tow ard s w hich one can aspire, a direction in w hich one can move. And it m ay be com prehensible to those who w ould not be able to grasp the



principles in the abstract. In the same way some people may find it easier to take the life of C hrist as a m odel, rather th an to achieve intellectual m astery of the principles o f C hristian ethics. Seen in this light there may not be m uch difference betw een som e Utopians and other kinds o f social reform ers, between th e politics of aspiration and the politics of principle. Bentham , for exam ple, was a systematic social reform er. He did not seek to rem odel society from w ithout but to suggest the principles by which society m ight reform itself from w ithin. He could perhaps have done this by describing a perfect utilitarian society. I f he had done this w ith any skill he m ight have escaped the charges of aridity which are often levelled against him. His theories would have had more appeal to the im agination. T h e danger would be th at readers m ight have been distracted by the peripheral details of his account and m ight therefore have lost sight o f the underlying principles.


The C retan colony

The transition from the m eandering discussions of the early book to the account o f th e im aginary city begins at 702a. T he A thenian summarizes w hat has been achieved so far. The object o f the discussion has been to discover how a city m ight best be ru n and how an individual m ight best m anage his life. B ut some test is needed to discover w hether all this has been worthw hile. C onveniently Cleinias, the C retan, announces th at he has been appointed to a legislative commission established w ith the object o f draw ing up a legal code for a new C retan colony. T h e com m ission may take their laws from any source, native or foreign, providing only that they choose the best. T he Cretan suggests, therefore, that he and his com panions should select from the topics they have discussed so far and ‘establish a city in words, founding it, so to speak, from the very b eg in n in g ’. T his will give the A thenian the kind o f enquiry he is seeking while Cleinias may be able to put w hat they say to practical use when he engages in his own legislative activity (702b-d). There are several clues here to P lato’s purpose in describing the city of the L a w s : (1) The reason the A thenian gives for describing the city is to test the principles which have been expounded in the earlier books, presum ably by dem onstrating how they might w o rk o u t in practice. So Plato seems to think o f him self as proceeding from the discovery of basic principles to their instantiation in the constitution of the city. He



d o es n o t first pain t an attractive picture and th e n try to deduce from it a political program m e: the principles have priority. (2) T h e fo undation o f th e new C retan city provides the occasion fo r exp o u n d in g a constitu tio n , b u t the co n stitution the A thenian suggests is n o t supposed to be th e constitution for that city. Though th e A thenian shows in terest in the particular circum stances o f the city a n d even addresses the subjects o f his im aginary legislation as ‘M ag n esian s’, C leinias does n o t ask him to do his legislative work for h im ; he merely hopes th at he will be able to make some use o f the A th en ian ’s suggestions. It looks as though there is good reason for this. T h e A thenian believes, no d o u b t rightly, th a t constitutions m ust be a d a p te d to the p articular circum stances o f each city. B ut to explore in g reat detail the exact circum stances o f the new city would involve Plato in discussion of a great deal of m aterial quite irrelevant to the needs of o th e r cities and to th e interests o f m ost readers. H e therefore offers us som ething of a com prom ise. A lthough the proposed co n stitution suits th e general situation of th e new C retan city, the A thenian does not investigate its applicability in detail. In deed he som etim es indicates th a t adaptations will have to be m ade to suit prevailing conditions (745e-746d, cf. 737c-d). (3) T he C retan legislative com m ission is n o t expected to be p articu larly original in the laws it proposes. It is to select w hat seems b e st from C rete and elsewhere. T o a considerable extent this is what th e A thenian him self does: m ost of his suggestions have some basis in existing G reek practice; such originality as he shows lies chiefly in his m an n er of incorporating and codifying these elem ents into a new co n stitution. T h u s Plato is n o t proposing to engage in pure speculation b u t to build upon the experience o f real m en in real cities.


The conditions fo r legislation

T h e A th en ian ’s first reaction, on being invited to propose a constitu tio n , is to m ake enquiries ab o u t the situation and circum stances o f the new city. T hese are described in sufficient detail to suggest that Plato has a quite specific region of Crete in m ind (704bc, 705c; M orrow , 1960, pp. 50-1, 95; Saunders, 1970a, p. 15). Later it is given a nam e, M agnesia. T h e geography o f the proposed site turns o u t to be perfect in th e A th en ian ’s eyes, largely because there will be little tem p tatio n for the citizens to engage in foreign trade or to establish a navy. T h e fact th a t th e colonists will be draw n from many d ifferen t cities will make it easier for them to accept a com pletely new



code of legislation, though it will make it more difficult to make them feel as one (708a~d). T h e Athenian goes on to em phasize the importance of what we m ight call contingent factors in legislation. Accident, war, poverty and disease are the universal legislators. G od is the suprem e ruler and under him comes chance. But there is also need for the skill of the hum an legislator who knows how to use opportunities offered (708e-709c). The skilled legislator w ould know what circum stances to pray for in order to further his work, i.e. he would know w hat is the most propitious situation for the introduction o f a really good legal code (709d-e). This turns out to be w hen a wise legislator can co-operate with a young b u t naturally self-disciplined ty ran t. A m ong other kinds of state the most suitable will be a constitutional m onarchy followed by some form of democracy. Oligarchies will be especially difficult to reform because in them the largest nu m b er of individuals have entrenched power (709e-711a). T yranny offers the best opp o rtu n ity for carrying through reform s because those who possess pow er are strongest and fewest in num ber. N o t only is a tyrant best able to force through legislation, he can also exercise the greatest m oral influence over the character and behaviour of his people. T hus the fundam ental requirem ent is to create a desire for tem perate and just institutions among those who exercise power, w hatever the basis of th at power. Although this great blessing has come only rarely to m en, when it does occur states can easily and quickly be transform ed. T here are a num ber of puzzling features here. One is the contrast between the realism the A thenian shows at the beginning of the passage and th e optim ism evinced at the end. He begins by asserting the im portance o f geographical, social and other factors which lie outside hum an control; the legislator’s skill lies in m aking best use of these. H e thus rejects, quite explicitly, the kind of utopianism which rests on the supposition that hum an planning can end all social evils. At the end of the passage, on the oth er hand, he suggests, quite unrealistically, th at a tyrant acting under wise guidance could easily change the habits of an entire city. A second source o f difficulty lies in the sudden switch from problems about the foundation of a new city to problem s about the reform of an existing one. T h e A thenian has been asked to ‘found a city in w ords’ and his enquiries about the geography and citizen body of the new C retan city relate to this request. B ut in talking about the tyrant who co-operates w ith the wise legislator he is really asking what conditions facilitate the reform of an existing state. T his is the more puz 2 ling because the situation in which Cleinias and his colleagues



have been asked to legislate looks m ore favourable than could ever be fo u n d in an established city. T h ey have ca rte blanche in their legislation a n d have to overcom e no resistance from entrenched interests. A th ird su rp rise is th e su d d en suggestion th at there should be both a w ise legislator a n d a young b u t disciplined ty ran t. In the rest o f the L a w s P lato w rites as th o u g h th e w isdom needed to discern w hat laws a re rig h t and th e political pow er needed to p u t them into effect will b o th be em bodied in one m an. H e talks o f the legislator as though he w ere all-pow erful. N ow he suddenly separates the two, b u t not for long. In 711d-712a he suggests th at wisdom and power may som etim es be com bined in th e sam e m an. T h ereafter the young tyrant disappears as quickly as he arrived. C haritably, one m ight suppose that h e was in tro d u ced in o rd er to make th e point th at w isdom and power are not th e sam e; or perhaps Plato believes that, while a young m o n arch will usually be too im m atu re for w isdom , an old one will be set in his ways. B ut neither o f these points is m ade in the text. I t is difficult to escape th e conclusion th a t Plato is here recycling m aterial from an earlier p eriod. T h e passage reeks of the Syracusan affair w ith P lato him self cast in th e role o f the wise legislator and D ionysius as the young ty ran t. In its context it is largely irrelevant, th o u g h it does make one or tw o im p o rtan t points, particularly the suggestions (a) th at th e fewer centres o f pow er there are in a city the easier it is to reform , and (b) th a t pow er is in fact m ost widely distrib u ted , n o t in dem ocracies, b u t in oligarchies.


The second-best sta te

In 712b-715d atten tio n shifts from the conditions necessary for successful legislation to th e form o f constitution w hich should be established. I t is here, as we have seen, th at Plato affirms m ost strongly his d octrine o f th e sovereignty o f law. I t is devotion to the law on the p art o f th e rulers that distinguishes a true constitution from a mere faction state. Law is itself th e em bodim ent of reason, the im m ortal elem ent w ithin us, a point w hich is reinforced by the story o f the age of Cronos. T h is assertion o f th e prim acy o f reason dem onstrates that, so far as th e ir m ost fund am en tal p rinciples are concerned, there is not m uch difference betw een the R ep u b lic and the L a w s. In each the ultim ate goal to w hich any co n stitu tio n -m aker m ust aspire is unity and harm ony u nder the control of reason. In the R epu blic this reason is em bodied in the persons of th e guardians, b u t the L a w s reinforces the



point made in the S tatesm an : there is no possibility in practice of obtaining wise rulers among men. T he direct rule o f fully rational beings would be attainable only in another age where conditions are quite different from those we know. W isdom m u st therefore be em bodied in law. But even th e city of the L a w s can be b ro u g h t into being only where circum stances are especially propitious: it requires, among other things a wise legislator who somehow possesses the power to get his proposals put into practice. At 739a-e there is w hat looks like a m ore explicit discussion of the relation between the R epublic and theL aw i. T h e context is a discussion of the choice of citizens, th e distribution of land and of restrictions on the holding of private property. T he A thenian rem arks th at the state which he and his com panions are now describing can have only a second-best constitution. H e expects this claim to com e as a surprise, though reason and experience will dem onstrate its inevitability (739a). The best city would be one in which all things are held in com m on. This w ould be possible only among gods or children of gods b u t it is still the paradigm , or standard, o f a constitution. T h e constitution now under consideration will be closest to the paradigm in im m ortality and second in m erit. We may also go on to consider the th ird best constitution (739a-e; cf. 853c). Since in the city o f the R epu blic the guardians practise com plete com m unity o f property, it looks as though we are here being told th at th at city is an ideal b u t n o t a serious possibility for ordinary hum an beings. A lthough the prom ise to describe th e third best constitution rem ains unfulfilled, it is not difficult to see why such a constitution m ight be required. In founding his im aginary state the Athenian assumes the best conditions likely to occur on earth b u t he is conscious that most legislators will not be so lucky. F ounders of bran d new cities are fortunate in that they do not encounter quarrels over land or property. T hose legislating in different conditions will have to be content with gradual change (736c-d). I f the population o f a city becom es depleted by war or disease, there may be no alternative but to take in ill-trained citizens from elsewhere (741a). It would be desirable for all citizens to arrive w ith equal property, but this is im possible in practice. The A thenian’s suggestions for the num erical division of the citizens can be observed only so far as possible (747e). T he general position is sum m arized in an im portant passage at 745e-746d. One m ust not suppose that a legislator will ever encounter such favourable circum stances that all the proposed arrangem ents can be applied in all their details. F or exam ple, they presuppose th at the citizens will willingly accept the proposed restrictions on property and



arrangem ents for the d istrib u tio n of land, they will also require the rig h t topography for the city and surrounding countryside. Does this m ean that the A thenian’s proposals are as unrealistic as dream s or m odels m ade from wax? T h e reply the A thenian offers on behalf o f his legislator is significant: ‘In considering program m es for fu tu re action it is only right to om it noth in g of w hat is tru est and best. W here one of these suggestions tu rn s out to be incapable of realization one need not a tte m p t to p u t it into practice. Instead one should try to bring about w hichever of th e rem aining possibilities is m ost closely related to the m o d el’ (746b-c).


P la to 's purposes

W e have seen th a t, according to th e L a w s, neith er the direct rule of philosopher kings nor com m unity o f property are possible am ong men as they actually are. T h u s the ideal state of the R epu blic could never in practice be attainable. T his m ay m ean that Plato has changed his m ind in th e direction of g reater realism , b u t it is equally possible th a t the R ep u b lic and the L a w s represent different kinds o f ideal. T h e citizens of th e R epu blic are recognizably hum an and live in a w orld like our own. B u t th eir rulers are m uch m ore rational than we could ever be, and the lower classes exhibit a quite unrealistic willingness to follow the g uidance of th e ir b etters. So in the R epu blic we are presented w ith a p ictu re o f how m en w ould live if they were capable o f organizing their lives w ith total rationality. In practice, no hum an com m unity has this capacity, b u t th a t does not m ean th at the R ep u b lic lacks any practical relevance. We can aspire to th e sort of political harm ony it exhibits even if we know th at we can never reach th at level. T his seems to be recognized in th e L a w s. We can im itate the rule of divine reason by su b m ittin g to th e ru le o fla w (7 1 3 e -7 14a); we can try as h ard as we can to find a form o f society w hich resem bles the com plete com m unity of th e first best city (739d-e). T h e city of the L a w s is m uch m ore closely related to practical experience th an th a t o f the R ep u b lic, b u t even it is not a b lu e p rin t for political action. As we have seen, the A thenian acknowledges th at it will not be possible to enforce his recom m endations to the letter. I t is m ore accurate to regard the im aginary city as a m odel based on the assum ption th at the conditions o f its foundation are as favourable as any likely to be fo u n d on earth . Since no h um an legislator is likely to en co u n ter conditions w hich are com pletely favourable in every respect, it is unlikely th a t th e proposed constitution could ever be put



into effect in all its details. B ut this does not prevent the L a w s being a practical guide to legislation. In teaching any kind of technique we usually begin with illustrations, often imaginary ones, in which everything goes right. In reality things seldom go com pletely to plan. We have to use our judgem ent in adapting the models we have learnt to the vicissitudes of ordinary life. But this does not prevent the use o f the model as a guide to action. Plato seems to have seen the city o f the L a w s in this light, as a m odel, not a bluep rin t. W h eth er it is a good m odel is, of course, open to dispute, but it does seem fitted for the role in the sense that m ost of its features could be adapted to the conditions prevailing in real cities. M any o f them in practice have been. Popper (1966, ch. 9) contrasts Plato as a utopian w ith what he calls the ‘piecemeal social engineer.’ So far as the L a v s is concerned this charge is unfair; its approach is neither clearly utopian nor clearly piecemeal. It is not, as Popper (p. 157) claims, a blueprint to be imposed on society; it leaves room for adaptation to the different needs of different com m unities and stresses the need for the legislator to use his judgem ent in dealing with the particular situation confronting him; it recognizes that on many occasion small-scale changes will be the only possibility. On the other hand, Plato w ould not think that piecemeal engineering was sufficient in itself. He stressed in 630d632d that the legislator m ust direct his activities systematically to a single goal. Certainly he overdoes this: a reform er does not need a complete program m e worked out in every detail. But he does need some kind of overall view o f what he is trying to achieve. W ithout this he would have no way of deciding his priorities or of ensuring that different parts of his legislation did not counteract one another. For example, th ere w ould be no· point in a g o v ern m en t’s reform ing housing law in order to prom ote home ow nership if its taxation policy made this more difficult. Attacks on utopian social engineering rest, in part, on the belief that large-scale schemes are liable to be m isdirected or to collapse under the weight of the bureaucracies they generate. T his may well be tru e if we think of attem pts to transform radically a m odern nation state. But, of course, Plato is concerned with the politics and institutions of city states which would seem tiny by our standards. In th e Greek world the opportunities for making a fresh start and for doing so successfully were m uch greater than in o u r own. T h e situation envisaged in the L a w s , where a brand new state is to be founded, would by no means have been unique. On occasion, established states also would adopt new constitutions. There is a tradition that Plato him self was invited to propose laws for the newly founded city of M egalopolis. T here are



som e parallels betw een the circum stances of this foundation and those envisaged in the L a w s. A num ber o f oth er G reek cities are also said to have sought legislative assistance from the Academy (M orrow , 1960, p. 8, n. 11). T h e L a w s m ight well be in tended as a handbook for such activities. O ne should n o t assum e th at it should be regarded in the sam e light as a m odern Utopian’s dream of the future. P o p p er com pares Plato w ith M arx, n o t only because he sees b o th as advocates o f large-scale social engineering, bu t also because he regards th e m as historicists, i.e. as believing th a t history has its own laws which g o v ern hum an developm ent. In P lato ’s case this charge has to be based o n th e R ep u b lic, Books V III and X I, w hich describe a progression of co n stitutions. M ost scholars regard this section as m etaphorical. In the L a w s , Book I I I , th ere is a general account o f world history b u t there is no a tte m p t to extrapolate from th e past to the future. T he new city will be th e outcom e, not o f a process o f historical developm ent, b u t of d elib erate h u m an intervention. I t is a product of skill operating on chance factors u n d er th e general guidance o f G od. Plato is not in the b usiness of predicting th e future course of history.


Society 704a-708d, 734e-747e, 771a-778a, 779d-785b, 804d-808c, 828a-e, 831b-832d, 842b-850d, 914e-915c, 918a-920c; Re p u blic, 369b-374e, 412b-424d, 427d-434d, 449a-466d, 543b-576b

L a ïc s,


The problem

The driving force of Plato’s social philosophy is his conviction that Athens in his day was a sick society and th at its m alaise could be cured only by a radical transform ation of the social fabric. In the L a w s, as in the R epublic, he stresses his view o f A thenian dem ocracy as lawless and devoid of moral restraint (700a-701d; cf. R ep u b lic , 555b-561e). In the L aw s this condition is attrib u ted , at least in part, to the lack of control over music. W hereas in its golden days Athens adhered to strict rules of m usic, now anything which earns the applause o f the m ultitude is praised; this lawlessness is liable to infiltrate all other areas of the city’s life. T h e aim of the L a w s is to re-establish respect for law as a fundam ental requirem ent of a healthy society. T h e lawlessness which troubles Plato is not only a m atter o f external behaviour; it perm eates the soul o f the dem ocratic m an who is driven this way and that by his unregulated desires (R ep u b lic, 560e-561e). Similarly, the G orgias contrasts Callicles’ ideal o f the man who seeks to gratify his appetites w ithout restriction with the law and order enshrined in Socrates’ conception of the just m an (491c-495b, 503d508c). As we have seen, the same em phasis dom inates the early books of the L a w s : lack of self-control is the root problem for all existing com m unities, including the Persians, the A thenians and even some of the Dorians. T he first principle enunciated by the legislator in his prelude is the requirem ent to honour one’s soul, n o t by the indulgence of desire, but by obedience to law (726a-728c). T his is the goal o f the laboriously discussed drinking parties and th e rest of the educational system. Plato connects the lawlessness and self-indulgence of contem porary Athens with economic causes. T h e simple city o f the R epu blic becomes



‘fev ered ’ w hen new trades are introduced to pam per the desires of the citizens (372e-373e). M o st of the social and economic changes d escrib ed in the R ep u b lic, V I I I and IX , are instigated by econom ic factors. M oney and desires go together: indeed Plato calls the appetites th e ‘m oney-loving’ elem ent in th e soul (580e). T h u s there is a sharp o p p o sitio n betw een th e love o f m oney and Plato’s own ideal of law fulness: ‘it is im possible to honour riches and at the same tim e to have adequate self-discipline (sophrosurte ) am ong the citizens’ (555cd ). T h e L a w s also attrib u tes m uch of th e contem porary malaise to the d e sire for gain (7 0 5 a-b , 714a, 727e-728e). But P lato’s belief in the econom ic basis o f social d isorder is m ost d ea rly indicated by his p ro p o se d rem edies; he places severe restraints on the capacity o f the citizens to own p roperty and accum ulate money. A ssociated w ith the supposed lawlessness of contem porary society is its disunity. T he purpose o f the co m m unity o f property and abolition o f th e family am ong the R e p u b lic ’s guardians is to make them feel as o n e , for d isu n ity is th e g reatest evil the state can en counter (4 6 le 466d). W e have seen how th is concern for internal peace, unity and harm ony form s one o f th e guiding principles of the L a w s. I t is hard to know how m uch justice there is in Plato’s criticism s of A th e n ia n society. B ut th e p ictu re he paints has m uch in com m on with w h at m any people have felt about m odern societies in which there is a h ig h degree o f personal freedom . T hey say that there is a lack o f firm , generally accepted stan d ard s, th a t people live only for the pleasure of th e m om ent, th at they value financial gain above everything else, that th e re is no sense of belonging to one another, no sense of com m on responsibilities or o f com m on concerns. One m ight dism iss these com p lain ts as reflecting th e conservatism o f m iddle age or the cow ardice o f those w ho, unable to live w ith freedom , seek the security o f a p atern alist au th o rity . B ut p erhaps there is m ore to it th a n that; th in k ers o f m ore th an one political persuasion have diagnosed the co n d itio n o f m odern m an as one o f alienation. It is w orth exploring the parallels betw een these ideas and P lato’s com plaints about contem p o rary Athens. M a rx ’s concept of alienation is rooted in his analysis of the capitalist m ode o f production; it would be anachronistic, therefore, to see P lato’s analysis o f the ills o f A thens as foreshadow ing M arx. On the other h a n d th ere may be parallels betw een some of the phenom ena w hich A larx saw as sym ptom s o f alienation and some o f the features of contem porary society th at troubled Plato. In particular, both com p lain o f the lack o f a sense o f com m unity and of th e preference for individualistic and com petitive, rather than for co-operative modes o f



activity. Perhaps we should also consider here Plato's idea th at m anual labour and trade are degrading, banausic. Plato is totally unsym pathetic to the manual worker, but he shares w ith the M arxists a belief that to labour for wages is incom patible with the true dignity and self-fulfilment of a hum an being. T here may be a closer parallel w ith D u rk h eim ’s conception of anomie. According to D urkheim , men in a state o f anom ie lack any feeling o f constraint upon their desires. T h e grow th o f individualism has replaced the conscience collective, that corpus o f com m on sentim ents and shared beliefs which was characteristic o f earlier societies. M en no longer have a sense of belonging, o f being integrated into th e com m unity. P lato’s picture too is of individuals left w ithout any guidance, external or internal, apart from th e ir own desires, having n o firm standards and no sense o f belonging to a larger whole. It looks very m uch as though D urkheim d etected in nineteenthcentury France some o f the features w hich Plato th ought he saw in fourth-century Greece. T his condition of anomie can be painful: it is hard to go w ithout the security and sense of certainty w hich our forefathers apparently knew. But it is not clear how we should react to it. D urkheim looked to the developm ent of organ ic so lid a rity , i.e. a system of functional interdependence based on the differentiation of social roles. At times Plato seems to have w anted simply to p u t the clock back. B ut perhaps we should regard anom ie as the price we have to pay for freedom . After all, the m odern individual is free precisely because he is not bound by traditional constraint. 2

P la t o ’s remedies

T h e city of the L a w s is intended to rem edy the defects o f contem porary society by creating unity under the control o f a firm ly accepted code of law and morals. T his purpose will be met in p art by the adoption of a new legal system and new style of lawgiving and by the creation o f an elaborate educational system. These are discussed elsewhere. The other rem edies suggested by the Athenian include: (a) population controls, (b) th e selection of the right physical environm ent for the new city, (c) controls over economic life, as well as (d) measures designed to foster feelings of unity and friendship through the family and other social groups. Population controls In view of the allegations that Plato is a racialist or advocate of a caste



system , one m ight expect a close atten tio n to the genetic m ake-up of th e citizen body, b u t, in fact, there is surprisingly little on this point. T h e new C retan colony will draw its in h ab itants from all over G reece, a lth o u g h D orians will be particularly welcome. T h e A thenian co m m en ts th a t this will be a disadvantage from one point of view b u t an advantage from another. C olonists draw n from different cities will n o t have th e same u n ity o f feeling th a t colonists from the same city w o u ld have had; on th e oth er hand, it will be easier to im pose new cu sto m s on th em th an it would be if they all came from the same back g ro u n d sharing a com m on tradition (707e-708d). L ater he draws a co m p ariso n betw een th e legislator and a stock breeder who m ust p u rg e his herd s of inferior stock if they are to prosper (735b-c), b u t the pu rg es he proposes are less drastic th an one m ight expect: sentences of d e a th or exile for the m ost serious crim inals and the sending out of new colonies w hen th e com petition for scarce resources threatens to cause civil strife (735d-736a). In the case o f th e new C retan city, o f course, su c h m easures will not be necessary; we have to select citizens from d iffe re n t sources and m ingle th em as we m ight w aters from different sp rin g s aim ing to keep th e m ixture as p u re as possible (736b-c). How a n d on w hat criteria this selection is to be done is not entirely clear: S au n d ers translates th e relevant passage (freely b u t intelligibly), ‘w h en we have screened th e bad candidates over a suitable period and g iv en th em every chance to be converted, we can refuse their ap p licatio n to en ter and becom e citizens of th e state; b u t we should g re e t th e good ones w ith all possible courtesy and kindness. ’ W hatever th is m eans, it seems th a t th e selection is on m oral rath er th an on racial g ro u n d s and th a t dub io u s candidates are to be given the chance to m ake good (736b-c). T h e A thenian is m ore concerned ab o u t the size o f the population. T h e ideal n u m b er of households is 5040. T his num ber is chosen because it is divisible by all num bers betw een 1 and 12 except 11 and is therefo re a very convenient num b er if one wants to divide the state into g ro u p s o f different sizes for different purposes (737e-738a). But p resu m ab ly these m athem atical considerations are n o t the only ones P lato has in m ind. H e m ust believe th at around five thousand househ o ld s will produce a state sm all enough to have a real sense of u n ity and to perm it decisions to be m ade and officers to be appointed o n th e basis o f a real personal know ledge and at the same tim e large e n o u g h to bestow the full benefits o f civilized life (738a-e). Various devices are available to m aintain the population at a constant level; th e se seem to consist largely o f m easures to persuade citizens to m ain tain a suitable rate o f reproduction. I f these fail it may be



necessary to send out a colony to get rid of excess citizens (740b-741a). It is w orth noting that no mention is m ade of the practice o f exposing infants (Plato’s apparent acceptance of this in th e R epu blic has been the source of a good deal of m odern criticism). Geography The A thenian is concerned about the geography of the new colony. He would have preferred it to be far from the sea, or at least to have no harbours; at any rate the city should not be established on the coast. It is a good thing that the countryside will produce all th at is necessary for daily life, b u t not in such abundance as to leave a surplus for foreign trade; it is also fortunate that the prerequisites for shipbuilding are not abundant (704a-705c). T he key points here are P lato’s hostility to trade and to sea power. The trouble w ith trade is th at ‘it fills the land w ith w holesaling and retailing, breeds shifty and deceitful habits in a m an ’s soul and makes the citizens distrustful and hostile, not only am ong them selves b u t also in their dealings with the world outside’ (705a, trans. Saunders). Presum ably th e poin t is th at those who are in m arket relations to one another act out of concern for cash values rather th an for law and morality and see one another as com petitors or as objects for exploitation rather than as fellow citizens or brothers. Elsewhere the Athenian seeks to lim it contacts w ith other states on the grounds that they introduce undesirable foreign habits (949e-950a). T he objection to naval suprem acy is that those who fight at sea see nothing w rong with retreating rather than bravely facing the enem y. Sea pow er has a bad effect on character because it relies on cunning and stratagem rather than on th e virtues o f courage, endurance and self-control. M oreover, the credit for naval victories goes not to the bravest soldier but to the skill of sailors who include men of all kinds w ith no particularly com m endable qualities. T here is an echo here of the oligarchic idea that A thens’ m aritime success put political power into the hands o f the poorest classes, which furnished the oarsm en, b u t that idea is given a new tw ist, in that naval tactics them selves are seen as the source o f m oral corruption (706a-707d). Economic organization Agricultural land is to be divided into lots of equal value so that there is one lot per household (763d-737d, 739e-741a). T he lots and the



e q u ip m e n t needed to work th em will be inalienable (741a-e), and the p ro p erty held by any citizen will be lim ited to a value four tim es that of th e lot. T h e citizen body will be divided into four classes, the lowest ow ning only th e lot and its e q u ip m en t, the higher classes having respectively tw o, three and four tim es this am ount (744b-745a; see S aunders, 1961). Fam ily law is directed to ensuring (a) th at there will always be a male heir (adopted if necessary) for each lot, and (b) that no m ale citizen will be left w ithout a lot. T hese m easures ensure th at citizens of the new state will all basically be farm ers. W hat is not clear is w hether they will work the land them selves or w h eth er they will e n tru st it entirely to slaves. At first sight, 806d looks decisive for th e latter view: the A thenian describes th e leisured life o f th e citizens who have given th eir farm work out to slaves. T h ere are, how ever, a nu m ber o f considerations pointing in the opposite direction: (a) th ere is no express prohibition of agriculture by citizens; (b) th e agricultural laws o f 842b-846a are explicitly (842d) d esigned for citizens who will them selves be engaging in agricultural labour (M orrow 1960, pp. 531-2); (c) the m easures for the election o f th e Council (756b-e) apparently presuppose that frequent attendance at the assem bly will be especially burdensom e to the lower classes, presum ably because they have to work; (d) as A ristotle (P o litic s , 1265a 15) points out, an enorm ous territory would be needed to support 5040 unproductive families. I t seems likely therefore that the m ajority of the citizens are expected to be working farm ers, and that m ost citizens will be m ore like prosperous peasants th an landed aristocrats. C itizens will be excluded from taking any p art in trade or m anufacture (846d-847b, 915d-920c); they will be prohibited from doing so even indirectly th ro u g h their slaves. Essential m anufactured goods will be provided by resident aliens (849b-d, 919c-920c) whose period o f residence will norm ally be lim ited to twenty years (850a-b). L ending at interest will be prohibited, as will the holding of gold, silver and ‘h a rd ’ currency, th ough there will be a local currency for use w ithin th e city (742a-c). T h e im port o f luxuries is prohibited, though the state will see to the im port o f essential m ilitary supplies (847b-d). T h e effect o f these regulations is to keep the com m ercial activity of the city to an absolute m inim um and to see that, at all costs, citizens take no p a rt in it. T h e reasons for P lato’s disapproval o f trade are not very explicit. In part, no do ubt, it reflects a general G reek a ttitu d e — a n u m b er of states did p ro hibit citizens from engaging in trade or m anu factu re, and it was generally regarded as undignified and unw orthy of a free m an (919d-e). A thens was unusually tolerant o f these p u rsu its. P lato’s m ain objection seems to be th a t com m erce



makes possible the unlim ited p u rsu it ofgain (7 3 6e, 741e, 847d, 918d). He is not against all gainful pursuits; external goods do have a value though they come below those of m ind and body. Even farm ing can be pursued for gain, though the am ount a citizen could m ake from his lot would obviously be lim ited. Presum ably the greed w hich com m erce permits can make us forgetful o f more valuable things and, no d o u b t, it appeals to the less rational elem ents in th e soul, thus destroying the discipline on w hich Plato places such a high value. B ut it is not only wealth that can corru p t, b u t also poverty,· since th at forces us into shameless and unw orthy ways of life (9 l9 b -c): Plato thus recognizes that commerce which perm its relative wealth m ust also perm it relative poverty. T he fullest account of P lato ’s attitu d e to trade is at 918a-920c. T here th e A thenian argues th a t trade, properly understood as the redistribution of goods where they are needed, is a laudable thing. But it is abused because few can resist opportunities for excessive m oney­ making. F o r exam ple, innkeepers who set up in rem ote areas treat travellers as prisoners to be held to ransom , rather than as honoured guests. T he objection here seems to be as m uch to the exploitation of one’s fellows as to excessive profit-m aking. The family and other social groups In the R epublic the existence o f distinct social groups w ithin the state is seen as a threat to its unity; th is is why the family is abolished. T he L a w s follows the opposite policy. T he state is seen as a union o f households, not as a collection o f individuals (M orrow I960, p. 118). M en are required to m arry betw een the ages of thirty and thirty-five, and m uch legislation is designed to strengthen the family bond and to ensure th at families do n o t die out. T here are family shrines and family courts. T h e hom icide legislation is based on the principle th a t the dead m an’s relatives m ust see th at he is avenged. Crim es against parents are punished w ith particular severity. T he citizen body is divided into twelve tribes, each w ith its own territory (745b-d). T here will apparently be twelve local subdivisions in each tribe, corresponding to the Attic demes (771b; cf. 746d). Each tribe and subdivision will have its own m onthly religious festival. These have a social purpose: they enable the citizens to get to know each other well (738d-e, 771d;cf. 828a-d). T here is a passing reference to phratrvs, which in Athens were groups o f associated clans (746d). T he A thenian takes it for granted th a t th e citizens o f the new state will continue the C retan and Spartan practice of eating common



m e a ls. H e also proposes th eir extension to w om en (780e-781d). T he d e sira b ility of this in stitu tio n is taken so m uch for granted th at the A th e n ia n sees no need to argue for it in the case of m en. T h e reason for e x te n d in g it to w om en is to prom ote respect for law and custom in w h a t we m ight norm ally reg ard as private life. Its purpose seem s to be p a r tly to in d o ctrin ate th e citizens by prom oting a com m on ethos and p a r tly to em phasize th e social being o f the citizens rather th a n their liv es as individuals. T h e com m on meals are, o f course, a feature o f the R epu blic as well as o f th e L a w s , b u t the acceptance o f th e fam ily and of tribal organization is a new d ep artu re for Plato. It is clear th a t they are not merely to le ra te d as unavoidable evils b u t w elcom ed and encouraged as things o f great value. N o explicit reason is given for this b u t im plicitly Plato seem s to be acknow ledging a poin t m ade in m ore recent tim es by D u rk h e im and others. T h e m ore closely individuals are integrated into social groupings, th e less they are likely to suffer from the feeling of roo tlessn ess and lack o f co n strain t w hich D urkheim calls anom ie. M e m b e rsh ip o f sm aller social groups m ediates our loyalty to and re s p e c t for the state. P a rt o f w hat belonging to a a country m eans to us is o u r m em bership o f sm aller groups w ithin a larger whole. T he p ro p o sal in the R ep u b lic to prom ote unity by abolishing these smaller g ro u p s is superficially attractive b u t quite m isguided, given m en as th e y actually are. I t is not clear w hether the recognition of this in the L a w s is due to a new aw areness on P lato’s part or w hether it is sim ply a co rollary of th e decision to legislate for m en as they actually are in the real w orld rath er than for those in ideal circum stances.


The p osition o f women

O ne of th e m ore notorious proposals o f the R epu blic is that, so far as the g u ard ian s, at least, are concerned, w om en should be treated equally w ith m en. T his proposal had a corollary in the suggestion for the abolition of the fam ily. T h e positive em phasis given to the family and th e generally m ore cautious tone of the L a w s preclude such drastic proposals. T h e A thenian still w ants to give wom en a m ore prom inent role th a n they could norm ally play in fo u rth -cen tu ry A thens, b u t it is n o t clear how far he is prep ared to go in this direction. Since the new city will be com posed of families rather than of separate individuals, it is assum ed th at m arriage will be the vocation of all w om en, th o u g h , as an innovation, m en will also be required to m arry betw een th e ages o f th irty and thirty-five (721b-d; b u t see



722d). As in Athens an unm arried woman will be legally under the guardianship of her father or nearest male relative, a m arried one under that o f her husband. But there are some smail m oves tow ards recognizing the legal personality of women. Those over forty will be able to act as witnesses in courts, and widows or divorcees will be able to initiate legal actions in their own right (937a). D ivorce law is also more favourable to women than in Athens (929e-930b), and it is envisaged th a t in certain very lim ited circum stances an heiress may be able to choose her own husband (925a-c). These measures would, at best, constitute m odest im provem ents in the status of women. M ore striking proposals are that they should bear arms on behalf o f the city and should receive the sam e education as men. T he argum ent here is th a t a state w hich fails to assign its wom en any m ilitary function is gratuitously weakening itself, since women constitute h alf th e population. But if women are to serve in the arm y they m ust also receive m ilitary training and take p a rt in athletic contests (804d-806d, 833c-d). W om en also have an im p o rtan t effect on the m oral ethos of the state. They should therefore receive the same moral training as m en and should have their own com m on meals (780c-781d). T his last suggestion would be an innovation: even in Crete and Sparta, the com m on meals were confined to m en. In making these proposals the A thenian assumes that women are in general inferior to men in respect of virtue, b u t he gives them special credit for orderliness and m oderation (802e). Since these virtues are central to the m oral doctrine of the laws, his attitu d e here seems barely consistent. It is not clear what political rights will be granted to wom en. At 785b it is suggested th a t wom en should be eligible for office from the age of forty (the corresponding age for men is thirty). M em bership of the assembly is open to those who have borne arms ;753b). It looks therefore as though older w'omen, at least, would be eligible for the full range of political offices. This is what one would expect in the light of the argum ent for wom en’s participation in the arm y. It would, one supposes, be just as serious a waste of hum an resources to exclude women from office as it would be to exclude them from bearing arms. Against this one may point to the fact th at the A thenian nowhere explicitly m entions female holders of the main offices of state. (This may not be very significant, since, in G reek, m asculine gram m atical forms can ‘em brace the fem ale’.) He does, however, refer occasionally to women who will act as supervisors of marriage and fam ily life (784a, 930a). It could be these he has in m ind when he speaks of women being eligible for office from the age of forty.



I : is difficult to see how one could decide betw een these two in te rp re ta tio n s. I t is possible th a t the am biguity is deliberate. Plato m a y realize th a t his argum ents com m it h im to the equal treatm ent of w o m e n and yet be relu ctan t to suggest openly such a bold change. B ut e v en if one supposes th at he does w ant w om en to hold political office, it w o u ld be w rong to classify h im as an early exponent o f w om en’s lib eratio n . T h e tru th is th at he was not interested in liberating anyone, m ale o r female; liberty, in the sense o f th e o pportunity for individuals to choose their own way o f life, is not one o f his values. T he fu n d a m e n ta l aim of th e proposals for the position of women in both the R e p u b lic and th e L a w s is th e full utilization of th eir abilities for the go o d o f the state as a w hole. Since w om en constitute half the p o p u la tio n , to neglect the education o f w om en is to risk the corruption o f th e whole com m unity. T o exclude th em from bearing arm s is to re d u c e substantially the m ilitary stren g th of the city. It follows that w om en m ust be educated and receive m ilitary training. 4

S la v e s a n d aliens

S ince m ost form s o f econom ic activity are forbidden to citizens, the n ew city will d ep en d heavily on the labour o f slaves and o f resident aliens or m etics as they are generally know n. O f these, the slaves will be u se d in three m ain capacities: (a) as the dom estic or agricultural serv an ts of citizen households; (b) as slaves o f metics engaged in trade or m anufacture; (c) as public slaves of the city used (e.g.) in the m ain ten an ce of public works. I t has som etim es been supposed th at the agricultural slaves will really be in a condition o f serfdom like th a t o f the Spartan H elots, a su b ject people held in p erm an en t subjection tied to the land. T he evidence for this view is at 806d w here th e A thenian describes the leisu red life of citizens whose ‘farm s have been given over to slaves’. T a y lo r translates this ‘th eir farm s let o u t to villeins’, b u t there is n o th in g to justify th is w riting o f th e feudal system into P lato’s text. As we have seen, it is likely th a t th e citizens w ould them selves be running th e ir own farm s using slaves as servants. T he A thenian him self is h ig h ly critical o f th e S partan system o f H elotage (776c). T he objection is, app aren tly , th a t the H elots, because they form ed a coherent class w ith a com m on language, offered a p erm an ent threat o f rebellion. T he A th en ian advises th at so far as possible slaves should come from d ifferen t countries and speak different languages (777c-d). In general, the A thenian regards slaves as sim ply a form of property (776b, 777c), as things to be used. M an is a particularly difficult kind of



animal to handle; one should, therefore, avoid treating slaves harshly, because that will .make them m ore troublesom e. On the o ther hand citizens should not live on fam iliar term s w ith slaves or a tte m p t to argue with them . One m ust always use the language o f com m and (777d-778a). T his attitude is borne out in the detailed legal provisions, particularly o f th e penal code. In m ost cases they are liable to different penalties from those imposed on free m en, usually some form o f capital or corporal punishm ent. T he A thenian’s m ain concern seem s to be to stamp on any act o f insubordination which m ight th reaten the privileged status o f the citizens. F or example, a citizen who kills a slave in anger is required only to purify him self for the p o llution he has incurred and to com pensate the slave’s ow ner for his loss. By contrast, a slave who kills a citizen in anger will be handed over to the relatives of the dead m an; they may treat him in any way they w ish, provided th at they do not leave him alive (868a-c). A free m an who kills in selfdefence will undergo no penalties; a slave who kills a free m an in selfdefence will be executed as though guilty of parricide (869c-d). Morrow (1939, 1960) has shown that where P lato’s proposals differ from contem porary A thenian practice they almost always do so in the direction of greater severity tow ards th e slave. Penalties are more severe, and slaves are deprived of alm ost all legal p rotection against wrongs and assaults suffered at the hands o f citizens. One m ight try to excuse Plato by arguing that, living in a slave society, he took the existence of slavery for granted. P erhaps, one might say, future generations will feel as horrified by (e.g.) our practice of eating m eat as we are by the institution of slavery. A nd no doubt those who were born or sold into slavery quickly acquired servile habits; it m ust have been easy, therefore, to believe th a t they were by nature inferior. Popper (1966, pp. 222,224,236) counters this defence by claiming th a t some o f P lato’s contem poraries saw th at slavery was unjustified and advocated its abolition. T here is very little evidence for this. One of the few passages which Popper cites in support of his claim is Euripides, Ion , 854, where it is said that a loyal slave can be as good as a free man. Plato him self recognizes this in 776d-777a, though he balances this adm ission w ith a reference to those who say th at slaves are inherently inferior. In any case he does not question or attem pt to mitigate the rigours of slavery; rather he attem pts to stren g th en the institution. W hat strikes us most in Plato’s treatm ent of slavery is his lack of hum anitarian principles. T his is in line with his general attitude. He does not see th e fact of a shared hum anity as creating any dem and for respect or sym pathy. T he slave is not an end in him self b u t a m eans,

1 08


a n d P lato consistently treats him as such. T o concede him political or legal rights w ould be inconsistent w ith this perspective. So there is a r u th le s s logic ab o u t Plato ’s proposals. Even so, one m ight expect him to have rath er m ore d oubts about slavery th an he actually shows. One o f th e guiding principles o f th e R ep u b lic is th at people should be used in th e role for w hich they are best suited by nature. T his principle is s till ev id en t in th e tre a tm e n t of w om en in the L a w s. It follows th at if o n e is to justify using freem en an d slaves in radically different ways, o n e should show th at they have radically different natures. Plato does n o t a ttem p t to do this. By com parison w ith his attitu d e to slaves, the A thenian’s treatm ent o f th e m etics, or resident aliens, seems very favourable. T h e presence o f these people in th e new city w ould not be unusual. M etics played an im p o rta n t p art in th e life o f m ost G reek cities, w ith th e notable ex cep tio n o f S parta. T h ey w ere particularly prom inent at A thens, the m ain com m ercial centre o f G reece. T h e A thenian proposes th at their residence in th e new city will be subject to two m ain restrictions: (a) th e y will not be able to stay longer than tw enty years unless they receive special perm ission (850b-c); (b) they will not be able to accu m u late w ealth in excess of th a t held by the th ird pro p erty class (91 5 b -c). In these respects they would be worse off th a n the metics of A th en s, some o f whose fam ilies had been resident for several g enerations and m any of w hom were wealthy. T he A thenian’s aim is clearly to p rev en t the developm ent o f a class o f non-citizens w ho were, nevertheless, p erm an en t residents and regarded the city as their home. T h e restrictions on w ealth w ould p revent them acquiring great political or social im portance as well as preserving the status of the citizens. On the o th er h and th e m etics will be exem pt from m ilitary service and taxation (850b-c). T hey will enjoy full protection of the law . O verall their situation will n o t be at all undesirable. In m any ways th e y will be the freest group w ithin the population. T his is not due to an unw onted access o f liberalism on P lato’s part. T he city needs the m etics and they have to be persu ad ed to come voluntarily, so their position has to be reasonably attractive.


C lass distin ction s a n d the d ivisio n o f labour

T h e principle th a t different kinds of task should be undertaken by d ifferen t kinds o f people is fundam ental to the R epu blic where it gives rise to the threefold class system o f guardians, auxiliaries and fa rm e rs/a rtisan s. In th e L a w s this principle is applied in two main



ways. Firstly it is the basis on which citizens are pro h ib ited from engaging in m anufacture or trade; the craft of the citizen, to m aintain the good order o f th e state, is a full-tim e occupation (846d). Secondly it requires that the metics who work as artisans should be forbidden from practicing two different trades. Some m odern critics have taken the sharp distinction betw een the citizens on th e one hand and the slaves and metics on the o ther as evidence th a t th e city of th e L a w s will be based on class divisions. Tw o different lines o f argum ent are possible here. (1) Saunders (1962) has argued that there are close parallels between the R epublic and the L aw s. T h e guardians o f the laws, he claims, correspond to the guardians in the R ep u b lic , the bulk o f the citizens to th e auxiliaries and th e slaves, and m etics to the farm er/artisan class. But although there are some resem blances here, the differences are at least as striking. In the R ep u b lic, fighting and ruling are seen as distinct tasks and both are m arked off sharply from farming as well as from trade and m anufacture. F arm ing here includes not only agricultural labour b u t also the ow nership of land. T he guardians are forbidden to own property because th e n they would become farm ers and householders. By contrast the citizens o f the L aw s share in all three kinds of function. T hey have a part in governm ent, they m ust serve in the arm y and they m ust own land. It is possible that the difference between th e R epu blic an d theLazw is due simply to the fact th at the L a w s is m ore closely tied to practicalities. Plato recognizes th at the sharp divisions proposed in the Republic could never become a reality, and therefore com prom ises by treating the position of citizen-farm er as a single role. B ut the differences in th e social stru ctu re o f th e two states could also reflect different views of hum an nature. T h e R epublic is based on the assum ption that hum an beings come in three varieties, depending on w hether reason, spirit or appetite predom inates in th e soul. D ifferent social roles require different capacities and so each m an m ust be given only that role for which he is suited. Since the citizens of the L a w s are allowed, indeed required, to operate in all three capacities, it would seem that this doctrine of different kinds o f men has disappeared. T here is a corresponding difference in the way in w hich the two states are supposed to hold together. T h &R epublic exhibits an organic unity. T h e different elem ents are like the different parts of an organism or a m achine such as a clock. T he parts have quite different characteristics b u t by interacting with one another prom ote the end for which the organism or m achine is designed. T he citizens o f the L a w s, on the other hand, do not differ in their nature or function. It seems to



be th e ir very sim ilarity, their com m on tasks, th eir shared custom s and tra d itio n s w hich will hold them together. T h e class divisions of th e R ep u b lic can be criticized not only because th e y restrict individual choice and create unequal privileges but b ecau se they ex p ro p riate part of th e individual personality. W hile it is n o t tru e th at hu m an beings come neatly packaged in three varieties, it is alm o st certainly th e case that if hu m an beings were subjected to the k in d o f division of labour envisaged in th e R epu blic they would develop th re e distin ct kinds o f character. Each person w ould be able to develop o n ly a small p art o f his hum an capacities. H e would not be a full man. T h is w ould not be th e case w ith th e citizens of the Laws¡, it w ould be possible for them to live a fully ro u n d ed existence. So, while there is little trace of h u m an itarian ism in the L a w s, it is a more human work than th e R epu blic. (2) I t has been argued m ore generally that in the L a w s Plato seeks to p e rp etu ate an d reinforce th e position o f the aristocratic class to w h ic h he him self belonged. H e therefore advocates a system in which an hered itary landow ning aristocracy (the citizens) rule over slaves and aliens who lack b o th land and political rights (W ood and W ood, 1978). T h is class system is, if anything, m ore firm ly fixed that that of the R e p u b lic , since each in dividual’s position will be perm anently fixed by b ir th ra th e r th a n by th e ru lers’ estim ate of his abilities. On this view, th e n , th e L a w s is a straightforw ard p ro duct o f aristocratic reaction. P la to , an aristo crat him self, looks for a state in w hich the pow ers and privileges of his class will be unshakeably guaranteed. T h e re are two m ain difficulties w ith this interpretation: (1) M u ch o f th e evidence suggests th at the citizens will be small fa rm e rs rath er th an aristocrats. T h ere are supposed to be 5040 citizen h o u seh o ld s in a relatively sm all an d infertile area. Even if we suppose th e y will not actually engage in agricultural labour, they will certainly n e e d to supervise th eir farm s. T h ey will not be rich and there will be ra th e r a lot o f th em . T his does n o t fit the conventional picture o f an aristocrat. (2) N e ith e r th e m etics n o r th e slaves are intended to form p e rm a n e n t classes w ithin th e state. T h e lim itation on their period of resid en ce w ould p rev en t m etics from com ing to regard the new city as a p e rm a n e n t hom e and from becom ing linked w ith one another by ties o f b irth or kinship. Sim ilarly th e provision th at slaves should, if possible, be o f different origin an d speak different languages implies th a t th e m ajority of them will be purchased from abroad rather than h o m e bred . A nd, if it is tru e th a t m ost citizens will work or at least supervise their ow n farm s, m ost slaves will presum ably live as adjuncts



to citizen households rather than have homes of th e ir own. T he A thenian’s com m ents on the Spartan system of H elotage make it clear that he seeks to prevent the developm ent of the slaves into a distinct social class. T here are more general reasons why it may be m isleading to th ink of the city of the laws as a state based on class distinctions. In the R ep u b lic the m em bers of the lower classes are nevertheless part of the state. T he city is seen as an organism in which each elem ent has its own function. For the city to do well, all parts m ust perform their own task and no class has a special claim to happiness. In the L a w s, on the o th er hand, the aliens and slaves are specifically excluded from m em bership of the state; they do not have even an inferior brand o f citizenship and have no right of perm anent residence. One cannot, therefore, say th at the state is composed of three classes. Rather it consists o f a single citizen class which relies on the labour of individuals who them selves stand outside the state. T hus, whereas the classes in the R epu blic may be compared to the parts of th e body, the role of the slaves and m etics in the L a w s would be more like th at of tools or possessions w hich are not part of the man b u t are needed by him to further his purposes. If they fare badly, that does not, in itself, mean that the state has been harm ed. Conversely, in seeking the good of the whole, the legislator is not essentially concerned with the good of these supporting groups. T he difference here might not be very im portant in practice. If you are a m em ber of an underprivileged group, it may not m atter m uch to you w hether you are officially a m em ber of th e state. But this point can be im portant when we try to u n d erstan d the basic patterns o f P lato’s political thought. As we have seen, he is not given to num anitarianism . It is not the function o f his legislator to secure the salvation of mankind. But he does have an obligation to secure the welfare of the city. From this point of view, it counts very m uch w hether a particular social group is seen as p art of the city, and therefore as falling w ithin the scope of the legislator’s concern, or as som ething external which may be used or discarded as suits the interest of the city proper. Seen in this perspective, Plato is not erecting a state dom inated byclass divisions b u t seeking to abolish such distinctions by excluding from the state all those who cannot live the life he judges w orthy of citizens. His ideal is one which has appealed to m any generations in political thought, the ideal o f an agrarian com m unity in w hich every citizen is a farm er. U nfortunately this could only be achieved by a sleight of hand. T he excluded groups are sm uggled in again as slaves and metics.

11 The System of Government L a w s, 751a-768e, 855c-856a, 875d-876e, 945b-948b, 951c-952b,


A fter considering th e social and econom ic conditions of the new city, th e A thenian tu rn s, at th e beginning o f Book V I, to consider w hat we w ould call its ‘co n stitu tio n ’. T h e plan he apparently intends to follow is set o u t at 75 la: we m ust first establish political offices and th en lay d o w n laws for the holders o f these offices to adm inister. M ost o f the m ain offices o f state are d escribed betw een 751a and 768e and the carefully m arked transition at 768e to the substantive legislation suggests that th e A thenian regards the first phase, the establishm ent of offices, as com pleted by this point. It may come as som ething o f a su rp rise , therefore, th at tw o new offices, the scrutineers (otherwise ‘ex a m in e rs’ or ‘au d ito rs’) and th e n o ctu rnal council, are introduced in Book X II. It has som etim es b een supposed th at these offices m ust be an afte rth o u g h t. B ut as early as 632c there was a h in t that some special b o d y w ould be needed to give th e state its intellectual u n d erpinning, a n d 8 1 8 a-d seems to prefigure th e n octurnal council’s program m e of stu d y . T h e in tro d u ctio n o f th e council at the very end o f the L a w s en ab les the dialogue to end on a suitable note of solem nity. T h e tidym in d e d may feel th a t, w ith th e in tro d u ctio n o f the scrutineers and the n o c tu rn a l council, th e top o f th e political pyram id becomes dangerously overcrow ded, b u t there is no inconsistency either at the theoretical or at the practical level. M ost of th e offices and institutions proposed in the L a w s have p reced en ts som ew here in th e G reek w orld, usually at A thens; so classical scholars have rightly expended effort in exploring the relatio n sh ip betw een P lato ’s proposals and the practice of real Greek cities (see especially M orrow , 1960, chapters V and V I). T hese studies are im p o rtan t for the light they cast on Plato’s m ethod of working, but, for philosophical purposes, it is m ore im p o rtan t to consider the u n d erly in g principles than the detailed provisions of the constitution. Its m ost im portant features are outlined in the Appendix.



1 M a g isteria l pow er The new state’s m ost striking departure from the practice o f fourthcentury Athens will be in the power bestowed on the m agistrates (i.e. on those elected to the principal offices of state'). In the A thens of Plato’s day, ultim ate authority rested with the A ssem bly and the large popular juries, though conservatives pined for the days w hen the Areopagus (a council of ex-m agistrates) had been the dom inant influence. In the new city the guardians o f the laws will be p re­ em inent. As their title suggests, their m ain task is to guard the laws (754d-755b). T his apparently am ounts to overseeing the other organs of state and initiating prosecutions of those m agistrates who step outside established law. T hey are to act as legislators by filling in details om itted by the original lawgiver and by initiating changes in the legal code on the rare occasions when this will be necessary (770a77 la, 772a-d). T h e large num ber of more specific duties m entioned as the dialogue progresses, taken together, give them a general power of supervision over every area of the com m on life, including social, commercial, religious and dom estic m atters, as well as what we would call political affairs (M orrow, 1960, pp. 198-204;. T he same shift of power from the popular in stitu tio n s to the magistracies is evident in the proposals for the adm inistration of justice. T he com petence of m agistrates to dispose o f m inor cases is increased and th e senior magistracies are involved closely in im portant cases: th e ‘select judges’ will be the ultim ate court o f appeal in civil disputes (767c-e) and, together with the guardians of the laws, they will form a special court to hear capital cases (855c-856a); three senior m agistrates will be responsible for the second of the three stages in the trial of public cases (768a). W hy would Plato transfer power to the m agistrates? T he im m ediate answer is that power m ust be exercised by those best qualified in term s of family background, character and training v751c-d). But what exactly are the relevant qualifications? In the R ep u b lic Plato had assigned suprem e power to the guardians on the grounds th at they alone would have the requisite philosophical understanding. To some extent th e same line of thought is evident in the L a w s : the nocturnal council, the m em bership of which overlaps extensively w ith the guardians of th e laws and the other senior m agistracies, will devote itself to philosophical and theological m atters. But throughout most of the work the em phasis is on correct m oral training rather than on philosophy. T he guardians of the laws and the scrutineers will have the advantage of age, since they m ust be at least fifty years old, and in


p ractice will probably have had a good deal o f adm inistrative experience. (T he offices of the cou n try w ardens and their assistants m ay be designed as a kind of training g ro und, and prom ising young m en will have atten d ed the nocturnal council.) But, although the citizens are enjoined to elect the best m en, there are no institutional g u arantees that only those who have proved them selves in this way will b e elected. As P lato him self sees, th e system will in the last resort d e p e n d on having citizens sufficiently well educated to make the right choices (751c-d). B ut th ere is m ore to P lato’s preference for m agisterial power than th e wish th a t decisions should be taken by those best qualified for the job. H e believed th at th e in stitu tio n s o f G reek dem ocracy were in h eren tly irrational (R ep u b lic , 487b-497a, 557a-558c; S ta te sm a n , 296a-297c). In the L a w s this belief is m ost evident in the discussions of judicial procedure. T h e A thenian com plains, explicitly or implicitly, a b o u t the upro ar in dem ocratic courts, the inability of jurym en to q u e stio n litigants and w itnesses, th e lack of tim e available for trials, a n d th e secrecy of voting w hen th e jury gave the verdict (766d-767a, 855c-856a, 876a-e). P lato’s ideal cou rt is one in w hich a relatively sm all n u m b e r o f judges exam ine th e issues in a calm and un h u rried w ay; they are able to questio n w itnesses fully and w hen the decision is tak en each judge takes personal responsibility for his vote. T he A th en ian departs from this ideal to th e extent o f giving a role to ‘trib a l’ c o u rts, apparently chosen by lot from th e whole people, b u t he seems to regard this as a concession m ade to give all the citizens a sense that they really are m em bers o f the state (768b). T hese points w hich the A thenian makes ab o u t the courts could equally be applied to the political process. W here large num bers of people are involved, as in dem ocratic assem blies, it will be difficult to achieve the careful, responsible deliberation Plato seeks. But he is equally hostile to bestow ing pow er on single individuals. T he A thenian generally talks as though the guardians o f the laws are to act jointly as a group. T h e same goes for m ost of the other m agistracies. In o th e r w ords, the m agistrates in the new city will be more like m em bers o f a com m ittee than executive officers. We are not told how these com m ittees will conduct their deliberations, b u t presum ably Plato believes th a t if they discuss m atters in a sufficiently calm and u n h u rrie d way they will eventually reach some kind of consensus. T h e R epu b lic has often been criticized for its assum ption that there are experts in politics to whom all pow er should be entrusted. T he L a w s shares the view of politics as a rational pursu it b u t interprets it m ore realistically by suggesting th a t good governm ent can be achieved



where a w ell-educated citizenry elects experienced officials and leaves them to pool their wisdom in taking decisions. For m uch the same reasons, most m odern states and m ost institutions and associations within the state now govern themselves through elected com m ittees.


Checks a n d balances

In discussing Peloponnesian history, the A thenian argued th at no individual should be allowed great or unm ixed power (693b: Saunders translates ‘pow erful or extrem e au th o rity ’), and th at the success of Sparta was due to the developm ent of a system in which different organs of state, owing their auth o rity to different sources, acted as constraints on one another (691b-693c). He then advocated a moderate state com bining m onarchic and dem ocratic elem ents (693c702e). As one would e x p e a , these argum ents are reflected in the constitution proposed for the new city. In the constitution there are a num ber of devices for restricting the power of individual politicians: (1) All officials on election will undergo d o k im a sia , an exam ination of their fitness (753d-e, 755d-e, 756e, 759c-d, 760a, 763e, 765b-c, 766b, 767d). Presum ably this exam ination would in m ost cases be conducted by th e guardians of the laws; it is not clear who would examine the guardians (753d). (2) At the end of their term all officials will undergo euthuna (variously translated ‘exam ination’, ‘audit’ or ‘scrutiny’ (946d-e). T he importance of the officials (‘scrutineers’) who conduct these examinations is em phasized by the extraordinary honours bestow ed on them (945b-948b). (3) All officials will be liable to prosecution for m aladm inistration or giving corrupt verdicts (762a-b, 767e, 846b). (4) All public acts will be subject to the general supervision o f the guardians of the laws. (5) T he requirem ent that boards of officials act jointly should restrict the excesses of individuals. (6) T here is a system o f appeal against judicial decisions (766d-767d). This list would be im pressive were it not th at all the item s, apart from the supervisory role of the guardians o f the laws and the system of appeals, had some kind of parallel in fo urth-century Athens. Yet Plato professes to consider the A thenian state lawless and unrestrained. T he point m ust be that in dem ocratic Athens there was no restraint on the



pow ers of th e assem bly and th e popular courts. But then one m ight o b ject th a t in th e new city th ere will be few restraints on the guardians o f th e laws as a group, th o u g h there w ould, no d oubt, be adequate sanctions to deal w ith an individual guardian who stepped out of line. M o rro w ( 1960, pp. 549-51 ) suggests that the guardians of the laws, the scru tin eers and the select judges are intended to act as restraints on one an o th er. T h is may be so, b u t in view o f the very sim ilar background an d m em bership o f th e th ree bodies they would be unlikely to check each o th er very effectively. So, while the power of individual m ag istrates is severely restricted, the A thenian is prepared to give alm ost u n lim ited au th o rity to w hat could be a clique of elderly men. N o d o u b t they would be scrupulously law abiding, b u t life would be d istin c tly bleak for any M agnesian w ith deviant or reform ist tendencies. In Book I I I the A thenian distinguished seven claims to rule (689e690c) and appeared to a ttrib u te the success of Sparta, not only to the restrictio n s w hich it im posed on individual pow er, b u t also to the ex istence, alongside th e tw o kings, o f a council owing its au thority to age and o f a quasi-dem ocratic body o f ephors (691 d -6 9 2 a ). One m ight draw from this th e lesson th a t in a successful state there m ust be d ifferen t organs of governm ent, able to restrain one another, and ow ing th eir auth o rity to d ifferen t sources. Polybius (V I, X I, 11-12) in te rp re te d th e poin t in this way and he was followed by M ontesquieu w h en he argued for th e separation o f the legislative, executive and judicial pow ers. P lato, on th e other hand, does not make any clear separation betw een different parts of the governm ent, though, as we shall see in a m om ent, he does take some account o f the different claims to auth o rity . T h e result is a state w hich, for all its safeguards against th e ty ran n y o f a single official, offers little protection to an individual o r group who falls out w ith th e authorities as a whole.


A m ix ed con stitu tion

Since th e proposed constitution lacks distinct m onarchic, oligarchic an d dem ocratic organs, it is clear th at in dem anding a mixed c o n stitu tio n , the A thenian is seeking to com bine different kinds of political p rin c ip le w ithin th e same governm ental structures. T his is confirm ed w hen, after describing th e complex elections to the council, th e A thenian rem arks, ‘T h e election m ade in this way will be a mean betw een a m onarchic and a dem ocratic constitution — and the c o n stitu tio n always should be m idw ay betw een th ese’ (756e). H ere



democracy and m onarchy are apparently com bined in the process of electing a single set of officials. T he A thenian then goes on to com pare two types of equality. One is ‘the equality of n u m b er, w eight and measure’ which can readily be achieved by the use of the lot (757a-b). This is apparently what A ristotle calls ‘arithm etical’ equality (P o litic s , 1301b29-1302a8; cf. N icom achean E th ics , 1131 b25—1133b28). It is a matter of treating everyone alike. T he A thenian’s second kind of equality apparently corresponds to A ristotle’s ‘geom etrical equality’. It consists in giving ‘due m easure to each according to his n a tu re ’, i.e. more to the greater and less to the lesser. I t requires us to give a greater share of honours and offices to those pre-em inent in virtue and education. T his is the true form of equality and constitutes justice. It needs the wisdom and judgem ent o f Zeus and brings great blessings to a city. But in practice, in order to avoid dissension, it is sometimes necessary to use the other form of equality, that em bodied in the lot (757e-758a). The distinction between these two types ofequality has been central to most subsequent discussions of justice. It is com m onplace to say that justice consists, not in treating all men alike, b u t in giving equal consideration to everyone’s claims. Those w ith equal claims are to be treated equally, while those whose claims are not equal may be treated differently. T he difficulty writh such a doctrine is that it does not tell us what kinds of claims are to count nor how different claim s are to be assessed. It is thus com patible with a wide variety of com peting political and social outlooks. In the present passage, the A thenian affirms that the m ain claim is that of virtue or goodness. But it is not at all clear wrhat this has to do with the election to the council. One line o f interpretation stems from Aristotle [P o litics, 1266a528). He com plains that the constitution is not, as Plato seems to suppose, a m ixture of democracy and m onarchy, b u t a m ixture of democracy and oligarchy. Aristotle says this because o f the attention paid to property classification, particularly in the elections to the council. He interprets these as oligarchic provisions designed to extend the influence of the wealthy. Barker (I960, pp. 388-90) follows this line of interpretation and therefore assumes th at, when Plato advocates his second ‘tru e ’ kind ofequality, he has in m ind the fact that greater pow er will be given to those of greater w ealth. He then, quite properly, com plains that, since virtue and wealth do not coincide, Plato should not describe his sytem as bestowing greater honours on those of greater virtue. In an obscure passage (744a-c), the A thenian claims that property classes will be necessary because o f ‘the equality of opportunities in the



city’. T hey will enable offices, taxation and distributions of public revenue to be assigned not only in accordance w ith the virtue o f an in dividual or o f his ancestors or in accordance w ith his physical a ttrib u te s, b u t also in accordance w ith his possession (or use) of wealth or his poverty. T his unequal, though proportional, distribution will prev en t dissension. Some com m entators have taken this to m ean that an in d iv id u al’s possession of w ealth is to be treated as a sign of his personal virtue. Saunders (1972, pp. 31-2) suggests that it is rather the use o f w ealth that indicates virtue. N either o f these interpretations pays sufficient atten tio n to the fact that the A thenian’s language in 7 4 4 b -c suggests a contrast betw een virtue, on the one hand, and w ealth or poverty on the other. T h e possession or use of wealth thus seem s to be an additio n al criterion to th at o f virtue, not m erely an index o f it. So the poin t seems to be th at wealth needs to be taken into account in th e organization of the state, even though it may be in d ep en d en t o f virtue. T his makes it very unlikely th at the account of th e tw o kinds o f equality in 756e-758a really involves the kind of confusion betw een w ealth and virtue th at is required by B arker’s interp retatio n . I t is not clear th at the co n stitu tion attaches as m uch im portance to w ealth as A ristotle (and Barker) suppose. T h e property classes enter into th e election of th e council in two ways: there are equal num bers of council m em bers from each class and the lower classes are exem pt from atten d in g th e later stages o f the nom ination process. T he first of these p oints w ould increase th e pow er of the w ealthier classes only if they were num erically sm aller than the lower classes. T his would be a fair assum ption in m ost states b u t not necessarily in one w ith m axim um and m inim um lim its on the holding o f wealth. T he ex em p tio n o f th e lower classes from the later stages o f nom ination seems prim arily in ten d ed to ease th e econom ic b u rd en of participation in public life. Its effect is difficult to judge b u t presum ably it would ten d to increase th e influence o f th e wealthy. T he same m ight be said of th e provision th a t th e lower classes will not norm ally be com pelled to attend the A ssem bly (764a). A few o f th e less im portant magistracies are also restricted to the w ealthier classes. B ut, in general, it is su rp risin g how little use is m ade of the property classes. T h e privileges of th e wealthy are very lim ited; in particular it is notable that the p rop erty classes play no special part in the courts of law or in the election o f the m ost im p o rtan t officers o f state, the guardians of the laws and the scrutineers, though Saunders (1972, p. 35) has read oligarchic intentions into the h in t at 753d that participation in the final stage of th e election for the guardians of the laws will be optional and



will involve some kind of sacrifice. (T here is some su p p o rt for this in the R epublic, 378a, where it is suggested that the know ledge of those myths which are of doubtful moral tendency could be restricted by the requirem ent that they be revealed only to those w ho have offered a large and expensive sacrifice.) On the whole, the function of the classes seems very m uch that suggested at 744c; Plato believes that there will be dissension unless the constitution takes some (albeit small) account of differences in wealth. If the ‘tru e’ equality o f 756e-758a is not a m atter of giving greater power to the wealthy, what is it? Perhaps the first point to notice here is that, according to the A thenian, the sim ple ‘arith m etical’ equality is characterized by th e use of the lot. He also says that it will in practice be necessary to com bine both kinds of equality. W e may therefore expect the m ethod of electing the council to involve the true equality in one way or another. These elections are, in fact, peculiarly complicated. After an elaborate nom ination process, twice the required num ber of candidates from each class is chosen by election and half of these are then chosen by the lot. T hus the process as a whole combines the lot with the vote. T h e natural inference is that the use of the vote is supposed to contribute the true equality by assigning greater honour to those o f greater w orth. This may seem strange to us because we regard the choice o f officials by election as characteristic of democracies. But the Greeks thou g h t differently. In their view the democratic m ethod was to choose officials by lot w hile elections were, if anything, an oligarchic procedure (759b; cf. Aristotle, P o litics, 1294b8-13). I t is not difficult to see why they th o u g h t this: if all men really are equal, then all are equally deserving of office, and it does not really m atter whom you choose: so why not choose at random ? P utting it round the oth er way, in holding an election you are asking the citizens to say who is most suitable to hold office and this implies that some are b etter qualified than others. We tend to miss this point because we think in term s o f choosing representatives. But the Greeks did not have representative democracies. T h e Athenian clearly shares this view of elections as a m ethod of choosing the candidates best qualified for office (75 lc -d , 945b-946c). Thus the most likely interpretation o f his talk o f two kinds of equality seems to be this. H e thinks that the ideal would be to ensure that honours and offices go to the virtuous and that in a w ell-ordered state this can be achieved by elections. But even here he thinks it desirable to make some concession to egalitarian sentim ent by incorporating an elem ent o flo t (756e-757a, 759b). T hus Barker (1960, p. 397) is quite wrong to assume that Plato cannot have supposed that the people



w ould be qualified to choose its own rulers and th at his proposals for elections are sim ply concessions to the masses. It is the lot, not the vote, w hich is used ‘to avoid the discontent of the people’. T h is leaves us w ith th e p ro b lem o f explaining how in 756e the A thenian can describe the election to the council as com ing m idw ay betw een dem ocracy and m onarchy. T h ere is no elem ent in the council or elsew here th a t could clearly be called m onarchic — nothing like the S partan kings for exam ple. T h e problem is similar to that raised in Book I I I , 693d-761d, w here b o th th e A thenian and the Persian states in th eir prim e are said to have com e midway betw een m onarchy and dem ocracy, even th o u g h A thens had no clearly m onarchic elem ent and Persia no clear elem ent o f dem ocracy. As we saw (chapter 7, section 6), th e only way to make sense o f th at passage is to suppose th at the A thenian is asking for a state w hich blends liberty w ith respect for a u th o rity . T h e sam e w ould seem to be true o f the present passage. ‘M o n arch y ’, it seem s, does not m ean ‘kingship’ b u t is applied to any system in w hich th e citizens are expected to show respect for their rulers.


D em ocracy

I t is som etim es said th a t th e L a w s shows greater sym pathy for the ideals o f dem ocracy th an does th e R epu blic. T o evaluate this claim we need to know w hat is m eant by the ‘ideals o f dem ocracy’. F ro m one point o f view, no G reek state could claim to be dem ocratic. In all o f th em th e w om en and slaves who m ade up a large p a rt o f th e population lacked significant political rights. T he city o f the L a w s w ould be no exception in this respect, since, even if w om en are enfranchised, slaves and aliens will still be excluded from the political process. But, quite ap art from this, there are other im portant differences betw een th e G reek and the m odern conceptions of dem ocracy. In th e G reek view, dem ocratic constitutions were characterized by the following features: all adult male citizens had the rig h t to serve on juries and vote in the assem bly, ultim ate political pow er rested w ith these popular in stitutions and m agistrates were generally chosen by lot. O n the m odern view a dem ocratic constitution is one where the g overnm ent is chosen by elections in w hich all adult citizens have a right to vote. T h e constitution proposed for the city of the L a w s contains certain features which a G reek w ould recognize as dem ocratic: all citizens have right to serve on juries and there is a (very lim ited) use of the lot in



elections. But since ultim ate power rests w ith the m agistrates and since these are chosen chiefly by election, it would hardly count as a democracy in the G reek sense. O n the other h and, because its rulers are-elected, it would be m uch more like a dem ocracy in the m odern sense. In addition to considering the outw ard forms o f the constitution we should also ask what principles underlie it. H ere again it is necessary to distinguish different views of democracy. Some dem ocratic theorists have advocated democracy on the grounds that everyone has an equal right to participate in governm ent. From this point of view the direct dem ocracy of th e Greeks m ight be seen as ideal, while a system in which th e citizens elect representatives may be regarded as a way of avoiding the practical obstacles to direct dem ocracy in a large m odern state. We may call those who advocate dem ocracy on these grounds dem ocratic idealists. Such thinkers may believe th at dem ocratic processes will result in decisions taken for the com m on good, b u t they are not prim arily interested in questions of efficiency. T hey emphasize the right to participate or the need to express the general will rather than the im portance of wise government. T h e second group of dem ocratic theorists includes all those who advocate the choice of rulers by election on the grounds that this is the form of constitution m ost likely to produce good governm ent. They are not prim arily concerned to ensure that governm ent decisions reflect the general will. They may even believe th at the ideal ruler would be an enlightened despot. T hey would m erely insist that there is no m eans of guaranteeing that a despot will be enlightened. T hey may be equally dubious about the political com petence of the average voter. But they will still m aintain that a system in w hich all citizens have the vote gives us the best chance of achieving a tolerable governm ent and avoiding the excesses of tyranny. O n the w hole, they believe, the electorate will vote, not for m en exactly like them selves, b u t for those w ith some degree of expertise and knowledge o f political m atters. Once a governm ent is in power, the fear of losing office at the next election gives it a powerful incentive to consider the interests of the people at large. M oreover, an electoral system gives the citizens a sense of ‘belonging’ and helps them to identify w ith their governm ent. Theories of this kind have been called ‘dem ocratic elitism ’ because they acknowledge that governm ent will in p ractice be in the hands of an elite and confine the role of the electorate to choosing rulers from am ong this elite. It hardly needs saying that Plato is no dem ocratic idealist; he does not thin k th a t all m en have an equal claim to p articipate in politics and



he has no real respect for th e will of th e people. On the other hand, he does have a good deal in com m on w ith the dem ocratic elitists. His prim ary concern is to bring about good governm ent in a stable, peaceful com m unity. H e sees elections as a m eans of ensuring that political offices are held by properly qualified candidates whose c h aracter and education enable them to do w hat is right. T h e ‘checks and balances’ in the constitu tio n will restrain any m agistrate with tyrannical leanings, while the com bination of dem ocratic and nondem ocratic elem ents in the constitution will give the citizens a sense of solidarity (756e-757a, 759b).

12 Education and the Arts L aw s, 632-64Id, 643a-650b, 652a-661d, 663d-674c, 700a-701c, 764c 766b. 788a-823a, 950d-952d, 960b-969d; Republic, 376c-412b, 521c534e, 595a-608b; Timaeus, 47c-e, 80b, 88b-c, 90d

Education, as Plato conceives it, is a m atter o f training the young citizens in the forms of goodness or virtue they will require as adults (643d-e;cf. 641-c). We have already seen th at the inculcation of virtue is also th e m ain aim of th e legal code as a whole (chapter 4 above). It follows therefore that education, in Plato’s eyes, is not just one among many functions of the state b u t in some sense em braces all the other functions. This educational focus is apparent from the beginning of the dialogue, w hich starts from a criticism of th e institutions by which the C retans and Spartans tried to instill th eir own ethos into their young. T he m eanderings of Books I and II serve mainly to show how virtue may be taught. These ideas influence every aspect of the constitution proposed for the C retan city b u t are picked up especially in Book V II w hich is largely concerned with the education of the young.


The aim s o f education

Speaking generally, education may be defined as a m atter o f training the young to be good at w hatever pursu its they will have to follow as adults (643b). T hus, if a child is to be a craftsm an, one will give him toy tools; if he is to be a soldier he will be taught riding and the like; in oth er words, children learn th ro u g h play (643b-c). O f course, Plato, as we would now expect, despises such ‘technical' training as ‘illiberal’ and ‘banausic’ (643e-644a), b u t he believes th a t the same principles apply to genuine education. One m ust encourage in children a passionate desire to be good citizens and to know how to rule and be ruled in accordance w ith justice. T his kind of education is the greatest of gifts that can be bestowed on good men (644b). Since virtue consists



in having our desires adjusted to w hatever reason judges to be right, th is citizen ’s ed ucation will involve so train in g a child’s feelings of pleasu re and pain th a t, even before his reason can grasp the nature of v irtu e, he loves w hat o u g h t to be loved and hates w hat ought to be h a te d (653a-c). T h e judgem ents o f reason are em bodied in law (645a) w h ic h rests on th e w isdom an d experience o f th e old. So in saying th at ch ild re n should be h a b itu ated to feel pleasure and pain in accordance w ith law th e A thenian is suggesting th at their pleasures and pains sh o u ld coincide w ith those o f wise old m en (659c-d). Books I an d I I suggest tw o devices for achieving this kind of education. T h e first is th e m uch m aligned drinking party. T h e A th en ian believes th a t, p ro p erly directed, such parties will provide ways o f b o th testing an d train in g th e young m en’s powers o f selfcontrol (641a-d, 649d-650b; cf. 634a-b). Since those who are m ildly intoxicated often reveal w hat they would norm ally take care to conceal, it is easy enough to see why the A thenian sh o u ld regard drinking parties as a m eans o f testing a young m an ’s character. I t is less easy to see how these parties could actually help one to develop self-control. T h e A th en ian ’s argum ent seems to be this: the D orians (and others) tra in th e young to resist pain by placing th em in pain fu l situations while encouraging them th ro u g h fear of disgrace not to give in; d rin k intensifies pleasures; so one may train people to resist pleasures by giving th em drin k and at the sam e tim e encouraging them to resist tem p tatio n th ro u g h a sense o f sham e (645d-650b; cf. 634a-b). T h e flaw in this argum ent is the assum ption th at drink is a m eans of intensifying pleasures. I t may be th a t, b u t it is m ore obviously a m eans o f lessening one’s pow er o f self-control (649a-b). So the proposal to stren g th en self-control by d rinking parties is not genuinely parallel to th e practice o f stren g th en in g som eone’s physical courage by su b jectin g him to pain. T h e re is th erefore no reason w hy drinking parties should help the young to overcom e pleasures in general. On the o th e r h an d , it is p ro b ab ly tru e th a t one cannot learn to resist the specific tem ptations of d rin k w ith o u t having experienced them . So, sub m erg ed in th e discussion o f d rin k ing parties, there may be a good po in t: one does n o t encourage a person to develop genuine self-control by pro tectin g him from all tem ptations. T his m ight be a sound criticism of th e m ore restrictive aspects o f the Spartan regim e, as it w ould be o f (e.g.) Calvinist Scotland. T h e second educational device is th e chorus. T o u n d e rsta n d the A th en ian ’s com m ents on th is, one needs to realize th at the G reek chorus perfo rm ed dance m ovem ents w hile singing in unison. T hree choruses are envisaged, one o f children, the second o f young m en and



the th ird o f older men. T he age of the th ird chorus seem s to increase as the dialogue goes on until it is very definitely a chorus o f th e old (664d, 665b, 670a-b, 671d-e, 812b-c). I t soon becomes a p p a ren t th at the Athenian is interested in the th ird chorus prim arily because it will set the standards by which the choral singing and dancing o f th e young is to be judged (666d-67 la, esp. 670c-671a). T hus m usic and dance will be th e chief device for bringing the characters of the young into accord with the judgem ents o f the old. M usic an d education are inseparable (642a).

2 M usic In 653c-654d th e A thenian offers two explanations o f w hy the chorus is needed. According to the first account it is needed to reinforce the education the citizens have received from childhood. E ducation consists in th e correct discipline o f pleasures an d pains; b u t this discipline grows slack as we get older; so the gods have given us festivals (involving th e chorus) as a means o f recreation and to restore the habits in w hich we were b rought u p (653c-d). T h e second argum ent treats m usic as a way of creating order w ith in us. C hildren resemble other young animals in being unable to keep still or quiet, but hum an beings are unique in having a sense of order in m ovem ent. We call this rhythm and melody (or ‘harm ony’: there is no exact equivalent of the Greek harm on ia) and it is a gift of those gods w ho lead us in singing and dancing. So a really educated man m u st sing and dance well (654b; cf. Tim aeus, 47c-e). But this is not sim ply a m atter of technical proficiency. O ne m an may be able to express accurately what he imagines to be good w ithout actually loving the good or hating the bad; another may have the right feelings of pleasure and pain w ithout being able to express w hat he has in m ind correctly in all respects. It is the latter, th e one w ith the correct feelings, who is to be preferred (654c-d). In other w ords, th e A thenian seems to be insisting th at the success o f th e perform er in conveying w hat is good does not m atter so m uch as th e sincerity and conviction w ith which he does it. Since m usic and dance are so im portant for education, it follows that in order to tell who is genuinely educated we need to know th e criterion of excellence in m usic and dancing (654d-e). T he im m ediate answ er is that there are tunes and dance forms associated w ith different kinds of character; those associated with goodness of soul or body are good and those associated w ith evil are bad (654e-655b). T hus, contrary to com m on opinion, w hat is good does not vary from person to person. If



we believe th a t it does so vary, th a t is because we falsely im agine th at goodness consists sim ply in giving pleasure and because we find most pleasure in w hatever happens to accord w ith our own character. Those w ho take pleasure in perform ances they really know to be bad are in special danger because they will ten d to grow m ore like the things they enjoy (655e-656b). T h e p o in t seem s to be th a t if we allow ourselves to enjoy m usic and dance ap p ro p riate to bad characters we will eventually acquire bad characters ourselves. Since m usic and dance can have this co rrupting effect it follows, according to th e A thenian, th a t they cannot be left uncontrolled. So he praises the Egyptians who long ago established rigid norm s o f m usic based on w hat is naturally right. We likewise should try to establish a code o f p erm itted m usical form s and should consecrate them to make th e m as p e rm a n e n t as possible. W e m u st resist the love o f novelty (656d-657b). T h is means th a t we can no longer treat m usical festivals as sim ple contests in giving pleasure. It is, rath er, the pleasure of the o ld and wise th a t is the criterio n o f education. So the judges in musical contests m u st resist th e applause o f the masses and try to teach them w hat is rig h t (657c-660a). T h e sam e principles apply to poetry. T he co n te n t m ust be correct and, in p articular, poets m ust teach that justice and happiness coincide (660a-664d). T h e A thenian retu rn s to th e topic o f correctness in m usic and dance w hile professing to give directions to the chorus of D ionysus (i.e. the ch o ru s o f in eb riated old m en). H e treats these arts as im itative (m im etic). T h e excellence of such arts, as o f everything else, depends e ith e r on th e pleasure they brin g , or on som e kind of correctness or on th e ir usefulness (667b-c). (This seems to be a good general point, th o u g h of course, som e philosophers, notably the utilitarians, would try to assim ilate th e different criteria of value). Some harm less pleasures m ay be judged sim ply as pleasures, b u t, in th e case o f the im itative arts, th e im p o rtan t th in g is th at they should possess ‘equality o f q u an tity and q u ality ’ (667d). M usic m ust therefore be judged according to w hether it exhibits ‘sim ilarity in its im itation o f the b e a u tifu l’ (668a-b). T h e p o in t seems to be th a t anything w hich seeks to im itate or portray som ething m ust accurately represent its proportions. So, the A thenian argues, in judging an artistic creation we m u st u n d e rsta n d its essential n atu re, w hich, apparently, am ounts to know ing th e in ten tio n w ith w hich it is represented and w hat it is sup p o sed to represent. T his point is illustrated w ith a discussion of visual representations w hich leads to the conclusion th at in assessing any kind of representation we need to know (a) w hat is represented, (b) how correctly the representation is done, and (c) w hether the m anner



of representation is a good thing (668d-669b . T h e p o int of this difficult passage seems to be that accuracy o f representation is essential but is not enough in itself: the work of art m ust have m oral as well as technical excellence. T h e A thenian now' launches into an attack on poets who are indiscrim inate in their use o f words, melodies, rh y th m s and dance m ovements. T hey do n o t make these consistent w ith one another or assign them appropriately to the characters they supposedly represent (669b-670c). So the th ird chorus will have to have sufficient understanding of musical m atters to choose a correct style for their own characters and to set an example to others (670d-671a). T here is no doubt th at Plato took seriously his claim th a t art (especially music) can be m orally corrupting and m ust therefore be strictly controlled. T here are corresponding argum ents in the R epu blic (376e-403e, 605a-608b, esp. 395b-396e, 605c-606d). H e ascribes the supposed m oral decline o f Athens to lawlessness in m usic (700a-701c), and he shows how these ideas could be p u t into practice in the new Cretan city (798e-802e, 812a-813a, 814d-816dj. In his description o f Athens, the A thenian appears to be claim ing as a m atter o f observed fact th at disorder in m usic leads to a m ore general lawlessness. O f course, it is obvious enough that there are correlations betw een musical and social change — it is no accident that ‘punk rockers’ do not dance m inuets or Viennese w altzes — and Plato is no doubt right in supposing th a t there was a connection betw een changes in m usic and the move towards a m ore perm issive society in Athens. We know also th at music can help to spread ideas and attitudes — this is why people write patriotic or revolutionary songs, for example. But presum ably m usic can have this effect only w hen social conditions are ready for it. So, although music may have a p eripheral role in changing people’s attitudes, it is natural to see changes of musical style as sym ptom s rather than as causes of changes in society. O f course Plato’s argum ent against innovation in m usic does not rest prim arily on observed correlations betw een m usic and social change. H e has a theoretical argum ent th at m usic is naturally associated with character and that in perform ing or taking pleasure in music we assim ilate o u r own characters to th at associated w ith the music. This argum ent rests on Plato’s concept o f mimesis. As Plato sees them , m usic and dance are im itative or m im etic arts (655d, 668a, 795e, 798d, 814e; cf. C ra ty lu s, 423c-d, R epu blic , 393a-c, 401a, S ta tesm a n , 306d), but, as everyone who has read the R epublic knows, P lato’s concept of m im esis , or im itation, is very obscure. W hen it is introduced in the R epu blic , mimesis is defined as ‘making oneself



like some other thin g in voice or ap p earance’ (393c). In th at passage S ocrates distinguishes betw een m im etic poetry, w hich p u rp o rts to give us th e w ords o f the characters rep resented in the poem , and purely n arrativ e poetry, where th e poet speaks in his ow n person (392d-394c). H e proposes to abolish from th e ideal state all m im etic poetry apart fro m th a t w hich involves th e im itation o f the m ost adm irable c h aracters (394c-398c). H e uses different senses o f m im esis a few pages la te r w hen he describes different m usical modes as ‘im itating’ d ifferen t kinds o f character in different circum stances (399a-b) and suggests th a t arch itectu re, as well as painting and scu lp tu re, can p ro v id e im ages o f th e virtues an d vices (401b). T he p ictu re is fu rth e r com plicated in Book X, w here Socrates attacks all m im etic art. Some o f his argum ents there seem to assum e th at the art in question is like a tr o m p e - l’oeil p ainting designed to make us take -appearance for reality (596a-598d, 602c-603b), w hile others w ould seem to apply to all art (5 98d-602b, 603b-606d). In these contexts it looks as though Plato is confusing a m ore general sense o f m im esis , in w hich it applies to all arts, w ith a m ore specific sense in w hich m im esis involves the attem pt to recreate th e appearance o f a thing. T h e im portance of m im esis is not confined to aesthetic contexts. P lato also uses it in the follow ing ways (am ong others): (1) T h e p articulars of everyday life stan d in the same relation to th e Form s as images do to originals (R ep u b lic , 5 0 9 d -5 1 le). T his kind of language is particularly pro m in en t in th e T im aeu s, where the dem iurge (i.e. th e creator) is said to make the visible universe in im itation o f the etern al Form s ( T im aeu s , 27d-29d). (2) In th e C ra ty lu s, Socrates argues th a t w ords im itate the things th e y stan d for. H e tries to dem onstrate this by show ing th a t the elem ents out o f w hich w ords are com posed are standardly correlated w ith elem ents in th e things they represent (422b-427d). (3) Law im itates the rule o f reason (S ta te sm a n , 300c-301b, L a w s 713b-714b). I t is obvious th at, in these contexts, im itation cannot consist in recreatin g th e appearance o f a thing. T he Form s, for exam ple, are not in tim e and space and therefo re in them selves have no visible appearance. I t looks as th o u g h w hat P lato m eans is th a t the Form s system atically determ ine th e characteristics o f the particulars th a t fall u n d e r th em , so th a t we can fully u n d erstan d why the particulars are as th e y are only by reference to th e F orm s. An analogy m ight be the relatio n sh ip of a b arom etric ch art to the variations in atm ospheric p ressures th a t it represents. T h e ch art does not literally im itate the p ressu re system , b u t th e pressure system determ ines the form o f the



chart and gives it its m eaning and significance. We could cap tu re this wider m eaning of mimesis by translating it as ‘rep resen tatio n ’ or even as ‘expression’. If this interp retatio n of m im esis is correct we can m ake sense o f the notion th a t m usic is a m im etic art. I t can represent certain characters, not in th e sense th at it literally resem bles them , b u t in th e sense th a t it exhibits in a different m edium the same fundam ental p atte rn s of order or disorder. T he m usic associated with bad character is thus intrinsically bad because it fails to reveal the right kind o f order and bad in its results because in perform ing and enjoying it we take its character into our souls. T here would be som ething especially w rong with a choral work in w hich the words, rhy th m , m elody and dance m ovem ents were inconsistent w ith one another in the sense th at they expressed different kinds of character. Such m usic w ould be sym ptom atic of some kind of m oral disintegration and w ould be especially abho rren t to anyone like Plato who saw th e state as having a single overriding moral purpose. So there is a stan d a rd o f correctness in m usic (657a, 667b-668d, 700e) w hich does n o t consist sim ply in pictorial accuracy. N eith er is it purely intrinsic to th e wrork o f art. M usic is ‘correct’ when it aims to em body forms of ord er characteristic of good soul and w hen it does this successfully. T he m ost obvious difficulty w ith this theory is th at of seeing how the character o f a hum an being could possibly exhibit the sam e kinds of order as a musical com position. A score will reveal w ith som e precision the elem ents o u t o f w hich a piece of music is com posed and the ways in which they are related to one another: these are expressible in num erical term s. Psychology offers us nothing like this. O f course there is evidence o f correlations. betw een m usical structures and particular hum an em otions or moods. T w en tieth -cen tu ry m usic has been criticized for neglecting these correlations and so losing the power to express any coherent p attern of feelings. T here may be parallels betw een these criticisms and P lato’s com plaints of incoherence in m usic, though the reflection that sim ilar com plaints were m ade about (e.g.) Beethoven should discourage us from accepting them too readily. At a less sophisticated level it is clear enough th at m usic can help to create a mood. C onsider, for exam ple, the stirring effect of m artial music and the use o f m usic to create atm osphere in films. A choral work w ould be incoherent if the emotions created by its musical elem ents were quite different from those dem anded by th e w ords. But there appears to be no evidence that m usic has a lasting effect on character.



W e know very little about G reek m usic b u t it is clear that there had b e e n elaborate theorizing about the em otional significance of p a rtic u la r m odes. In th e R ep u b lic (398c-400c) Plato proposes to ex clu d e those m odes he judges to be in ap p ro p riate. In the L a w s he is less precise and leaves th e m atter to th e ju d gem ent of the th ird chorus (802 a-e, 811b-c). T h is looks like a recipe for sim ple conservatism .


E d u ca tio n an d p la y

P la to m akes use of a p u n n in g association betw een p a id e ia , education, a n d p a id ia , play, to suggest th a t education is or involves play (643b-c, 653c-654a, 803c-804b). H e also uses the contrast betw een p a id ia and spou de w hich suggests seriousness, effort or zeal. T his contrast works in tw o ways: (a) play as som ething enjoyable can be contrasted w ith w h a t is irksom e or involves effort and (b) play as som ething w hich has n o value beyond th e enjoym ent it brings may be contrasted with serious activities u n d ertak en for an end. U sing these associations Plato can establish the paradox th a t p a id ia , play, is sp o u d e , a serious m atter. I t is so p artly because th e ed ucation of children th ro u g h play is a most im p o rta n t m a tte r an d p artly because th e m ost serious w ork of adults is a k in d o f divinely inspired play. T h e m ost obvious application of these ideas is in the suggestion th at ch ild ren are to be educated th ro u g h play. W e have already m et the idea th a t children should play at th e activities they are to undertake as a d u lts (643b-644a). A ccordingly, th e educational program m es of the new city should provide for careful control over ch ildren’s games. In p a rtic u la r innovation in ch ild re n ’s games is to be pro h ib ited because it encourages a dangerous desire for change (797a-798e). A sim ilar idea is app lied to adults w hen it is suggested th a t religious festivals are g iven, n o t only as a m eans o f recreation, b u t also to reinforce and resto re th e ir education (653c-d). B ut th e no tio n of play also works in an o th e r way. T h e ideas th a t m an is a p u p p e t or plaything o f th e gods a n d th a t gods lead th e celebrations in religious festivals (653e) are co m b in ed in 803c-804b. H u m an affairs are no t of any real im portance — only G od really counts — and as m an is th e plaything o f the gods he o u g h t to live in a way appropriate to th a t role, engaging in the noblest k in d o f play. T h e p u rp o rt o f th e obscure sentences at 803d seems to be th is : we now th in k th a t w ork or effort should be perform ed for th e sake o f play; so th e ‘serious’ w ork o f w ar is u n dertak en for the sake o f peace; b u t play or education is really th e m ost im p o rtan t and serious thing a n d th is will be fo u n d only in peace; so m an ’s tru e vocation is to live in



peace and to spend his tim e in form s of play, nam ely sacrificing, singing and dancing. These ideas are echoed in 806d-807e w hich describes the citizen’s vocation to a life of serious leisure and, most strikingly, in 817a-d, where the A thenian describes legislators as poets who produce th e finest dram a. A t least, the new state is in ten d e d as an expression or im itation (mim esis) o f the noblest and best life w hich is supposed to be the tru est poetry. (T he w ord here is literally ‘tragedy’ but does not carry the im plications o f unhappiness involved in the English word.) These passages show an interw eaving of religious, political, ethical and aesthetic ideas. As we shall see (chapter 15), the order of the universe dem onstrates the existence o f a god or gods, and tru e religion requires an understanding o f this cosmic order. T h e same p atterns of order w hen transferred to the hum an sphere becom e law, w hich should shape both the city and the character of the individual (967e, see above, chapter 3, section 3). M usic and dance are also forms of order which help to ‘harm onize’ the soul (653c-654a;cf. T im a e u s ,A lc e, 80b). So the same form s of order are expressed in the heavens, the w ell-constituted city, th e character of th e good m an, and in the finest music (cf. T im aeus, 88c-d). T h e appreciation of this p o in t casts light on Plato’s theory b o th of politics and of art as they are expressed in the L aw s.

Plato’s attitude to art has often seem ed paradoxical. H e is him self a great artist, yet, it may seem , th e censorship he im poses on the arts w ould reduce poets, m usicians and painters to th e role o f party hacks serving th e purposes of th e state. P art of the tro uble here is th at we are apt to see tru e art as consisting in self-expression, while Plato sees its essence in th e expression of the order w hich underlies the universe, and th e m oral life alike. His position th erefore has m uch m ore in com m on w ith th e religious artists of th e p re-R om antic era who saw their work, not prim arily as expressing th eir ow n em otions, b u t as an expression of the divine tru th . A lthough this perspective imposes lim itations on th e subject m atter and style of art, it is certainly not incom patible w ith genuine artistic excellence. Similarly, on the political plane, one m ight com pare the state, as Plato conceives it, with a m onastic com m unity seeking to glorify G od in the o rd er of its daily life and in the beauty of its liturgy. W hile there is a good deal to be said for this com parison there is at least one im p o rtant difference between the two kinds of com m unity. T he m em bers of a religious order accept th e ir vocations voluntarily and willingly give up their freedom . Plato also expects the citizens o f his state to live willingly, and perhaps



joyouslyj w ithin th e bounds set for them , b u t this will be out of co n d itio n ing rath er than free choice.


O rg a n iz a tio n a n d curriculum

T h e system o f education proposed for the new city in general follows th e p rinciples laid dow n in the earlier books. E ducation will be the respon sib ility of the state and all children, male and female, will be re q u ire d to atten d school (804d-e). T h e system will be un d er the d ire c tio n o f a m inister of education chosen from am ong the guardians o f th e laws by a vote of all the m agistrates (apart from the m em bers of th e council and its executive com m ittee). Since this will be the most im p o rta n t office o f state, th e electors m ust choose for it the citizen who seem s m ost excellent in every respect (765a-766b; cf. 809a). T o assist h im he will have officers to take charge of schools and gym nasia and to act as judges in m usical and gym nastic contests (764c-765d, 813c). T h e teachers them selves will be salaried foreigners (804d, 813e) — the G reek s seem generally to have h eld th e job o f a schoolm aster in low esteem . T h e A thenian is concerned w ith th e treatm en t children will receive fro m th e ir earliest years. Small children and even the u n b o rn m ust ex p erience plenty o f m ovem ent (790c-791c). T hey should be neither ‘sp o ilt’ n o r repressed by un d u e strictness — the general aim is to avoid excesses o f eith er pleasure or pain (791d-793a). F rom the age o f three th e y w ill atten d ‘play gro u p s’ w here they will be supervised by one of th e fem ale officials (793e-794d). F orm al education will begin for both sexes at th e age of six and will com prise m usic, dance, literature, gym nastics and m ilitary exercises as well as som e arithm etic and the elem ents o f reading and w riting (809a-810a; cf. 794d-796d on gym nastics and m ilitary exercises; 8 1 0 b-812aon literature, 814e-816d on dance; and 812a-813aon music). T h e A thenian appears to suggest a tim etab le w hereby th e years ten to thirteen will be devoted to literature a n d th irte e n to sixteen to playing th e lyre. B ut presum ably he cannot m ean th a t these years will be devoted exclusively to those pursuits. It is n o t clear how young citizens betw een sixteen and tw enty-five years old w o u ld sp en d th e ir tim e. One w ould expect a period of m ilitary service like the A thenian ephebate, b u t this is not m entioned explicitly. T he office o f assistant to th e country w ardens seems intended as a m ilitary an d political train in g b u t this is left surprisingly late — the m inim um age is tw enty-five (760c). W ith in this program m e m ost atten tio n is paid to m usic (812 a - 8 13a;



cf. 799a-802e) and literature (810b-812a). As we w ould expect, both are to be strictly regulated. In th e case of literature the basic text will be the Laws itself — the A thenian seem s to have in m ind p articu larly the preambles to the laws. Teachers will learn these by heart, and the M inister of Education will use them as a guide in deciding w hat other forms of literature are to be tau g h t (81 lc-812a).

5 H igher things Everyone, the A thenian believes, should know the essentials of arithm etic, geom etry and astronom y, th o u g h it is n o t necesssary for the population at large to study these subjects in great detail (818a; cf. 747a-b, 809c-d). T here is em phasis on the practical value o f these studies (809d; cf. 747a, 818c) b u t it is also m ade clear from the beginning that they are an essential preparation for m ore exalted forms of inquiry (747b, 809d, 818c-d). T he A thenian bewails the ignorance of th e Greeks in these m atters com pared w ith the E gyptians (819a-b). In particular he criticizes his contem poraries for failing to acknowledge the existence o f incom m ensurables (819d-820d), and for believing th at the planets ‘w ander’ instead o f following a fixed circular p ath (822a-c). It becomes obvious later why he is so concerned about this latter error, since th e supposedly rational m ovem ents o f the heavens play a vital role in the argum ent for the existence o f G od (897b-898e). Only at the end of the dialogue, in the account of the nocturnal council, do we learn about the select few who will study m athem atics and astronomy in depth. This council, which owes its sinistersounding nam e sim ply to the fact that it meets in the small hours of the m orning, is m entioned briefly at 908a and 909a, reappears at 95 l d 952c, and is discussed at length from 961 a onw ards. T here are slightly different accounts of its m em bership but it is clearly intended to include th e m ost distinguished m em bers o f the state including the ten oldest guardians of the laws (95 le, 961a). These ‘senior m em b ers’ will each bring w ith them one young m an aged betw een thirty and forty, though their choice will be subject to the approval of the council as a whole. The first task assigned to the council is that of re-educating atheists (909a). At 951d-952a it is described as ‘the council o f those who oversee the laws ’ and the m ain topic of investigation is given as ‘the laws of their own city plus anything significant about these m atters th at can be learned from elsewhere’. In this capacity the council hears



th e rep o rts o f th e ‘o b serv ers’ — citizens aged betw een fifty and sixty w h o have been allow ed to go abroad to see if anything can be learned fro m o th er cities (95 la -9 5 2 d , 96 la). T h e longer account of the council in 961a ff. is prefaced by a passage dealing w ith the need for som e body to keep the laws safe from change (960b-e). One m ight think th a t this fu n c tio n had already been assigned to the guardians of the laws and the A th e n ia n is very vague about the relationship betw een these two bodies — at 966b he seems to lose sight o f any distinction betw een th e m . In view o f th e overlap in th e m em bership o f the two bodies this is n o t a serious erro r. T h e point seem s to be th a t the guardians p ro tect th e laws by overseeing th eir day to day operation w hile the no ctu rn al c o u n c il preserves th e ir in tellectual basis. I t studies the laws b u t leaves to o th ers the task o f p u ttin g its discoveries into practice. So the older m em b ers can be seen as th e reason o f the state while the younger m em b ers are th e senses w hich provide the inform ation th a t reason req u ires (964e-965a; cf. 961d). In discharging this function th e council will have to consider firstly w h a t is th e aim o f th e state, then th e m eans by w hich this aim is to be a tta in e d , and finally w hat laws or m en give good advice (962b). T he e n su in g passage reiterates th e need for legislation to be directed to a single goal, nam ely v irtue (962c-963a). T his leads into the problem of th e u n ity o f virtue. In talking o f four virtues — courage, wisdom , justice and tem perance — we im ply th at they are distinct b u t also in som e sense the same. O ne d u ty o f the council is to u n d erstan d this u n ity (963a-964a, 965b-e). In doing so they will need not m erely to know th e nam es o f things b u t also to give a definition or account of th e m (964a-b). T his know ledge will especially qualify the council to e d u cate the state in v irtue (either by instruction or by punishm ent) (9 64b-c). As tru e guardians o f th e laws they m ust know in w hat sense good is m any an d in w hat sense one, they m ust know the tru e n a tu re of th e laws and m u st be able to teach and live th e ir own lives accordingly (966a-b). T o fulfil this task an essential requirem ent will be a com plete m astery o f th e tw o doctrines by w hich the A thenian w ould prove the existence o f a god o r gods — the doctrines th at soul is prior to anything else and that the m ovem ents of th e heavens are rational. T hey will need to have com pleted th e essential prelim inary studies, to u n derstand w h at these have in com m on, to apply th eir know ledge in fram ing ap p ro p riate m oral and legal rules, and to be capable of giving a rational account o f these m atters where th a t is possible (966c-968a). T h e re is an in terestin g am biguity in the G reek text of 967e. I f we take th e passage literally, it looks as though th e A thenian is saying th at th e m em bers o f th e council m ust be able to connect th e ir other studies



with their knowledge of rnusic. M ost translators take it in this way. But Cherniss (1953, p 377) who is followed by S aunders (1972, pp 9-10) points out th a t Plato sometimes says ‘m usic' w hen he means ‘philosophy’. In the present passage the am biguity may be deliberate. It is presum ably philosophy th at enables th e n o ctu rn al council to unify its studies, b u t m usic, too, can reflect the fundam ental order of the universe. T he Athenian refuses to describe in detail the course of study that will be followed by the nocturnal council, ostensibly because only those who have participated in such a course could u n d erstan d the reasoning behind it (968d-e). T here are some obvious parallels with the curriculum laid down for the guardians in the R epu blic. Like the members of the nocturnal council, the guardians of th e. R epu blic would begin their higher education w ith the study of m athem atics (521c531 d), b u t w ould eventually achieve a wider insight into the relationship betw een the one and the m any (531e-534e; cf. L a w s, 965b). T hey would undergo their first period o f philosophical training betw een the ages of th irty and thirty-five and w ould re tu rn to their studies at th e age of fifty. At this stage they w ould finally achieve a grasp of the Form s (539e-540a). T hese ages correspond to those o f the junior and senior mem bers of the nocturnal council (951a-e, 961a-b). Against this one has to acknowledge th at there are discrepancies between the account of the nocturnal council and the description of the philosophic training in the R epu blic. In th a t dialogue Plato not only dem ands th at the philosopher should be able to und erstan d what the many particular instances of a certain kind have in com m on b u t talks as though the one Form of (e.g.) the beautiful or the just is distinct from the particulars th at fall under it. T he F orm truly is while the particulars come betw een being and n o t being (475e-480a). T he Form s belong to an ‘intelligible’ w orld, n o t to the visible w orld of everyday experience (508b-c, 509d, 517b-c). N othing in the L aw s suggests this kind of separation of Form s and particulars. T he need is sim ply to show how we can talk of the virtues as one (963d-e), and this merely requires us to discern som ething th a t is th e same in all of them (965c-d). In the Republic the knowledge o f the good provides the means of understanding all reality (505a-509c), while in the L a w s it is apparently reduced to the m ore obvious role of enabling the nocturnal council to give moral and political guidance. T he guardians of the R epublic have to tu rn away from the everyday w orld of becom ing tow ards the ‘tru e ’ being of the Form s (5 1 8 a-5 19b, 521c-d); the nocturnal council needs the factual inform ation provided by its younger m em bers and is concerned w ith the practical problem s of



legislation (962b, 964e-965a). T h e R ep u b lic despises observational astronom y and seeks to replace it w ith a purely theoretical study (529a-530c); the L a w s , on the o th er hand, requires a study o f the visible heavens since it is th eir regularity and o rd er th at proves the existence o f G od (966e-967a). T h ese differences betw een th e R ep u b lic and the L a w s w ould not be p articu larly significant if th ere w ere strong independent grounds for h o ld in g th a t Plato m aintained to th e end o f his life the theory of Form s as ex p o u n d ed in th e R epu blic. A fter all, th e discrepancies are mainly d u e to absence from th e L a w s o f p oints m ade in the R epu blic. These om issions m ight have been m ade for th e sake o f sim plicity. B ut it is n o ta b le th a t th e P a rm en id es offers som e convincing argum ents against th e th eo ry o f F orm s so conceived, an d th a t later dialogues (with the ex cep tio n o f th e T im aeu s whose d ating is disputed) no longer use language th a t suggests a separation o f F orm s and particulars. On the o th e r h a n d they show a stro n g in te re st in the problem o f the one and th e m any. T h e S o p h ist, for exam ple, shows how to define a ‘sophist’ by progressively dividing th e category o f an art or skill. This m ethod w o u ld show how m any different species could all be grouped un d er on e general kind. T h e A thenian’s b rie f rem arks on the differentiation o f courage and w isdom (963d-e) w ould fit this p attern . So there are gro u n d s for holding th at, in spite o f superficial sim ilarities betw een the R e p u b lic and th e L a w s , the S o p h ist and the S ta te sm a n give us a b etter id ea o f w hat the nocturnal council w ould actually do. We may find fu rth e r evidence for this in 630e-631a, w here the A thenian, in language recalling th e S o p h ist, m aintains th at the tru e legislator would p u rsu e his task system atically, setting out his laws in their various classes. P resum ably his own legislative proposals are intended to illu stra te this m ethod. A concern for system is particularly apparent in Books IX and X, w here th ere are system atic classifications o f the causes o f injustice, o f hom icides and w oundings and of the various kin d s o f im piety (see below, chapters 14 and 15). It is not unduly fan cifu l to suppose th a t Plato sees this as an application of the m ethod w h ich he advocates in th e S o p h ist and th e S ta te sm a n and th at he would w ish th e n octurnal council to adopt th e same approach.

13 Punishment Lazes, 728b-c, 731b-d, 735d-e, 853a-882c, 908e-910d, 933e-934c, 941b-942a, 957e-958a; Protagoras, 323c-326e; Gorgias, 472d-480b, 507c-508b, 525b-c; Republic, 380a-b, 409e-410a, 591a-b

The Athenian takes it for granted that, as well as having persuasive pream bles, laws also require sanctions to give them their force or necessity (722b); so laws embody a kind of threat (722e). It follows that the entire legislative code is to some extent concerned with punishm ent, though the m ore im portant pronouncem ents on the theory of punishm ent mostly come in Book IX where the A thenian describes what we would call the ‘crim inal’ law of the new state.


The penalties

The penalties th at th e A thenian proposes differ considerably from those available to m ost m odern courts. T he chief penalties suggested are as follows.

Death Capital punishm ent is prescribed for a wide range o f offences including deliberate m urder (871d), w ounding a parent, brother or sister w ith intent to kill (877b-c), persistent atheism (909a), serious acts of im piety (910c-d), theft from tem ples (854e), theft o f public property (942a), harbouring exiles (955b), waging private war (955c), taking bribes (955d) and obstructing the judgm ent of a court (958c ). It looks as though these proposals would extend the use of capital punishm ent beyond w hat was curren t in Athens. T he offences against the state discussed in Book X II are handled with notable ferocity.



M onetary penalties T h e se take a variety o f form s. In som e cases, for exam ple, failure to m arry by the approved age (774a), the am ount o f the fine varies according to the o ffen d er’s p ro p erty class: the w ealthier citizens pay m ore. F o r offences w hich involve dam age to the interests of, or injury to , another person, th e general principle is that those convicted should pay com pensation plus an additional sum by way o f punishm ent. T hus in th e case of th e ft all offenders are req u ired to give double com pensation (857a-b), a provision borrow ed from A thenian practice. T h e chief innovation is th e stipulation th at no one shall be fined so heavily th at he can n o t afford to keep the equ ip m en t necessary for ru n n in g his farm and th a t no one is to be subjected to com plete confiscation o f p ro p erty (855a-c), m easures th at are essential if th e full n u m b e r of households is to be m aintained. Im p riso n m en t So far as we know , G reek states did n o t generally im pose im prisonm ent as a penalty in its own right, though one could be d etain ed while aw aiting trial or sentence. Plato introduces it as a pen alty for several kinds o f offender including those who strike som eone m ore than tw enty years th eir senior (880b-c), citizens who engage in retail trad e (919e-,920a) and those who prevent witnesses or litigants atten d in g a c o u rt (954e-955a). Im prisonm ent o f a rather special kind is im posed on atheists (908e-909b). In some cases im prisonm ent seems to be envisaged as an alternative penalty for those w ho cannot or will not pay fines (855b, 857a). D ish o n o u r and deprivation o f rights G reek cities used a w ide range o f penalties u n d er the general heading o f a tim ia (literally d ishonour) . In Athens this generally m eant the suspension of civic rig h ts, som etim es including the loss of citizenship. E lsew here it could be some kind o f public disgrace. In the code of the L a w s , a tim ia som etim es involves m inor deprivations of rights (754e755a, 784d) b u t generally m eans having o n e ’s nam e posted or being in som e o th er way publicly b ran d ed as a bad character (755a, 762c, 774c, 784d). I t is stip u lated th a t no one is ever to forfeit his citizen rights com pletely for a single crim e (855c).



Exile Sentences of exile were com m on in G reek states. T h e A thenian uses them for aliens, b u t in the case of citizens he w ould, w ith rare exceptions, restrict their use to cases of hom icide w here the underlying assum ption is that the continued presence of the killer will lead to religious pollution (864e, 867c-868e). Corporal punishm ent This was rarely, if at all, applied to the citizens of G reek states, though it would, one assum es, have been used widely on slaves. T h e A thenian proposes that citizens should be beaten for physical assaults on their parents (881d). In a few cases he would allow' th ird parties to beat offenders w ith im punity (762c, 784d, 845c). C orporal punishm ent is to be used m ore extensively for slaves and foreigners (854d, 879e, 881c, 882b, 917c-e).


The aim s o f punishm ent

Plato’s views on the purposes of punish m en t need to be seen in the light o f what we know about the views of his contem poraries. So far as can be gathered from our main source, the speeches o f the A thenian orators, popular tho u g h t recognized three m ain ai ms that punishm ent might accomplish: (1) An orator will sometimes ask that a penalty be im posed as an example to stop the criminal him self from repeating his crimes and to discourage others from em ulating them. This is. o f course, the idea em bodied in the m odern concept of deterrence. (2) Sometimes courts were asked to penalize crim inals to ‘restrain’ them or ‘bring them to their senses’. W hile this can sound like the m odern concept of reform it probably refers to m ore prim itive notions — som ething like the dem and that crim inals be given a ‘short, sharp shock’. (3) T he orators make frequent use of th e concept of tim oria. This has certain analogies w ith the m odern concept of retribution, though it is clear th at th e orators were not espousing a retributive theory of punishm ent as th a t would now be understood. M odern retributive theorists have been careful to distinguish th e im partial requirem ent of retribution from dem ands for personal vengeance. R etribution is im personal and is based simply on the idea th at it is right for those who



have done w rong to suffer. T h e G reek orators, on the other hand, often m ake it clear that they are b ringing prosecutions o u t o f private resentm ent and they call u p o n juries to share their anger. T h e L a w s is by no m eans th e first dialogue in which Plato discusses th e problem o f p unishm ent. H e p u ts a particularly notable account in to th e m outh of th e sophist Protagoras: ‘N o one’, say the sophist, ‘punishes sim ply for the sake o f the past; th at w ould be to exact blind vengeance ( tim b ria ) like a beast. A rational m an punishes to prevent the crim inal him self and those who are aware o f his punishm ent from d oing w rong in fu tu re ’ (P ro ta g o ra s 324a-c). T h ere are two im portant points here: (1) Protagoras dism isses m ere tim b ria and he does so in term s that w ould com m it him to rejecting n o t only p u n ishm ent as personal vengeance b u t also th e kinds o f supposedly im personal retribution d em anded by m odern retrib u tiv e theorists. (2) F o r tirn eria, Protagoras w ould su bstitute a preventive view of p u n ish m e n t, w hich seems to rely largely on deterrence, though it could leave room for elem ents of reform . Elsew here in the same speech Protagoras allows th at cities may, by penalties of death or exile, rid them selves of those evil characters who are incapable o f acquiring justice (322d, 325a-b). He also describes the p u n ish m en t o f children w hich he likens to the process of straightening a w arped plank by blows (325d-e). O verall, Protagoras’s attitu d e is th a t p u n ish m en t ensures th a t people will conform to social rules; it th u s secures the good of th e co m m unity as a whole. In the G orgias and th e R ep u b lic th e focus is rath er different. T he key doctrine in both dialogues is th a t injustice is b ad , n o t only for society, b u t also for the u n ju srm a n him self. T h e u n ju st m an is therefore like the sick m an and p u n ish m e n t benefits him by curing his sickness. T he em phasis here is on individual, rath er than social, good. O f course, the two approaches are com patible: it could be th a t pu n ish m en t does good b o th to the in dividual, by curing him o f injustice, and also to society by enforcing social rules. In the L a w s, the A thenian’s attitu d e to punishm ent becomes ap p aren t in the prelude to th e legal code (726-734d). H aving exhorted th e citizens to honour th eir souls, the A thenian’s im aginary legislator goes on to describe w hat som e, he thinks, m ight call the greatest ‘ju d g m en t’ (d ik e ) on those who behave badly. T his judgm ent consists in th eir becom ing like the w icked and so cutting them selves o ff from th e com pany o f the good (728b). T h e A thenian then corrects himself:



N ow this kind of suffering is not a judgm ent, for justice and judgm ent are fine things. T his is rather p u n ish m en t [or ‘retribution’: thew ord is um~ona\, a form o f suffering consequent on injustice. Both th e person w ho has experienced this and the one who does not are w retched. T he one is not cured; the other is destroyed that many may be saved. (728c) The fundam ental point is clear enough here: the A thenian is distinguishing betw een a genuine judgm ent that does good and the mere infliction of suffering which may do no good at all. W hat is not clear is why both those who undergo this suffering and those who do not are supposed to be w retched. T he most plausible explanation is suggested by Saunders: the person who undergoes suffering is said to be w retched because that experience does nothing for his soul, while the person who avoids it is equally wretched because, as he continues his life of crime, he will eventually be executed. A few pages later the legislator advises the citizens on how they should deal w ith the injustice of others. T hey should be gentle with and pity those whose u njust deeds can be cured. A fter all, no one is voluntarily a w rongdoer, for no one willingly accepts the greatest of evils into his soul. But when we come across som eone who is thoroughly wicked, then we should give vent to our anger. T h e em phasis here is strongly on doing good to the crim inal by curing him of his injustice. Only when he is beyond cure should other considerations come into play. T here are many other passages where the A thenian expresses the same fundam ental attitude to punishm ent — th a t it m ust do good prim arily by curing the crim inal. (See, e.g., 735e, 843d, 854d-855b, 862d-863c, 933e-934c, 9 4 Id, 957e.) T he underlying principles are discussed in detail in Book IX . T hree passages in particular seem to offer a full and explicit account: (1) In 854c-855a the A thenian is discussing the treatm en t to be given to those who have stolen from tem ples (a particularly horrendous crim e). Slaves and foreigners are to be b randed, w hipped and cast naked out of the land. T h e A thenian hopes that this penalty will bring the crim inal to his senses and make him a b etter m an, for, he says, penalties are not imposed by law to do harm ; they make the criminal more virtuous or less wicked. Citizens who steal from tem ples must be p u t to death since if they com m it such a crim e after the upbringing they have received they may be deem ed incurable. For these people death will be the least o f evils and they will benefit others by serving as an example to them. (2) In 862e-863c some new elem ents are added to this penology.


A nyone who has done dam age will be required to com pensate those he has injured. B ut w hen the act results from injustice in the agent there will need to be additional m easures in order to cure him of this ‘disease’ in his soul. T h e A thenian goes on to explain how this should be done: . . . w henever any m an com m its any unjust act, great or small, the law shall in stru ct him for the future either never willingly to dare to do such a deed, o r else to do it ever so m uch less often, in addition to paying for the injury. T o effect this w hether by action or speech, by m eans o f pleasures and pains, honours and d ishonours, m oney-fines and m oney gifts, and in general by w hatsoever m eans one can em ploy to make m en hate injustice and love (or at any rate not hate) justice — this is precisely the task of laws m ost noble. B ut for all those whom he perceives to be incurable in resp ect of these m atters, w hat penalty shall the lawgiver enact, and w hat law? T h e lawgiver will realize th at in all such cases not only is it b e tte r for the sinners them selves to live no longer, b u t also that they will prove o f a double benefit to others by q u ittin g life — and th u s he will o f necessity inflict d eath as th e chastisem ent for th eir sins, in cases o f this kind, and o f this kind only. (862d-863a, trans. Bury) T h e difficulty here lies in the contention th at the legislator may use any m ethod in handling crim inals. R eading this in the light of m odern penal theory we m ay credit Plato w ith the insight th a t curing or reform ing a crim inal need n o t involve doing anything unpleasant to him . It is even conceivable th a t one could reform him by rew ards or h onours. T his in terp retatio n is th e m ost natural one if the passage is taken by itself, and S aunders w rites it into his translation. B ut th ro u g h o u t th e L a w s th e A thenian takes it for gran ted th a t unpleasant penalties should be im posed on w rongdoers. T h e only other m ethod is in th e case of honest atheists w ho, w hile undergoing detention, will be ‘re -e d u c ate d ’ by m em bers of th e n octurnal council (909a). T he G o rg ia s , 478a-e, 525b, and th c R e p u b lic , 380a, 591a-b, likewise assume th a t injustice can be cu red only by the infliction of pain. A less ‘progressive’ reading o f th e p resen t passage would thus fit b etter into th e dialogue as a whole. T h e A thenian would not be denying th a t all crim inals should be punished. R ath er he w ould be asserting th at, in ad d itio n to th e p u n ish m en t o f crim inals, all the o th er m echanism s of th e state m ust be geared tow ards th e prom otion o f justice. T his will be done by words in th e case of th e pream bles to the laws and m any of the educatio n al provisions. I t will be done by deeds, not only w hen the



wicked are punished, but also when the jusi receive rew ards and honours. T hus the citizens in general will be steered into the paths of virtue and criminals will be given an incentive to reform . (3) In 933e-934c there appears perhaps the clearest sum m ary o f Plato’s penology. The Athenian insists first th a t those who have injured others by theft or violence should pay a penalty proportionate to the am ount of dam age in order to restore the ioss. (T he w ord here is literally ‘cure’.) T h en the crim inal m ust pay an additional penalty to bring him to his senses. This penalty should be less for those who, because o f their youth, or for some sim ilar reason, have succum bed to the influence o f others. It should be greater for those who have gone wrong as a result o f their folly or lack of self-control. T hey m ust pay the penalty not for the sake of their wrongdoing !w hat has happened cannot be undone) but for the sake of the future. T h e object is th at they themselves and those who see their punishm ent sh o u ld come to hate injustice or at least show some recovery from th at ‘disease’. The A thenian goes on to emphasize the need for care and precision in choosing the penalty for each offender. T h e foundation of P lato’s penology, as it em erges in these passages, is the claim that injustice is bad for us and may th erefore be likened to a disease. T he prim ary aim o f the penal system m u st be to cure this disease but, particularly in the case o f the incurable, there is room for other considerations such as deterrence and the need to rid the city of evil m en. Superficially, at least, such a system m ay seem sophisticated and rational — Saunders has laid particular em phasis on its advanced ‘scientific’ character (1972, pp. 18-20; 1973a an d b, 1976, 1981). T here are certain analogies w ith the m odern progressive'idea that crim inals should be treated by psychiatric or sim ilar m ethods — an idea defended by Lady W ootton (1963) and p o p ular among some psychiatrists and social theorists. It is im p o rtan t for us to consider w hether the form idable practical and theoretical objections th at have been raised against these m odern treatm ent theories would also serve to underm ine Plato’s penology.


Punishm ent as cure

Any curative theory of punishm ent m ust rest on three assum ptions: (1) T h a t crim inality resembles a disease in being bad for the person who has it. (2) T h a t the acts designated as crimes are sym ptom s of some internal disorder.



(3) T h a t the form s of p u n ish m e n t or trea tm e n t proposed can in som e way rem edy th a t disorder. A ssum ption (1) causes p articular problem s for those m odern penologists who take for granted th e u tilitarian assum ption th at an act sh o u ld be regarded as crim inal only if it is harm ful to other individuals o r to society as a whole. O n this view th e re is no basis — apart from a sim ple confusion of individual and collective good — for regarding crim e as b ad for th e crim inal him self. So the analogy betw een crim e a n d disease im m ediately breaks dow n. Plato is in a m uch stronger position because he has m aintained (a) th a t the virtues are divine goods a n d th a t to be dep riv ed o f th em is th e g reatest evil th a t can befall us, (b) th a t the just life is pleasanter th an the unjust, (c) that the unjust m a n will be separated from th e com pany o f the good and will do badly in later existences, and (d) th at no one therefore w ould voluntarily choose to be un ju st. (See chapters 4, 5, 6 above.) I f these claims were estab lish ed , the analogy betw een injustice and disease w ould be en tire ly defensible since injustice clearly w ould be bad for the un ju st m an . T h e difficulty, o f course, is th a t n one o f th em is based on any really firm line of argum ent. A ssu m p tio n (2) is im p o rtan t because the A thenian, like most other legislators, imposes penalties only on those who have com m itted offences. O n a curative theory this requires explanation. Should we n o t seek to diagnose those w ith d iso rd ered souls so th a t we may cure th e m before they com m it offences? A nd why should we assum e th at th e acts designated as crim inal arise from disorders in the soul? M ight n o t th e sam e act arise from very different in ner conditions? P la to ’s theory can go a long way tow ards m eeting these conditions. T o begin w ith he could p o in t to th e m any devices by w hich he w ould seek to educate citizens in the way o f virtue. T hese, he m ight claim, sh o u ld be sufficient to cure those w hose psychic disorders have yet to re su lt in crim e. I f a crim inal act has been com m itted, th at is p rim a fa c ie evidence th a t m ore serious (an u n pleasant) m ethods of cure are re q u ire d . O n th e o th e r h an d he does n o t believe th a t all who com m it offences should be p unished. As we shall see in the next ch apter he distinguishes carefully betw een dam age or injury and injustice. C u rativ e m easures are req u ired w here the act issues from an u n ju st state o f th e soul b u t sim ple dam age requires only com pensation. T he A thenian then distinguishes different causes o f injustice and goes on to a long account o f th e laws on hom icide and w ounding. T he general p o in t is th at courts are to pay close attention to motive and circu m stan ces of each act and to vary the p u nishm ents accordingly. T h e re are indications th a t he w ould wish to apply this to the treatm ent



of all crimes though there are only a few cases in w hich this is made explicit. (See, e.g. the laws on atheism (908a-910e) and on im proper advocacy (938b-c).) These points make it clear that Plato does not sim ply assum e that acts forbidden by law m ust stem from a diseased state o f the soul. Rather he would see the occurrence of these acts as an occasion for investigating w hether the p erpetrator was suffering from injustice in the soul and for giving appropriate treatm ent. B ut it is significant th at Plato uses crimes o f violence (hom icide and w ounding) to illustrate his points, for these are the kinds of crime that it is m ost natural to attrib u te to psychic disorder and which therefore fit the curative m odel best. T he m odel is m uch less n atu ral in the case o f m ost o th er crimes, e.g. failure to m arry, planting trees too close to one’s neighbours’ property, engaging in retail trade or w aging private war. T hese may be contrary to the particular form o f social order the A thenian wishes to im pose b u t they do not, on the face of it, look like sym ptom s of m ental illness. After all, the fact th at a legislator has chosen to make an act illegal does nothing to show th a t those w ho do it m ust be suffering from some psychic disorder. T h e same point can apply to some m odern reform theories. M edical treatm ent may be appropriate for some crimes (e.g. com pulsive shoplifting) b u t one cannot infer that it will be equally appropriate for others (including, e.g., m inor m otoring offences). A ssum ption (3) applies to any curative theory — one m ust give some grounds for supposing th at the treatm en t given to crim inals can cure their evil tendencies — b u t it poses special problem s for a theory like Plato’s w hich proposes to cure crim inals by p u n ishing them rather than by psychiatric or sim ilar kinds o f treatm ent. N o t only is there no obvious reason why the suffering implicit in punishm ent should im prove our character, it can even be argued th at traditional punishm ents make criminals worse by intensifying their hostility to society. Plato m ust therefore dem onstrate th a t the kinds of punishm ent he envisages would have the required effects. T h ere are hints o f an answ er to this problem in the G o rg ia s, 504b505b, and the R ep u b lic , 591a-b, both o f w hich suggest th a t the experience o f punishm ent curbs or restrains the desires. These passages could be taken in conjunction w ith the T im aeus doctrine that disorderly motions of the lower soul may affect the reason (42e-44d, 69d-70d; see above chapter 5, section 2). C ould one not suppose that punishm ent somehow brings these motions to order? T his idea may be attractive so long as one thinks in term s of corporal punishm ents which may literally create m otions w ithin us, but the A thenian does not



n o rm ally prescribe corporal p u n ish m en t for citizens. T he physio­ logical passages in w hich T im aeus argues th at the balance of fluids in th e body m ay affect the soul (86b-87b) likewise afford no help because th e re is no reason to suppose th a t p u n ish m en t affects the bodily fluids. W ith in th e L a w s th ere is no explicit account of how punishm ent c u re s, th o u g h we m ight c o n stru c t an account out o f som e rem arks m a d e d u rin g the discussion o f d rinking parties. T he A thenian there argues th a t the fear of disgrace or p u n ish m ent may stop us fleeing from d a n g e r. T his ‘good’ fear may thus help to train us in the habit of fearlessness (646e-648c). T h is idea could be generalized. Perhaps we initially refrain from crim e for fear o f p u n ish m en t and thereby becom e h a b itu a te d to just behaviour. T h e tro u b le w ith this suggestion is th at it m akes th e reform ative effect of p u n ish m en t dependent on deterrence. P u n ish m e n t w ould not benefit th e individual who is punished so m uch as th o se o th er people who are d eterred by the exam ple of his p u n ish m e n t and thereby acquire good habits. T h is is clearly the m ajor weakness in P lato’s account o f punishm ent. U n less he can provide su b stan tial grounds for supposing th at p u n ish m e n t rem edies th e disorders o f th e soul, there is no reason to accept a curative theory.


T he th eory in p ra c tic e

E v e n if it was g ranted, for th e sake o f argum ent, th at punishm ents m ig h t cure psychic disorders, th ere w ould still be serious difficulties in th e n o tio n o f a penal code based on curative considerations. C o nventional codes prescribe punishm ents f o r offences , b u t the leg islato r w ho aims to cure th e crim inal should be concerned prim arily w ith th e u n derlying m ental condition. W e m ight therefore expect his code to p rescribe specific treatm en ts for each kind of m en ta l disorder. A lternatively, if he had sufficient confidence in the expertise o f the c o u rts, he m ight leave all treatm en ts to th eir discretion. P lato is aw are of these poin ts. In Book IX the laws on hom icide and w o u n d in g involve quite elaborate distinctions betw een the different sta te s o f m in d w hich may have p ro m p ted the crim e. In each case the m ain division is betw een acts done in anger and acts th a t are fully v o lu n tary . T h e section is prefaced by a discussion of the causes o f in ju stice (863a-864d) and it looks as if these two laws may be intended as exam ples o f the kind of th in g th at legislators and courts should aim at in o th e r cases (876a-e, 934a). B ut Plato seems reluctant to press th e se points to their logical conclusions. F or m ost o f the offences



mentioned in the L a w s, the Athenian assigns a specific penalty w ithout any indication of how it m ight be varied to suit different m ental conditions. Even the laws on homicide and w o u n d in g include elements w hich w ould fit badly w ith à strictly curative policy. For example, the penalties are varied, not only according to the m ental condition of the offender, b u t also according to the civic s tatus (slave or free) of the offender and his victim and the degree o f family relationship, if any, subsisting between them . I f som eone kills a free man in anger w ithout prem editation he will be exiled for tw o years; if he kills a slave he will have to pay double damages to the ow ner; b u t if he kills his own parent he will be put to death, unless, curiously enough, the parent manages to forgive the killer before he dies (868e869a). Considerations of this kind m ight have some relevance for a curative theory — perhaps the anger of those w ho kill parents is usually m ore incurable than that of those who kill strangers or slaves. But it is possible on some occasions that the killing o f a stranger could show m ore serious depravity7 than the killing of a p arent. So there seems no justification for rigid distinctions here. T he provision th a t a parricide may be acquitted by his victim is particularly anomalous since it has no relevance whatever to th e curability o f the offender. In 875d-876e there is a digression about the need for courts to exercise discretion. T he A thenian’s position is th a t an ideal ruler would rule w ithout laws. In practice this is im possible, b u t where there are well educated judges, as m uch as possible should be left to their decision. Since th e judges in the new C retan C ity will be well educated, one m ight expect them to be given w ide discretion in sentencing. Some of the A thenian’s measures do allow for this (847a, 867d-e, 880c, 928c, 941a). These cases however are relatively unusual. The A thenian’s general practice is to lay dow n a p articu la r penalty for each offence w ithout any explicit indication th at judges m ight vary it to suit the m ental state of individual offenders. T here are other indications th at the A thenian does n o t adhere single-m indedly to his curative theory. This is evident, for exam ple, in his use o f the death penalty which is supposed to be restricted to incurables. T he A thenian takes this restriction seriously in his treatm ent o f thefts from temples and public p ro perty (854d-e, 941e942a). Citizens who com m it these offences after th e education they have received should be executed as beyond cure, b u t slaves and foreigners may be allowed to live. T he rationale of this distinction is clear enough (though one m ight object to its crudity — can a ll citizens but no slaves or foreigners who com m it these offences be deem ed incurable?) A similar' point is made in the case of im p ro p er advocacy



th ro u g h avarice (938b-c). It is reasonable, too, to suppose th a t those w ho rem ain ed unreform ed after three convictions for perjury are b e y o n d cure (937c). B ut in m ost cases th ere is no obvious justification, w ith in th e A th en ian ’s official theory, for th e use o f capital p u n ish m e n t. T h e A thenian decrees the death penalty for offences like h a rb o u rin g an exile, waging private w ar, taking bribes (955b-d) and o b stru c tin g the judgm ent o f a cou rt (958c), w ithout any attem p t to sh o w th a t those w ho do these things m ust be incurable. It is notable, to o , th a t, in spite of th e cases ju st m entioned, in w hich he acknow ledges th a t slaves are less likely th a n citizens to be beyond cure, h e is norm ally m uch m ore severe on th e form er. O th e r features o f th e penal code also look troublesom ely irrational. I n d ealin g w ith hom icide cases (865a-874c), the A thenian assumes th a t th o se who kill others, even accidentally, are polluted by the b lo o d sh e d and m ust therefore be excluded for a tim e from th e haunts o f th e ir victim s, as well as undergoing cerem onial purification. T hus m o st killers are req u ired to spend a period in exile. T h ere are even p ro v isio n s for th e trial o f anim als and inanim ate killers (873d-874a). T h e A thenian repeats stories about the vengeance o f the gods and o f th e souls o f th e dead even th o u g h he officially regards vengeance as irra tio n a l (729e, 865d-e, 870d-e, 872e-873a). M o st o f the divergencies betw een th e penal code and the official th e o ry could be explained only on th e supposition th at the A thenian a tta c h e s w eight to o th e r considerations besides th e need to cure the crim in al. Even th e explicitly penological passages m ention deterrence a n d th e need to rid th e city o f evil m en. T h e latter purpose could be se rv e d by sentences o f exile (735d-e). T h e fact th a t the A thenian so o fte n prefers th e death penalty m ay suggest th at deterrent co n sid eratio n s weigh heavily w ith him . T h ere are even passages w hich im p ly th a t a d e te rre n t function is b u ilt into his conception o f law. In d istin g u ish in g betw een laws p roper and pream bles he argued th at laws im p o sin g a penalty em body a kind o f th reat. T hey may therefore be classified as using force rath er than persuasion (721e, 722b; cf. 784c, 8 5 3 b -c). H e illustrates th e poin t by reference to those who fail to m a rry at th e appropriate tim e. H e hopes th a t th e penalty will prevent th e m seeing th e single life as easy and profitable (721d). Clearly at this p o in t d e te rre n t considerations are u p p erm o st in his m ind. T h e A thenian also expects p u n ish m en t to have an educational effect. I n 631 b -6 3 2 d he argued th a t th e prim ary task o f the legislator is to in s tru c t th e citizens in v irtu e as a whole and he im plied th a t one way o f doing this is by th e correct d istrib u tio n of praise and blam e, rewards a n d penalties. T h e th o u g h t m u st be th a t penalizing a form of action



teaches us th at it is wrong, just as rew arding it teaches us th at it is right. Potential offenders are unlikely to be deterred by the th reat o f punishm ent unless they know in advance w hat penalties are likely to follow the crimes they contem plate. So, in its d eterren t aspect, punishm ent presupposes some kind of tariff of penalties for offences. Similarly, if penalties are to have an educational force, instructing us in what is right and wrong, they will need to reflect the seriousness of the offences for w hich they are prescribed. If the penalties for a type of crime varied from case to case for reasons not obvious to the ou tsid er the effect would be to confuse rather than to stren g th en ethical judgem ents. So, on educational as on deterrent grounds, there are good reasons for discouraging too many individual variations in penalties. In practice the A thenian seeks, not only to make the citizens virtuous, b u t also to make the state unified and orderly. F rom this perspective even some of the m ore irrational-looking penalties can have a point. For example, slave offenders m ust be pu n ish ed severely because only through fear can the slave population be kept in subordination. T h e belief in pollution could also serve a social purpose by encouraging people to look w ith horror on u n n atu ral deaths. T he fact that the A thenian would exem pt from pollution those who kill unintentionally in war, public games o r m ilitary training and doctors who fail to save their patients’ lives (865a-b) suggests this kind of explanation — in these cases a belief in pollution w ould be socially counterproductive. T he sam e may be tru e o f the passages suggesting that the gods or the souls o f the dead avenge themselv es on those who have killed. Such a belief may be irrational in itself b u t it could certainly be useful. Even the provision th at animals and inanim ate objects which have caused death should be tried like crim inals could have a utilitarian basis. Such trials m ight fulfil functions like those o f a m odern coroner’s court (M acdowell, 1963, pp. 85-9, 143-4). Once the w ider social purposes th at punishm ent m ight achieve are taken into account, it ceases to be surprising th at the A thenian does not carry to its logical conclusion his belief in the curative value of punishm ent. In practice he expects the legislator to act on considerations that give the penal code a q u ite different shape from any which a purely curative system would bear. T his may be inevitable, since the m o s t plausible accounts o f how punishm ent could cure psychic disorder seem to presuppose that th e institution is already accepted on other grounds. Some m odern scholars have held the A thenian’s account of the aims of punishm ent to be particularly enlightened, because it rejects ideas



o f vengeance and insists on th e need to cure the crim inal and discourage fu tu re crim es. T hey have th en regretted that, by com parison, the actual penal code is reactionary and even barbaric. T h e assum ption here is th a t it is m ore enlightened to pu n ish for some fu tu re good th an as a response to w hat has happened in the past. B ut th e re are pow erful m oral objections to m ost forw ard-looking views of p u n ish m e n t. T o p u n ish som eone as a d e terre n t is to tre a t him , n o t as an en d in him self, b u t as a m eans to som e supposed social good. ‘C u rin g ’ or ‘reform ing’ th e crim inal may in practice m ean ‘b rain w ash in g h im ’ or rem oulding his character to a p a tte rn approved b y the authorities. T h e penal code o f th e L a w s highlights these dangers. A lthough the A th en ian ’s use o f quasi-m edical term s such as ‘cu rin g ’ suggests th a t he is concerned w ith the welfare o f the individual crim inal, in practice m ost o f th e crim es for w hich he will be punished are not harm ful to him as an individual; they are generally offences against th e social o rd er o r deviations from the approved norm s of th o u g h t and behaviour. T h is is especially obvious in the case o f offences such as innovation in m usic (799b) and atheism (908e-909a). T h e A thenian sees p u n ish m en t as a m echanism for enforcing social conform ity. In some ways it m ight have been preferable if he had stuck to ideas o f vengeance, for although such ideas may be prim itive and b arb aric they involve no th re a t to th e integrity of the individual.

14 Responsibility Laws, 731b-d, 734b, 859c-873b, 874b-877c, 878b-879b; P rotagoras , 352a-358e; Gorgias , 466d-468e

A lthough Plato in the L a w s no longer seems to accept th e Socratic denial of akrasia, he still maintains th at all injustice is involuntary (731c, 734b, 860d; see above, chapter 5, section 2). T his gives rise to a very obvious problem in the construction of the legal code. All legislators have found it necessary to distinguish voluntary from involuntary wrongs. But it looks as though this distinction could not be m aintained if all injustice was involuntary. So it seems th at Plato m ust either abandon his belief in the involuntariness o f injustice or do without one o f the fundam ental assum ptions o f crim inal law. He makes the A thenian face this problem fairly and squarely in 859c-864c. In this context th e words ‘voluntary’ and 'in v o lu n tary ’ are translations o f the Greek w ords hekon m d a k o n . T he G reeks used these words quite freely. Alm ost any act done w ith less th a n w hole-hearted enthusiasm could be said to be done akon. So besides ‘volu n tary ’ and ‘involuntary’ there are several other pairs of E nglish term s which could translate hekon and akon, for example 'in te n tio n al’ and ‘unintentional’, ‘willing’ and ‘unw illing’, ‘w ith his consent’ and ‘against his will’. T ranslators use these term s m ore or less indiscrim inately. F or m ost purposes this does not m atter, since the average English speaker is equally indiscrim inate. B ut philosophers and jurists som etim es try to give these w ords precise m eanings and to distinguish th eir meanings very carefully. So those who read the L a w s in an English version m ust rem em ber that, since Plato is working in term s of a single p air of opposites, hekon and akon, any variations are the work o f the translator.


152 1

T h e pro b lem

A t 859c th e A th en ian introduces w hat we may call ‘the paradox of just p u n ish m e n t’. H is argum ent seems to be as follows: (1)

(2) (3)


A nything w hich is just m ust also be kalon (a difficult w ord to translate: it can m ean ‘good’;, ‘noble’, ‘beautiful’ or ‘h o nourable’). Being justly p unished m ust therefore be kalon. But th e things done to us w hen we are p u nished are the very reverse o f those norm ally considered kalon (i.e. they are norm ally considered bad, ignoble or disgraceful). T h u s th ere appears to be a contradiction in ordinary ways of speaking.

T h e A thenian makes it clear th a t he and his com panions will have to take issue w ith o rdinary tho u g h t on this question (859c-d), b u t he does n o t explain exactly how he w ould solve the paradox. In the G orgias, 476a-477a, Socrates is made to raise a sim ilar problem . I t is solved th e re by the argum ent that, since just punishm ent is kalon , it m ust, in spite o f appearances, be good (i.e. beneficial) for the person who undergoes it. N o d o u b t th e A thenian has a sim ilar point in m ind here, b u t his language at 860c and 861a suggests th at he also means to make som e kind o f logical point. Perhaps, as Saunders (1968, p. 422) suggests, he m eans th at p u n ish m en t is good in one aspect though not in another. A t 860c th e A thenian claims th at a sim ilar inconsistency may appear in his own view th a t bad m en are in all respects involuntarily bad. This belief com m its him to saying th a t injustice is involuntary, w hich in tu rn , seems to im ply th at no one voluntarily does an u n ju st act. But this conclusion raises serious problem s for th e A thenian in his role of legislator. I t is norm al in legislation to distinguish voluntary and involuntary acts o f injustice, penalizing the form er more heavily than th e latter. I t looks as though the A thenian’s philosophical belief that all injustice is involuntary w ould invalidate this distinction. T hus there appears to be a contradiction betw een th a t belief and the practical requirem ents incum bent on the A thenian as a legislator. At this point (861a) the A thenian recalls his tre atm e n t o f the paradox o f just p u n ish m en t. H e seems to be suggesting th at the p ro b lem of v o luntary and involuntary acts o f injustice can be solved in th e sam e kind o f way. H e cannot abandon his belief th a t injustice is involuntary b u t he can show th at there is som e o ther kind of distinction to be m ade betw een acts. T h e suggestion he offers (86ld 862c) is th a t we should distinguish betw een (a) acts o f injustice and



(b) harms or injuries. People sometimes assume th a t all injuries are injustices and therefore treat cases in w hich one m an involuntarily injures another as cases of ‘involuntary injustice'. T h e A thenian, on the other hand, would not call such involuntary injuries ‘injustice’. Equally, he would claim that there are acts of injustice w hich, so far from doing harm , actually confer benefits. H is point is th a t w hat makes an act u njust is n o t its h arm ful nature, but the character or disposition from which it stem s. Accordingly, he m aintains th a t w hen harm has been done there will be need for restitution but th at cases of injustice call for additional measures to cure the w rongdoer (862d-863a). It is not difficult to see in outline w hat the A thenian has in m ind and how this connects with his theory of punishm ent. In his view the legislator, in addition to com pensating those who have suffered injury, must also attem pt to cure the un ju st souls of those who do wrong. This means th a t he cannot confine his attention to the external aspects of actions but m ust also consider w hether they result from a disordered psychological condition. T h e kinds o f act th at we norm ally call ‘involuntary’ in everyday life are not the outcom e of the agent’s choices and do not therefore have any bearing on th e state o f his soul. T hus they do not call for any special treatm en t. T he A thenian w ould regard these as m ere ‘h arm s’. T he acts which are norm ally called ‘voluntary’, on the other hand, do stem from the agent’s choices and do therefore reveal his m ental condition. I f these acts are contrary to the requirem ents of justice they show th at the agent has an unjust soul and needs treatm ent. It is these th at the A thenian w ould call ‘acts of injustice’. T hus, even though the A thenian believes th at no one voluntarily acquires an u njust soul, he still needs to make a distinction between those acts which are indicative o f a bad m ental state and those which are not. This is the function of his distinction between acts of injustice and harm s. In its practical effect this distinction w ould be very close to th e conventional one betw een voluntary and involuntary crimes. U nfortunately, all is not as plain sailing as this w ould suggest. The A thenian does not simply abandon the conventional distinction between voluntary and involuntary crim es in order to replace it with a distinction betw een acts o f injustice and harm s. R ather, he talks as though acts o f injustice m ust be voluntary while m ere harm s may be involuntary. H e attacks the very idea th at an involuntary act should be called unjust (862a). In his penal code, too, he insists on the distinction betw een those illegal acts which are voluntary and those which are not. Thus he seems to be saying both th at all injustice is involuntary and

1 54


th a t acts o f injustice m ust be voluntary. In other w ords, he still seems to be left w ith th e co n trad ictio n he d etected betw een his belief in the in v o lu n tarin ess of injustice and th e assum ptions he makes about the req u irem en ts o f legislation.


T he solution

B efore considering how the A thenian m ight solve his problem , it may b e help fu l to set out schem atically the argum ents w hich generate the a p p a re n t contradiction. F irst there is th e argum ent to th e conclusion th at all unjust acts m ust b e involuntary. T h is seems to be as follows: (A l) (A2) (A3) (A4)

T he u n ju st m an is bad. N o one is voluntarily bad. T herefore no one is voluntarily unjust. T herefore no one voluntarily does an unjust act.

O p p o sed to this is th e arg u m en t th a t at least som e acts o f injustice m u st be voluntary. T his seems to ru n as follows: (B l) T he legislator m ust p u n ish injustice. (B2) In doing so he m ust tre a t those who act voluntarily more harshly th an those w ho do not. (B3) T herefo re the legislator’s activity presupposes th a t people voluntarily do u n ju st acts. G iv en th e A thenian’s com m itm en t to legislation, the conclusions (A4) a n d (B3) seem to involve him in a contradiction. T his contradiction could be avoided eith er (a) by rejecting at least one prem ise o f one or o th e r of th e argum ents, or (b) by denying the inference from one or o th e r o f the sets o f prem ises to th e ir supposed conclusion, or (c) by show ing th a t th e tw o conclusions do n o t really contradict one another. T h e prem ises (A l), (A2), (B l), (B2) are all accepted by the A thenian m ore or less explicitly. So is the inference from (A l) and (A2) to (A3). T h u s th ere are only th ree routes by w hich the contradiction could be resolved: (1) W e could reject the inference from (A3) to (A4). (2) We could reject the inference from (B l) and (B2) to (B3). (3) We could try to in te rp re t th e conclusions (A4) and (B3) in such a way th at they do not contradict one another. I f he took ro u te (1), th e A thenian would be denying the inference fro m ‘N o one is voluntarily u n ju s t’ to ‘No one voluntarily does an



unjust act’. T his inference is not, in fact, logically com pelling as it stands. One could insist that there is a difference betw een being unjust and acting un ju stly, and th at there is, therefore, no inconsistency in saying that the crim inal’s u n ju st state is involuntary while the acts which arise out o f it are voluntary. This, if I u n d erstan d him rightly, is the interpretation adopted by Saunders. In his view, the A thenian means to distinguish betw een the psychological state and the crim inal act. The act may be voluntary or involuntary in the sense th a t one has choice and freedom o f action. T he state of m ind th a t makes one commit the offence is, on the other hand, always involuntary in th at it is not the sort o f thin g one could ever choose (1968, pp. 423-4; cf. 1970a, pp. 367-8). One difficulty with this interpretation, as Saunders him self acknowledges, is that the A thenian makes no effort to m ark the distinction betw een states of m ind and acts, even though there would be no great linguistic difficulty in doing so. T here is also a rather obscure sentence at 860d which can be interpreted in such a way as to make the A thenian deny that there is such a distinction (Shorey, 1928). M ore im portantly, th e reasons that the A thenian offers for saying that injustice is always involuntary would seem to apply to unjust acts as well as to unjust states of mind. H e believes that injustice is involuntary because no one who really understood w hat he was doing and who had control over him self would choose to be unjust. But, since unjust behaviour can help to make our souls u n ju st, th e A thenian would presum ably say also that no one who really u n d ersto o d w hat he was doing and who had control over him self would do an u n ju st deed. Anyone, therefore, who accepts the A thenian’s argum ents about the involuntariness o f injustice ought, it seems, to accept th at no unjust acts are voluntary. T hus, if the A thenian th o u g h t he could solve the paradox via route (1), he would be seriously m uddled. I f th e A thenian m eant to solve the problem via route (2) he w ould be accepting that the legislator has to concern him self w ith both voluntary and involuntary acts , b u t denying that th ere is such a thing as a voluntary act o f in ju stice. T here is one obvious way in w hich one could support this claim. Suppose someone voluntarily com m itted a crim e, let us say an assault, but did not realize that this act was an act of injustice. C ould we not then say th at w hat he did was voluntary under the description ‘harm ing som eone’ but involuntary u n d er the description ‘com m itting an act o f injustice’? I f th at were so, then the legislator’s concern to distinguish voluntary and involuntary acts would carry no suggestion that there are ever any voluntary acts o f injustice.



T h e tro u b le w ith th is idea is th a t it runs co u n ter to com m on sense a n d to Platonic theory. C om m on sense suggests th a t m any un ju st m en realize th a t they are behaving unjustly. Plato w ould not, I think, deny th is. W hat he w ould say is th at these people do n o t realize th a t their u n ju st behaviour is bad for th em in th e sense o f being against th eir true in terests. T h u s so long as th e A thenian sticks to his view th at the legislator m ust distinguish voluntary and involuntary acts he seems co m m itted to saying th a t there are voluntary acts o f injustice. T h e A thenian could solve his problem via ro ute (3) if he was p re p a re d to ad m it th a t, w hen th e w ord ‘v o luntarily’ appears in his claim th a t no one volu n tarily does an un ju st act, it bears a different sense from th a t w hich it assum es in th e edicts o f th e legislator. I t is not d ifficu lt to see w hy one m ig h t wish to say this. As we shall see in the n ex t section, th e A thenian believes th at we becom e unjust if our souls are beset by ignorance or are overcom e by anger or by pleasures. T he ignorance th a t can make us u n ju st is not, o f course, ignorance o f particu lar facts, b u t ignorance of w hat is really good. W hen the A thenian claims th a t injustice is involuntary, he is assum ing th at everyone really w ants w hat is tru ly good for him . F ro m this it w ould follow th a t acts done in ignorance of the good or th ro u g h the d o m in atio n o f pleasures cannot be w hat we really w ant. H ence they can be deem ed involuntary. B ut in saying this kind of th in g he is assigning new senses to w ords such as ‘volu n tary ’ and the like. These new senses are o f no help to judges in courts o f law, for they have to decide w hether th e acts on w hich they are called to pass judgm ent are vo luntary in th e sense o f ‘arising out o f the agent’s underlying m ental c o n d itio n ’. F o r th is p u rp o se judges have to know w hether the accused actually w anted to do th e act, n o t w hether he w ould have w anted to do it if he had know n w hat was good for him or had had control over his passions. T hus th e senses o f ‘volu n tary’ and ‘in voluntary’ th at are req u ired by legislators and by judges are quite distinct from those involved in th e Platonic d octrine th a t no one is voluntarily u n ju st. One can adm it th a t th ere are v oluntary acts o f injustice in the ‘legislator’s sense w ith o u t having to adm it th a t there are such things in the special ‘P lato n ic’ sense. In this way th e contradiction betw een th e two conclusions, (A4) an d (B3) can be rem oved. T h e re is, I th in k , little d o u b t th a t the A thenian ought to have d istin g u ish ed th e tw o senses o f ‘voluntary’ and ‘involuntary’ th at I have ju st described. In th e G orgias, 466d-468e, Plato shows his aw areness o f a closely related point: he distinguishes w hat a m an really w ants (the good) from w hat m erely ‘seems good’ or ‘pleases’ him . In th e p resent passage, how ever, th ere is nothing to suggest th a t we are



expected to make distinctions of this kind. T hus it is im possible to be sure w hat solution, if any, the A thenian is supposed to have in m ind. It may be that Plato him self has not clearly d istinguished the different possibilities or perhaps he finds such precision foreign to his immediate purposes.


The sources o f injustice

At 863a-b the A thenian agrees to give m ore explanation o f his distinction betw een injustice and harm and betw een the voluntary and the involuntary. T he passage w hich follows, one of the most controversial in the L a w s, falls into three sections. 863b-e This gives a prelim inary classification of the causes o f error. T hese are anger, pleasure and ignorance. This classification recalls the tripartite division of th e soul (R epu blic , 436a-441c) though the A thenian is unsure whether we should speak of ‘p arts’ or of ‘states’ of the soul. T he point seems to be th a t injustice may arise through a fault o f any o f the main parts or tendencies w ithin the soul. Anger and pleasure may both be said to ‘overcom e’ us, though pleasure does so by persuasion and deceit while anger acts by brute force. Ignorance, although it is not said to ‘overcom e’ us, is also a source o f error. I t may be divided into two kinds, ‘sim ple’ and ‘double’. T he distinguishing m ark of the doubly ignorant m an is that he erroneously thinks he has knowledge. W hen accom panied by strength and power this double ignorance is the source of th e most serious errors; when it is w ithout power, as in the case of children and the old, it m ay be treated leniently. The distinction betw een kinds of ignorance recalls the Socratic doctrine that the greatest ignorance is to think one has knowledge w hen one does not. It w ould be interesting to know w hat kinds of people would count as ‘doubly’ ignorant. Perhaps those w ith strength are fanatics, so convinced of th eir own rightness th at they cannot be called to reason, while those w ithout strength are people who make themselves ridiculous through lack of self-awareness. 863e-864b Here the A thenian says that he is going to give a clear account o f justice and injustice, b u t the passage is rem arkably obscure. H e begins by



saying th a t the dom ination of the soul by anger, fear, pleasure, pain, envy and desires is injustice w hether or not it leads to actual harm . T h is seem s to be sim ply a m ore elaborate restatem ent of the point th at in ju stice m ay be due to anger or pleasure. T he point is that injustice can consist in an im proper d o m ination o f the soul by its lower elem ents. As we w ould expect, the next sentence seems to describe the kin d s o f injustice due to ignorance, b u t it is not all clear w hat exactly is bein g said. (A corru p tio n in the m anuscripts does not help, though fo rtu n ately m ost recent editors agree in their versions.) Literally tra n sla te d th e sentence reads: B u t if th e opinion o f w hat is b est, by w hatever means a city or private citizens th in k they can attain it — if this is suprem e in th e ir souls and governs every m an, th en, even if it goes w rong som ehow , w hatever is done in this way and w hatever in each m an is obedient to this is to be called just and best for the whole o f h u m an life, though m any people consider a harm of this kind an involuntary injustice. T h e key phrases here are ‘the opinion o f w hat is b e st’ and ‘if it goes w rong som ehow ’, each of w hich is susceptible of a variety of in te rp re ta tio n s. T h ree readings of th e passage as a whole have some plausibility. (1) ‘T h e opinion of th e good’ means ‘th c in d iv id u a l’s conception of w h at is good’ and ‘if it goes w rong som ehow ’ m eans ‘if th at conception is in fact m istaken’. T h e passage as a whole th en says th at a person who acts according to his ow n conception o f the good will not be called u n ju st even though his conception is m isguided. T his is the favoured in te rp re ta tio n o f R itter (1896, p. 283) and o ther older com m entators w ho saw it as a recognition (at last!) by Plato o f the c la im s o f conscience. O n this reading a genuinely conscientious action cannot be called u njust. U n fortunately this neat conversion o f Plato into a p ro te sta n t hero does not square w ith th e intolerance o f dissent evinced in th e L a w s and in particular w ith th e attitude to atheists adopted in Book X. I t is also unlikely th a t P la to ’s contem poraries w ould have d escrib ed conscientious b u t m isguided actions as ‘involuntary in justice’. (2) T h e ‘opinion of w hat is b e st’ m eans the ‘correct conception of th e good’ and ‘if it goes w rong som ehow ’ means ‘if it is erroneous in points o f d etail’. T h e character described in the passage would th en be som eone w hose passions and desires are correctly o rd ered in accordance w ith the judgem ents o f reason b u t who makes incorrect decisions on some m inor m atters. I t is possible th at Plato w ould be



sym pathetic to such people, b u t again it is im plausible to suppose th at his contem poraries would have described them as ‘involuntarily unjust’. (3) T he ‘opinion of the best’ means ‘the correct conception o f the good’, b u t ‘if it goes w rong som ehow ’ means ‘if it happens to ad o p t the wrong m eans’. T he person described would th en be som eone o f good character who goes w rong through a m isjudgem ent o f the p articu lar facts. T he doctors who fail to save their patients’ lives (865a-b) m ight be an example. This interpretation w ould give the best sense to the suggestion that the people here described w ould often be called ‘involuntarily u n ju st’. A fter all, this is precisely the kind o f case the Athenian seems to have in m ind when he distinguishes betw een injustice and harm (861e-862c). It also makes the A thenian genuinely fufil his prom ise at 863a to explain, n o t only th e n atu re of justice and injustice, but also the distinction betw een the voluntary and the involuntary. His position would be that injustice m ay arise from a disorder o f the lower soul or from an erroneous conception o f the good. Where, however, a person goes wrong through ignorance o f the particular, he will not be called unjust even though he does harm . 864b-c This purports to be a resum e of w hat has gone before. Again we are told th at there are three m ain kinds of error though these are described in a slightly different way. T he first kind is painful and is called anger or fear. T h e second kind consists of pleasures and desires. T h e th ird is clearly meant to be ignorance, though a corruption o f the text makes the exact description unclear. Saunders makes the A thenian describe this kind o f error as involving ‘hopes and opinions — a m ere shot at the truth about the suprem e good’. An alternative and perhaps preferable reading (following des Places and Dies) makes it consist in ‘the loss of hopes and true opinion about the good’. W hatever the correct version, it does not look as though the passage is intended to add anything to what has gone before. Ignorance is then divided, in accordance with 863c-d. into (a) simple ignorance, (b) double ignorance with power and (c) double ignorance w ithout power. All acts m ust also be classified into those done openly w ith violence, those involving secrecy and fraud, and those involving a com bination of the two m ethods. T he point is apparently that judges m ust take all these points into account in deciding how to treat criminals, b u t there is no explicit reference to these categories in the ensuing sections of th e penal code. Presum ably these m atters are to be left to the court’s discretion.




R esp o n sib ility in p ra ctice

T h e account of th e law concerning hom icide and w oundings and assaults, w hich occupies the second p art o f Book IX (864c-882c), illu strates th e practical application o f P lato’s theory o f responsibility. T h e first cases dealt w ith are those w here the accused is insane, senile o r infantile. T hese people are not to be punished, th o u g h they will be re q u ire d to give com pensation and, if they have killed som eone, they w ill have to sp en d a year in exile to avoid pollution (864c-e). Plato o fte n talks as th o u g h injustice is a kind of extrem e folly or stu p id ity or even a m ental disease. T his passage shows th at he does not regard the u n ju st as literally insane; he is prep ared to distinguish the m ad from th e bad. T h e A thenian divides b o th hom icides and w oundings into three m ain categories: th e v oluntary, the involuntary and those done in anger. H e now here gives a precise account of the criteria w hich w ould justify us in regarding an act as involuntary, though in the course of the legislation he does give som e exam ples of the kind of thing he has in m ind. C ertain categories of exile will be deem ed to have retu rn ed in v o lu n tarily if they are shipw recked on the coast or b ro u g h t in forcibly by others (866c-d). C itizens will be excused for losing their w eapons in w ar if th is happens as a result o f being knocked unconscious, being th ro w n from a height or being overw helm ed by ru sh o f w ater d u rin g a sto rm at sea (943d-944d). In all these cases the acts are judged involuntary because they result from an overw helm ing external force. Elsew here th e A thenian pays m ore attention to the circum stances th an to th e causes o f the involuntary act. T here are special exem ptions for those who involuntarily kill som eone (other th a n an enem y) in a contest or in public gam es, in war or m ilitary train in g , and for doctors whose patients die. As those who killed in these circum stances w ould be m ore likely to do so through accident, m istake or ignorance th a n as a resu lt of an external force, it looks as th o u g h th e A thenian w ould recognize a fairly wide range o f excuses. T h e m ost su rp risin g feature o f this p art o f the L a w s is the treatm en t o f killings and w oundings done in anger (866d-869e, 878b-879b). T h ese are placed betw een the voluntary and the involuntary. In the case of killings in anger there is a fu rth er subdivision. T h e first category includes those done w ith o u t prem editation on the im pulse of th e m o m en t and w here the killer im m ediately repents. T he second category includes cases w here one person seeks to kill another in revenge for some provocation and where there is no repentence (866de). T h e crucial difference lies in the presence or absence of



prem editation. Acts done with prem editation are m ore like those done voluntarily. T hose done w ithout prem editation are m ore like the involuntary and may be treated m ore leniently. T h e basic penalty for the first category (killings in anger w ithout p rem editation) is two years’ exile. T he basic penalty for the second category (w ith prem edi­ tation) is three years’ exile. Both penalties would be very lenient com pared w ith those for voluntary killings which norm ally carry the death penalty. But the A thenian recognizes that these rough and ready categories may not do justice to the particular circum stances. So the guardians o f th e laws may, after due investigation, prolong the term of exile indefinitely (867c-868a). T he treatm ent o f woundings is in general like th a t o f killing but im ports some problem s of its own (874e, 876e-877c, 878b-879b). The A thenian first suggests a fourfold division into those done (a) involuntarily, (b) in anger, (c) th ro u g h fear and (d) voluntarily and with prem editation (874e). But the detailed legislation does not give any special place to woundings done in fear. Presum ably these are to be lum ped together w ith those done in anger. T here is no parallel for the distinction made in the case of killings betw een acts done in anger with and w ithout prem editation. Perhaps this is d ro p p e d for the sake of brevity. The A thenian’s account o f killing in anger has been exam ined in detail by A. D. Woozley (1972) whose m ain concern is to discover the exact m eaning of hekcn and ak o n , the words I have been translating ‘voluntary’ and ‘involuntary’. A lthough translators have often rendered these as ‘intentional’ and ‘u nintentional’ W oozley shows that they cannot m ean exactly w hat a m odern jurist or philosopher o f m ind would mean by those terms. T he crucial point is that, although the A thenian thinks of those who act in anger as being betw een the hekon and the akon , we certainly could not say th at these acts w ould be betw een the intentional and the unintentional. M ost acts done in anger are fully intentional. Equally the A thenian seem s to think that prem editated acts are more like those done hekon than are unprem editated ones, but prem editation is not relevant to w hether an act is intentional. For rather m ore complex reasons, Woozley also argues that hekon and akon cannot m ean exactly the same as ‘voluntary’ and ‘involuntary’. He eventually concludes th at an act is done hekon if the agent could have helped doing it and akon if he could not. Woozley’s article is w orth reading for its precise analysis o f English words such as ‘intentional’ and ‘voluntary’. But as an account of the L a w s it is flawed by tw o assum ptions. T he first is th a t one can expect a high degree of precision in Plato’s use of the words hekon and akon. It is




R esp o n sib ility in p ra ctice

T h e account o f the law concerning hom icide and w oundings and assaults, w hich occupies the second p a rt o f Book IX (864c-882c), illu strates th e practical application of P lato ’s theory of responsibility. T h e first cases dealt w ith are those w here the accused is insane, senile o r infantile. T hese people are not to be p unished, though they will be re q u ire d to give com pensation and, if they have killed som eone, they w ill have to spend a year in exile to avoid pollution (864c-e). Plato o fte n talks as th o u g h injustice is a kind of extrem e folly or stupidity or even a m ental disease. T his passage shows th a t he does not regard the u n ju st as literally insane; he is p rep ared to distinguish the m ad from th e bad. T h e A thenian divides b o th hom icides and w oundings into three m ain categories: the voluntary, the involuntary and those done in anger. H e now here gives a precise account of the criteria w hich would justify us in regarding an act as involuntary, though in the course o f the legislation he does give som e exam ples of the kind o f thing he has in m ind. C ertain categories o f exile will be deem ed to have retu rn ed inv o lu n tarily if they are shipw recked on the coast or brought in forcibly by others (866c-d). C itizens will be excused for losing their w eapons in w ar if this happens as a result of being knocked unconscious, being th ro w n from a height or being overw helm ed by ru sh o f w ater during a storm at sea (943d-944d). In all these cases the acts are judged involuntary because they result from an overw helm ing external force. Elsew here th e A thenian pays m ore attention to the circum stances th a n to th e causes of th e involuntary act. T here are special exem ptions for those who involuntarily kill someone (other th a n an enem y) in a contest or in pub lic games, in w ar or m ilitary training, and for doctors whose patients die. As those who killed in these circum stances w ould be m ore likely to do so through accident, m istake or ignorance th a n as a resu lt of an external force, it looks as th o u g h th e A thenian w ould recognize a fairly wide range of excuses. T h e m ost surprising feature of this p art of the L a w s is the treatm en t o f killings and w oundings done in anger (866d-869e, 878b-879b). T h ese are placed betw een th e voluntary and the involuntary. In the case o f killings in anger th ere is a fu rth er subdivision. T he first category includes those done w ith o u t prem editation on the im pulse of th e m om ent and where the killer im m ediately repents. T he second category includes cases w here one person seeks to kill another in revenge for some provocation and w here th ere is no repentence (866de). T h e crucial difference lies in the presence or absence of



prem editation. Acts done w ith prem editation are m ore like those done voluntarily. Those done w ithout prem editation are m ore like the involuntary and may be treated more leniently. T h e basic penalty for the first category (killings in anger w ithout prem editation) is two years’ exile. T h e basic penalty for the second category (with prem edi­ tation) is three years’ exile. Both penalties w ould be very lenient com pared with those for voluntary killings which norm ally carry the death penalty. B ut the Athenian recognizes that these rough and ready categories may not do justice to the particular circum stances. So the guardians of th e laws may, after due investigation, prolong the term of exile indefinitely (867c-868a). The treatm ent of w oundings is in general like th at o f killing but im ports some problem s of its own (874e, 876e-877c, 878b-879b). T he Athenian first suggests a fourfold division into those done (a) involuntarily, (b) in anger, (c) through fear and (d) voluntarily and w ith prem editation (874e). But the detailed legislation does not give any special place to w oundings done in fear. Presum ably these are to be lum ped together w ith those done in anger. T here is no parallel for the distinction made in the case of killings betw een acts done in anger w ith and w ithout prem editation. Perhaps this is d ro pped for the sake of brevity. The A thenian’s account of killing in anger has been exam ined in detail by A. D. W oozley (1972) whose main concern is to discover the exact m eaning o f hekon and a k o n , th e words I have been translating ‘voluntary’ and ‘involuntary’. A lthough translators have often rendered these as ‘intentional’ and ‘unintentional’ W oozley shows that they cannot mean exactly w hat a m odern jurist or philosopher of m ind would m ean by those term s. T he crucial point is that, although the A thenian thinks of those who act in anger as being betw een the hekon and the akon , we certainly could not say that these acts would be between the intentional and the unintentional. M ost acts done in anger are fully intentional. Equally the Athenian seems to think th at prem editated acts are more like those done hekon than are unprem editated ones, but prem editation is not relevant to w hether an act is intentional. F or rath er more complex reasons, W oozley also argues that hekon and akon cannot mean exactly the same as ‘voluntary’ and ‘involuntary’. H e eventually concludes th at an act is done hekon if the agent could have helped doing it and akon if he could not. Woozley’s article is w orth reading for its precise analysis of English words such as ‘intentional’ and ‘voluntary’. But as an account o f the L a w s it is flawed by two assum ptions. The first is that one can expect a high degree of precision in Plato’s use of the words hekon and akon. I t is



tru e th a t the English term s W oozley considers have been given very p recise m eanings b u t only lawyers and philosophers adhere to these at all closely. T h e m an in the street uses ‘v o luntary’, ‘in tentional’ and the like very freely, often treatin g th em as equivalents. T he G reek words in q u estio n were, if anything, used even m ore freely, particularly since th e re was no tra d itio n of ju risp ru d en ce or philosophy th at could preserv e precise m eanings for such term s. T h e second dubious assum ption is th at P lato’s concerns are m uch th e sam e as those o f m o d ern jurists. M odern legal practice distin g u ish es fairly sharply betw een tw o questions: (a) ‘D id the agent do th e act alleged?’ and (b) ‘Should he th erefore be punished, and if so h o w ?’. In th e A nglo-A m erican system questions of the first type are fo r th e jury to decide. In m ost circum stances the jury can find the accu sed guilty only if it has been show n th a t he acted know ingly and in ten tio n ally . T h e accused may defend him self by arguing, for exam ple, th a t the act was accidental and was not therefore done in ten tio n ally . O nce the accused has been found guilty, the judge has to d ecid e w hat is to be done to him . Judges often have great freedom in choosing a sentence — from an absolute discharge to im prisonm ent. T h e accused m ay attem p t to lessen th e sentence by m aking a plea in m itig atio n , arguing, for exam ple, th a t he acted un d er provocation or th a t he had hitherto led a blameless life. M o d ern practice assum es at least an elem ent of retributivism : it is based on the assum ption th a t m en deserve punishm ent if and only if th e y have intentionally done w rong. So the question o f pu n ish m en t arises only after it has been fo u n d th a t th e accused intentionally broke th e law. T h e penal code o f th e L a w s , on the other hand, makes no d istin ctio n betw een the question w hether the agent intentionally did w ro n g and th e question o f how he should be punished. T his is w hat we w o u ld expect in view o f P lato’s theory o f punishm ent. H e rejects any elem en t o f retributivism and looks on pu nishm ent as a device for doing social good, chiefly by curing th e crim inal. F rom this perspective the basic q uestion is sim ply ‘W hat form of treatm en t will do m ost good?’ In considering this question one w ould expect him to distinguish three m ain kinds o f case. F irst th ere are cases w here what has happened is in n o sense indicative of any fault in th e agent. This m ight be either because the act did not arise from his character at all or because he acted in ignorance of th e facts. T h e n th ere are acts w hich are fully indicative o f serious faults o f character. In betw een these will come th o se acts w hich are signs of a fault in th e agent b u t not such a serious fau lt as m ight norm ally be supposed. W ithin this category there will obviously be m any graduations.



The A thenian’s treatm ent of hom icide makes it d e a r th at he has this kind of point in m ind. H e begins by discussing involuntary hom icide. As we have seen, he does not explore w hat makes an act involuntary but it looks as though he has in m ind those done u n in ten tio n ally or through excusable ignorance. At the end o f the discussion o f hom icide he also considers cases which we w ould call justifiable hom icide. Those who kill in these circum stances will be treated as innocent. T he crucial point about both these categories is presum ably th a t they are no evidence o f a bad character on the part o f the agent. Between 869e and 873b there is a discussion o f voluntary hom icide. The A thenian makes it clear th at these are acts of hom icide arising out of a depraved character. T hey involve all kinds o f injustice and are prem editated — they result from yielding to pleasures, desires and envy (869e). T h ere follows a digression on th e main causes o f injustice — inordinate desire for riches, am bition and guilty fears (869e-870d). Homicides in anger come betw een the voluntary and the involuntary, because an act in anger is not n o rm ally indicative of totally depraved character. T he tw o classes the A thenian distinguishes here are those done on the spur o f the m om ent as the result o f sudden emotion and those w here the agent has exacted p rem editated revenge for some provocation. It is entirely natural to suppose th at the second category would generally be sym ptom atic of a more serious fault of character than the first. But both would be treated by a m odern court as cases where there were m itigating circumstances. This interpretation is confirm ed by w hat m ight otherw ise seem to be some of th e A th en ian ’s m ore peculiar proposals. K illings in selfdefence, for exam ple, are treated as a subclass o f those done in anger and carry the same dim inished penalties, w hich is reasonable enough if this category is intended to cover all cases where there are m itigating circum stances. But the norm al rule that those who kill in anger receive lesser penalties is not applied to those w ho kill th eir parents even in self-defence; th e same goes for slaves who kill their m asters. This would be nonsense if the A thenian’s prim ary concern was the degree to w hich the agent could or could not help doing w hat he did. The anger a son feels tow ards his father or a slave tow ards a free m an is no less pow erful an em otion th an any other. T he p o in t m ust be th at in these cases penal policy cannot allow provocation as a m itigating circum stance. K illing one’s father is such a terrible thing that one should p u t up with anything rather than com m it this crime. So parricide even un d er provocation is evidence o f a particularly vicious nature. In the case of slaves who kill free men, the A thenian may be thinking not so m uch of the depravity o f character revealed in such



a cts, as o f th e need at all costs to prev en t slaves becom ing insubordinate. A n o th er anom aly, on W oozley’s in terp retatio n , w ould be in the A th e n ia n ’s tre a tm e n t o f fear as a cause o f crim e. In the account of w o u n d in g s, acts done in fear are treated alongside those done in anger, as com ing betw een th e voluntary and the involuntary. B ut in his tre a tm e n t o f killings he expressly treats m urders done th ro u g h a guilty fear, as in th e case o f a crim inal w ho kills to conceal his crim e, as being p rim e exam ples o f th e voluntary (870c-d). Yet guilty fears are no less co m p ellin g th a n innocent ones. T h ere are, o f course, perfectly good reasons, w hich a m odern cou rt w ould easily u nderstand, for treating th e tw o acts differently. T h e guilty fear in itself results from depravity o f character and there are also good d eterren t reasons for being especially h ard on this kind o f killer.


R esp o n sib ility a n d th e tre a tm e n t o f crim inals

P la to ’s handling o f th e problem s o f responsibility resem bles in some resp ects th a t o f B arbara W ootton (1963). B oth authors start from the assu m p tio n th a t th e aim o f the penal process is to treat crim e, and that th erefo re th e only q uestion relevant in deciding w hether and how to p u n ish som eone, is ‘W hat sort of pun ish m en t w ould cure the co n d itio n w hich led to this crim e and p rev en t its rep etitio n ?’ W ootton accordingly argues th a t th e n otion o f responsibility m ust be elim inated. T h e concern o f th e courts should be sim ply to discover w h eth er th e acts alleged to have occurred really did occur. T he crim in al sh o u ld th e n be h an d ed over to a body of experts to determ ine how he should be treated. P la to ’s proposals seem less radical b u t rest o n sim ilar assum ptions. Q uestions o f the voluntariness or involuntariness o f th e crim e are not particularly relevant in them selves. W hat counts is w h eth er the act was evidence of depravity. T h e aim sh o u ld th en be to find an ap propriate way of treating this depravity. Plato does n o t propose the establishm ent o f separate panels o f psychiatrists or th e like to consider this question, b u t he does think th a t ideally it w ould be left to th e discretion o f a body o f wise and ex p erien ced people. T h e rules he suggests seem to be intended prim arily as guidelines for the use o f these ‘experts’. W o o tto n ’s proposals have been heavily criticized, largely because th e y w ould lead to consequences w hich m ost o f us w ould find morally unacceptable. O ne particu lar objection is th at the length and un pleasantness o f tre a tm e n t need n o t correspond to the seriousness of



the condition it is supposed to cure. I t is quite conceivable th a t one who had com m itted a dreadful m urder could be treated quickly and easily while one who had com m itted a m uch more m inor crim e would require along and unpleasant course o f treatm ent. In this respect Plato does not seem to see the im plications o f his own position. A lthough he claims that any kind o f treatm en t m ay be considered, he in practice takes it for granted th at the length and unpleasantness o f the sentence will be proportional to the degree o f depravity involved in the crime. This is why those who kill in anger on the spur of the m om ent get lesser sentences. T he Athenian could perhaps defend him self by claiming that more serious crimes require heavier sentences for d eterren t reasons but this is not necessarily so. As B entham saw, the logic of deterrence is that, where th ere were especially strong m otives for a crime (e.g. some form o f provocation), sentences should be increased in order to outw eigh th e tem ptation. So the considerations traditionally regarded as m itigating circum stances should really be seen as aggravating the offence. T he fact th a t Plato fails to see these points suggests th a t he has not seen th e full im plications of his own position and has not entirely em ancipated him self from retributivism .


Religion L a w s, 6 2 4 a -6 2 5 b , 7 1 5 e -7 1 8 c , 7 3 8 b -e , 7 5 9 a -7 6 0 a , 8 2 8 a -d , 88 4 -9 1 0 d ; P ka ed ru s, 2 4 5 c -2 4 6 a ; Tim aeus, esp ecially 2 7 c -4 7 e , 6 8 e -6 9 d , 92c; P hilebus, 28 c-3 0 d ; E pinom is

I t is n o t by accident th a t th e w ord iheos, ‘god’ is placed em phatically at th e beginning o f th e first sentence o f the L a w s . T he A thenian and his com panions are on a pilgrim age to th e cave w here M inos was supposed to have received th e laws o f C rete from no less a being than Zeus him self. T h e message is p retty obvious: law-giving is a religious task an d m ust be g rounded on an adequate theology. This religious em phasis is m aintained th ro u g h o u t the dialogue. It is particularly evident in th e provisions m ade for th e foundation of tem ples (738b-e, 761c, 778c-d, 848c-d), for th e choice priests and other religious officials (759a-760a) and for th e holding o f festivals (771d, 828a-d). T h e m ost striking expression o f religious sentim ent is in 715e-718c, a passage th at reads m ore like a piece o f prophecy th an sober philosophy. Its doctrin e, in so far as it is capable o f sum m ary, is th at G o d controls everything an d is th e m easure o f all things. H is constant com panion is justice. A nyone w ho wishes to be truly happy m u st make h im self like G od by being virtuous. T he wicked cut them selves off fro m G od. T h e first object o f w orship m ust therefore be the gods o f O lym pus, th e n those of th e underw orld, spirits, heroes, household gods and finally o n e’s paren ts. In other w ords, although the beginning o f th e speech may suggest m onotheistic ideas to us, the A thenian is len d in g his su p p o rt to traditional G reek religion — Plato never seems to w orry over th e question w hether there is one god or many. In Book X the supposed need for a law against atheism provides the o p p o rtu n ity for a m ore philosophical approach to religion. By way of a pream ble to this law the A thenian offers a long and detailed proof o f th e existence o f a god or gods.




The dangers o f atheism

The first few pages of Book X (884a-891d) serve as an intro d u ctio n to the m ain argum ents on the existence and nature o f the gods. T heir chief purpose is to establish th a t these m atters really are o f concern to the legislator. They do this by arguing th at atheism or incorrect religious belief is a serious cause of vice, and can resu lt from intellectual error as m uch as from m oral corruption; it is therefore the duty of the legislator to convert the unbeliever. At one level th e section is profoundly depressing. T he old m en give vent to th e ir prejudices, assuming that atheism is necessarily connected w ith vice, th a t it is a product o f youthful pride and folly and th a t it m u st be suppressed. But, even so, the A thenian does insist, against the m ore sim plem inded C retan, th a t th e existence o f the gods cannot be taken as obvious (886a). T here is a serious intellectual case for atheism and this must be m et in its own terms by argum ent. Two passages in this in troduction are w orth special attention. The first is at 885b, where the A thenian m aintains th a t those who u tter impious words or do impious deeds can do so only because they believe one or other o f three erroneous doctrines: (a) th at there are no gods, (b) that th e gods are indifferent to th e affairs of m en, or (c) th at the gods can be won over by prayers and sacrifices. T h e lawgiver m ust convince these people of their errors by persuading them (a) th at there are gods, (b) that these gods are concerned w ith the affairs o f m en and (c) that they cannot be bribed. T he A thenian him self follows this plan in 891e-907b. I t is not surprising that a respectable G reek should insist th at the gods exist and are not indifferent to hum an affairs. B ut th e attack on the idea that th e gods can be b ribed, like the criticism o f poets who represent the gods as behaving im m orally (886b-c; cf. R ep u b lic, 377b392c), shows Plato as a reform er aware of som e of th e less acceptable aspects of conventional G reek religion. T he passage recalls R epu blic , 362d-367e, w here A deim antus lists the same three varieties of erroneous belief about the gods. But there are im p o rtan t differences in the use of the two dialogues make of this classification. In the R epublic Adeim antus calls on Socrates to show th a t there are other reasons for being just besides the fear of divine retribution. In th e present passage, by contrast, the A thenian assumes that people cannot be expected to behave them selves w ithout this fear. So although the L a w s officially agrees w ith th s R epublic in m aintaining that virtue is the greatest good, it concedes th at, in practice, sanctions are still necessary. The second passage to dwell on is the A th en ian’s statem ent of the



case for atheism (888e-890b). T h e atheists, as he represents them , b e g in by dividing all things into th ree categories depending on w h e th e r they com e ab o u t by (a) nature (phusis ), (b) chance ( tuche ) or (c) a rt (techrie ). A ccording to th e A thenian the atheists a ttrib u te m ost th in g s in th e universe to n ature or chance. A rt in th eir view is only a seco n d ary cause an d its pro d u cts, w hich include politics and legislation, are o f little im portance. F rom this they conclude th at the gods, as products o f art, have no real existence, th at the good according to th e law (a m atter o f art) is different from the good according to n a tu re , th a t justice is a hu m an invention liable to constant controversy a n d change, and th at natural rightness consists, n o t in subduing oneself to o thers, as the law dem ands, b u t in m aking oneself m aster over them . A ttem pts have been m ade to identify a particular thinker or thinkers w ho m ight be in P lato ’s m ind here. I t seems m ore likely th at he is conflating the doctrines o f several different authors. His prim ary ta rg e t is th e im m oralism he had earlier p u t into the m ouths o f Callicles a n d T h rasym achus and w hich he supposed to be a direct consequence o f th e sophistic d istin ctio n betw een nomos and p h u sis. In this he has alm o st certainly been unfair. T h e belief th a t m orality is a hum an in v e n tio n does n o t in itself im ply th a t m orality is a th in g o f no im p o rtan ce, and it is likely th at many o f those who accepted the n om osp h u sis distinction would nevertheless have su pported custom ary stan d ard s o f behaviour — Protagoras, for exam ple, may well have taken this view. As P lato sees it, this supposed im m oralism rested on the attem pts of th e p re-S o cratic philosophers to investigate th e universe in a purely n atu ralistic way w ith o u t reference to any divine agency. T h e basic assum ptions o f these philosophers, as Plato sees them , are em bodied in th e distinction betw een nature, chance and art. T hey conceived each th in g to have its ow n n atu re or ph u sis so th at w hatever arose o u t of its n a tu re m ight be said to be a p ro d u ct o f p h u sis. A rt, by contrast, includes everything th a t involves purpose or intention; b u t it may be said to make use o f n atu re in th e sense th a t we use th e natural p ro p e rtie s o f things to carry out o u r purposes. T he th ird category, th a t o f chance, includes everything n o t directly attributable to either n atu re or art. T h u s it w ould seem th a t the art o f the chariotm aker makes use o f th e natu re o f w ood and m etal, b u t if tw o chariots accidentally collide, th at happens by chance, because it is a direct result neither of n a tu re nor of art. T h e d istinction betw een natu re, chance and art th erefo re cuts across the m odern distinction betw een events w hich are an d events w hich are not causally determ ined. Som e events w hich we w ould regard as causally determ ined could be attrib u ted to chance.



The Athenian indicates at 889a that he is unw iljing to accept the division of the w orld into the products o f nature, chance and art. T he reason soon becomes clear: he wants to m aintain th a t everything is ultim ately th e prod u ct of art and th at the threefold classification is therefore misconceived. In o th er words he is going to argue th at the world cannot be understood w ithout giving a fundam ental place to notions o f intention, purpose or design. B ut the way in w hich he expresses this aim sounds curious to o u r ears. H e is going to dem onstrate th at the first existents are not th e elem ents o f fire, earth, air and water to which the atheists give the nam e ‘n a tu re ’, b u t soul and its works which are to be term ed ‘art’. It is these th at are truly natural (892c-893b). H e attem pts to prove this by m eans of an argum ent from change (893c-897b) that in some respects resem bles A ristotle’s argum ent to the first mover and A quinas’ ‘first w ay’. (T he most im portant difference is that A ristotle and A quinas b o th believe that the first mover m ust itself be unchanged: G od therefore moves the universe as a final cause by being the object of desire. In Plato, on the other hand, ‘soul’ moves itself as well as o th er things and thus seems to operate as an efficient cause.) Having established in this way that soul has priority, th e A thenian goes on to argue that the order o f the universe shows it to be governed by a soul th a t possesses goodness and rationality (897b-899b).


The p rio rity o f soul

The argum ent for the priority o f soul (893c-897b) appears to have the following as its m ain stages: (1) (2)


(4) (5) (6)

Some things are at rest, others in m otion (893b-c). T here are two main kinds of m otion: (a) th a t w hich moves other things but not itself, and (b) th at w hich moves itself as well as moving other things (894b-c). (a) T he first m otion cannot be the kind th at moves other things w ithout m oving itself (894e). (b) I f there was a state o f com plete rest then the kind of m otion that moves itself w ould have to be the first to arise out o f th at state (895a-b). T herefore th e first kind of m otion is that w hich moves itself (895b). A thin g th at moves itself is said to be alive (895c). T h at which has soul is alive (895c).

170 (7) (8) (9) (10) (11) (12)


Soul is, by definition, th e m otion that moves itself (895e-896a). T herefore soul is the first source of m ovem ent and change in everything (896b-c). T herefore soul is prior to m aterial things (896b-c). T herefo re th e attrib u tes o f soul are also p rio r to m aterial things (896c-d). Soul, since it is th e source o f everything, m ust be the source of values — good and bad, fair and foul, just and unjust (896d). Since soul dwells in and governs all m oving things, it m ust govern the universe (896d-e).

P rem ise (1), th a t som e things are in m otion and others at rest, may n o t be the tru ism it seems. Parm enides had denied the reality of m o tio n w hile H eraclitus m aintained th a t everything moves. O n the o rth o d o x in terp retatio n Plato ’s ow n m iddle dialogues m aintain th at th e Form s, w hich are th e only reality, are unchanging and unm oving. P rem ise (2), the distinction betw een th e m otions th a t are and are not self-m oving, is reached after a detailed classification o f kinds of m otion a n d change. T hese appear to be (a) m otion in one location (i.e. circular m o tio n like th at o f a top w hich revolves while staying in one place ), (b) m otion in m any locations (i.e. w hen som ething moves from place to place), (c) com bination, (d) separation, (e) increase, (f) d im in u ­ tio n , (g) generation, (h) destruction, (i) self-m oving m otion, and (j) m otion th at is not self-m oving. T w o points are w orth noting in this o b scu re passage. F irstly the A thenian talks as though all kinds of chan g e h appen as th e result o f th e m ovem ent o f particles. So u n d er the gen eral heading ‘m otion’ he w ould seem to include anything w hatever th a t may be said to happen. T h e second point is th at the kinds of m o tio n are not m utually exclusive. In particular, any of the first eight m otions could be either self-m oving or caused by some other m otion. So ‘self-m oving’ and ‘non-self-m oving’ are basic categories into w hich all m otion and changes can be divided. T h e non-self-m oving m otion has the power to move other things b u t n o t itself, while th e self-m oving m otion can move both o ther th in g s and itself. We m ight p u t th e point by saying th at the non-selfm oving m otion results from a cause other th an itself while the selfm oving m otion does not. U n d ersto o d in this way it is a tru th o f logic th a t all m otions are eith er self-m oving or non-self-m oving, b u t the term s carry w ith them the seeds o f dangerous confusion. T he obvious exam ples o f self-m oving m otions are the m ovem ents o f hum an beings an d anim als w hich d ep en d on know ledge, desires, intentions and the



like — i.e. they involve agency. B ut i f ‘self-m oving’ is taken to imply agency, it is no longer a logical tru th th at all m otions are either selfmoving or moved by som ething else. Logically speaking, there is no reason why a totally inanim ate thing should not sta rt spontaneously into motion. Stages (3) and (4) offer two co-ordinate argum ents to show th at the first m otion m ust be the self-m oving variety. T he first arg u m en t (via 3a) makes the unw arranted assum ption that there m ust be a first motion, i.e. th at there cannot be an infinite regress of m otions. The second argum ent (via 3b) assumes that rest is the natural state o f things and that m ovem ent requires explanation. This w ould be reasonable enough if one accepted the very natural belief that things in m otion are always tending to slow down and stop. T here w ould th en be need for some spontaneous motion to keep the universe moving. B ut, of course, modern physics rejects this belief. Rest stands in need of explanation as m uch as motion. The argum ent in (5), (6) and (7) is confused. We m ight expect the Athenian to argue: ‘T o be self-m oving is to be alive; to be alive is to have a soul; therefore w hatever is self-m oving has a soul’. W hat he in fact says is: ‘(5) W hatever is self-m oving is alive; (6) w hatever has soul is alive; (7) soul is by definition self-m oving m o tio n .’ O n this form ulation (5) and (6) are redun d an t because ;7) is established directly by appeal to the supposed definition o f soul. This obscures an im portant weakness in the argum ent. I t is probably true th a t w hatever is alive is, in some sense, self-moving, b u t it does not follow that whatever is self-m oving is alive — th at would be true only if we took ‘self-m oving’ in a m ore restricted sense than it has borne so far, so that it im plied not merely spontaneous m ovem ent b u t a gen cy. Sim ilarly, although everything that has a soul may be alive, it does not follow that everything w hich is alive has soul, unless, that is, one takes soul in such a wide sense that lower animals and even plants have souls. In Tim aeus, 77a-b, plants are credited with the lowest (appetitive) kind of soul, but for the purposes of th e present argum ent Plato needs a more restricted sense of ‘soul’ which implies intelligence. A notable feature of this passage is an attem pt to elucidate the notion of tru th by definition (895d-e). T he A thenian distinguishes (a) a thing’s being or essence (ousia ), (b) its definition or account (logos), and (c) its nam e. T he definition and the nam e both refer to the same thing. T hus, since the definition o f ‘soul’ is ‘th e m ovem ent th at moves itself we can ask for no stronger proof that the soul is the first source of movement.



F ro m (8) onw ards th e arg u m en t is hopelessly em broiled in the am b ig u ities o f ‘self-m oving’ and ‘soul’. Stage (8) itself, ‘Soul is the first s o u rc e o f m ov em en t’, is su p p o sed to be derived from (4), ‘T h e first m o tio n is th e self-m oving k in d ’, tog eth er w ith (7) ‘Soul is the m otion th a t m oves its e lf. B ut in (4) ‘self-m oving’ has to m ean sim ply ‘moving w ith o u t a prior cause’ w hereas in (7) it has to m ean som ething like ‘b e in g alive’. T o m ake m atters w orse, m ore and m ore is b u ilt into the c o n c e p t o f soul. U p to this p o in t soul has been treated as a kind of m o tio n b u t in (9), ‘Soul is prior to m aterial th in g s’, it is suddenly tre a te d as an in d ep en d en t substance, as though there could be m otion in d e p e n d e n tly of anyth in g ’s m oving. Stage (10) treats th e soul not sim p ly as th a t w hich makes things alive b u t as essentially involving ra tio n a l capacities. In (11) we learn th a t because soul is the cause of e v e ry th in g it m u st be th e cause of things good and bad, fair and foul, ju s t a n d un ju st, i.e. of all values, b u t we have been given no reason for s u p p o sin g th a t there are values and no explanation of w hat kind of e x iste n c e they have. Stage (12) moves illegitim ately from the claim th a t soul moves everything to th e claim th at it dwells in and governs th e universe. I f th is analysis is co rrect the arg u m en t for the priority of the soul b re a k s dow n for two m ain reasons. F irstly , it fails to take seriously the p o ssib ility o f an infinite regress o f m otions. Secondly, it confuses ‘selfm o v in g m o tio n ’ w ith soul in th e traditional sense o f an independent su b sta n c e capable o f exhibiting agency and personality.


F rom soul to g o d

By 896e the A thenian has reached the conclusion th at the universe is c o n tro lle d by soul. H e th e n tantalizingly asserts th at there m ust be at least tw o souls — one responsible for good and another for evil. It has so m etim es been supposed th a t Plato is here postulating the existence o f a b ad world soul, b u t this reading has no real su p p o rt in the text. It w o u ld be sufficient for th e A thenian’s purposes if there were m any im p e rfe c t souls who, betw een them , could account for the evil we see in th e w orld. But this passage obliquely refers to one o f the m ain p ro b le m s in P lato’s theology. H ow , on the supposition that the world is g o v ern ed by a ratio n al god or gods, can we account for evil? In the T im a eu s (42e-44d) th e doctrine seems to be th at there are disorderly m o v em en ts w hich resist th e control o f reason, b u t th en we have the p ro b le m , ‘W here do these m ovem ents come from ?’ I f we suppose th at th e y are in h eren t in th e m aterials o u t o f w hich the universe is



constructed, we have a counter-exam ple to P lato’s claim th at all m otion is ultim ately initiated by soul. T h e alternative is to postulate some kind of irrational soul, b u t this disrupts the strong association Plato sees between soul and rationality. We cannot tell how , if at all, he would solve the problem. The A thenian now turns to the question ‘W hat sort o f soul governs the universe? Is it wise and virtuous, or neither of th ese?’ (896e-897c). H e argues that the soul or souls that govern the universe m ust possess both wisdom and virtue because their m otion is m ost like th at o f nous (intelligence, reason or thought). T hey exhibit motion in one place (i.e. circular motion) w hich here, as in the Tim aeus (35a-37c, 40b, 42e44d), is associated with reason. T he connection, to say the least, is not obvious. T h e Athenian draws attention to the fact that circular m otion, like the working of reason, is always m ovem ent in the sam e respects, in the same m anner, in the same place, around the sam e things, in relation to the same things and in accordance w ith one rule and system (898b). So the fundam ental point seems to be th at both are, as we say, rule-governed. U nfortunately, we can concede that there is a certain parallelism here w ithout concluding that the universe is under the control of a rational soul. We now explain natural regularities, such as the movements of the heavens, by reference to laws of n ature w ithout presupposing th at any kind of intelligent activity lies b eh in d them . But even if we conceded that natural regularities result from the activity of some kind of intelligence we would be entitled to infer only th at the intelligence governing the m ovements o f the heavens is sufficient to understand these m ovements. T here would be no grounds for attributing any higher degree o f intelligence or anything approxi­ mating to m oral virtue. Although th e A thenian is em phatic that soul moves th e universe, he leaves open the question how soul does this (898e-899c). It could be that there are souls w ithin each o f the heavenly bodies. Alternatively soul m ight move the heavenly bodies externally by physical contact or by some other means. T hus he com m its him self to no particular view about the num ber of gods (one or many) or about their precise relations to heavenly bodies (w hether we should see these bodies as gods or as things under the control of gods). But he is sure that the soul or souls in question m ust be divine. In this sense he echoes the rem ark attributed to Thales, ‘Everything is full of gods’ (899b). 4

The significance o f the p ro o f

Book X cannot be intended as a com plete account of P lato’s theology.



I n th e T im aeu s, for exam ple, we learn about a ‘dem iurge’ or ‘c r a fts m a n ’ w ho is the creator b o th o f th e w orld soul and o f souls for th e heavenly bodies. T h e P h ile b u s (28c-30d) appears to offer a similar p ic tu r e . T h e L a w s gives only fragm ents o f this system , though there a re h in ts o f w hat lies in th e background. At 904a the A thenian speaks o f a ‘K in g ’ who governs the universe. T his eternal ‘K in g ’ is d istin g u ish e d from th e gods recognized by law, who are not eternal th o u g h they are im perishable. Several tim es, too, there is a suggestion t h a t soul, although prior to m atter, has nevertheless come into being at so m e p o in t (892a; cf. 967d). T h e n atu ral inference is th a t the A thenian re a lly sees soul as the p ro d u c t o f th e sam e eternal creative pow er of re a s o n th at Tim aeus calls ‘the dem iurge’. T h e A thenian is also som ew hat reticen t about the world soul. It is n a tu r a l to see him as referrin g , in 896e-897a, to a single soul actuating th e universe as a whole, b u t, w hen he goes on to consider how soul m o v es th e heavenly bodies, he concentrates on the individual souls t h a t are sup p o sed to m ove bodies such as th e sun. All this w ould make se n s e if th e A thenian was presupposing the system of the Tim aeus w h ile being relu ctan t to say m uch about ideas as alien to conventional re lig io n as those o f the dem iurge and w orld soul. T h e re is a related problem ab o u t th e status of the traditional gods. E lsew h ere in th e L a w s th e A thenian stresses the im portance of tra d itio n a l pattern s of w orship and th e official object o f the present s e c tio n is to su p p o rt a belief in the gods recognized by law (885b). B ut th e argum ent o f Book X offers no reason for believing in the tra d itio n a l deities. T h e cosmology o f th e Tim aeus likewise has no real p lace for the gods o f popular religion, though Tim aeus (ironically?) su g g ests th a t these gods m ust exist since some people, who ought to k n o w th eir own ancestry, claim descent from them (40d-e). T he E p in o m is explicitly avoids laying dow n any particular doctrine about th e n a tu re of th e traditional gods, th o u g h it allows for the existence of b e in g s w ith bodies of air or aether w ho are interm ediate betw een o u rselv es and the gods o f the heavens (984d-985b). H ow ever this p assag e is suspect, because now here in the undoubtedly authentic dialogues does Plato treat aether as a separate elem ent. T here are thus tw o possible views o f the role o f th e traditional gods in the L a w s. One p o ssib ility is th a t they are view ed as interm ediate beings, b u t the m ore likely alternative is th a t th e A thenian accepts the language o f tra d itio n a l beliefs as a concession to the masses. O rdinary m en (in clu d in g , no doubt, Cleinias and M egillus) w ould have difficulty in u n d e rsta n d in g the sophistications of th e genuine Platonic theology; so,



for their sake, religious sentim ents m ust be expressed in m ythological terms. This distinction betw een Platonic and popular religion has a w ider importance. Plato has been seen as a reactionary in his attitu d e to scientific knowledge. T he great achievement of the pre-S ocratic philosophers was, we are told, to seek naturalistic explanations o f the world instead of appealing to m yths of divine intervention. Particularly adm ired by those who argue in this way is th e m aterialistic system o f the atom ists, D em ocritus and Leucippus. B ut Plato attacks the work of these philosophers precisely because of their preference for naturalistic rather than theological explanations. Is he not therefore urging a retreat into obscurantism ? In considering this charge is is vital to see th at P lato's deities, unlike the gods of popular religion, operate in strict conform ity to rational laws. T hus w hat excites Plato most about astronom y is th at the movements of the heavens are com prehensible in geom etrical term s. In his view, this rational order is the strongest evidence for the existence of gods and, conversely, to suppose that gods could act irregularly is blasphemy. Plato’s religion thus has the opposite effect from the conventional piety of the Greeks. W hile conventional religion encouraged m en to explain events in term s o f arbitrary divine activity, Plato’s religion requires us to seek the laws underlying natu ral phenomena. Indeed, since the sole characteristics Plato attributes to the gods are the desire and capacity to organize things in a rational way, his religion am ounts to little more than the conviction th a t the universe is ultim ately law-governed. This Platonic insistence th at the universe be explained by means of m athem atically expressable laws is as im portant for the developm ent of science as the pre-Socratic philosophers quest for naturalistic rather than m ythological explanations. T he fact that Plato expresses his belief in theological language makes no difference to its effect on scientific practice. WTe can see this also in the work of N ew ton and his followers. T hey were certainly influenced by the atomists b u t they were also actuated by a desire to provide th e m ost satisfying m athem atical account of the movements of the heavens. Like Plato, they supposed th at in doing this they were revealing the workings o f the divine m ind. 5

The g o d s’ concern fo r man

Once he has established the existence o f a god or gods, the A thenian turns his attention to those who say that the gods do not concern



them selves w ith hu m an affairs or th a t th e gods can be bribed. T he arg u m en t is quite sim ple. T h e gods know everything, have all power a n d are suprem ely good (901d-e). B ut if they neglected the affairs of m en , th a t could only be th ro u g h ignorance, inability or vice. So the gods m u st care for m en (901e-903a). Sim ilarly anyone who supposed th a t th e gods could be b rib e d w ould be regarding th e m as inferior to h u m a n governors, w hich is ab su rd (905d-907b). T h e difficulty w ith all th is is th a t no justification has been given for a ttrib u tin g to the gods th e characteristics they are now sup p o sed to have. At m ost, the earlier a rg u m en ts could establish only th a t th e gods have th e kinds of perfectio n needed to secure the o rd er o f the heavens. I t does not follow fro m this th a t they m u st therefo re have th e m oral perfections. Sim ilar p ro b lem s affect m ost o th e r system s o f n atu ral religion. T h e god o f the philosophers is not th e god o f conventional religious belief. T hose who hold th a t th e gods are indifferent to h u m an affairs will, th e A thenian supposes, have been led to this view by seeing th a t the w icked apparently prosper. T hey believe th at this could not happen if h u m a n affairs w ere u n d e r divine supervision and therefore conclude th a t th e gods have no concern for us (899d-900c). T he details o f the passage (903b-905d) in w hich Plato attem pts to deal w ith this version o f th e problem of evil are obscure, th o u g h the m ain outline is clear en o u g h . T h e ru ler o f th e universe arranges everything w ith a view to th e preservation and good of th e whole. So we m en are created not p rim arily for our ow n good b u t for th a t o f the universe o f w hich we are p a rts. W e should n o t, therefore, com plain of our situation, since it m u st be best b o th for us and for th e universe as a w hole. We all have m any existences, so it is a sim ple m atter for the ru ler of the universe to ensure that,¡ as our characters change, we are transferred to an a p p ro p ria te place in th e cosm os. I f we becom e virtuous we will go to a h ig h er place to live am ong b etter souls, and if we becom e vicious we w ill m ove to a lower region to join th e worse souls. B ut we ourselves are responsible for the ways in which our characters change. In th e background of this passage is the assum ption th a t in a rationally organized w orld the w icked m ust do badly and the virtuous m u st prosper. T his doctrine is shared by those who defend a retrib u tiv e theory of p u n ish m en t. B ut, since Plato rejects the re trib u tiv ists’ idea th a t penalties should be inflicted on people sim ply because they have done w rong, he cannot m aintain a conventional d octrin e o f divine retrib u tio n . An alternative, tried in the G orgias, 52 5 b -c, and the R ep u b lic, 619b-e, is to suppose th at divine p u n ish m e n t cures us o f vice or deters us from doing w rong in later existences. In view o f th e account o f p u n ish m en t in the L a w s , one



m ight expect a sim ilar theory to be tried here, b u t th e A thenian attem pts instead to show th at wicked souls do badly and virtuous ones prosper, not as the result o f being deliberately p unished, b u t as the natural and inevitable outcom e o f th eir own choices. O bviously it is im possible to offer any evidence for such a b elief — it has to be accepted, if at all, as a m atter of faith. But even so, there are problem s in P lato’s account. He quite illegitim ately takes it for granted that, ‘because of our com m on origin’, w hat is good for the universe as a whole m ust be good for each of us as individuals (903c) — a line of argum ent pervasive in P lato’s th o u g h t though nonetheless fallacious. T here is also a good deal of obscurity about the no tio n of souls m oving to a b etter or worse place: 904c-d uses the conventional language of rewards in heaven and punishm ent in H ades, b u t 904e-905a offers the neat idea th a t the punishm ent o f the wicked and th e rew ard of the virtuous may consist simply in being sent to live am ong souls like themselves and so receiving the same kind of trea tm en t w hich they give to others — a solution th a t w ould give us all the advantages of Heaven and H ell w ithout the difficulties norm ally im plicit in these ideas.


The law against im p iety

T he discussion o f religion so far has officially been m erely a prelude to the law against atheism . The law itself is based on a distinction betw een tw o kinds of atheist. T h ere are those who are basically honest but m isguided. T hey openly adm it their atheism b u t m aintain their sense of justice. But there are also those who com bine their atheism with a corrupt character. T hey keep their disbelief secret and are liable to impose themselves on the public in all sorts of ways, but particularly, it seems, by going in for bogus m agic and religious cerem onies (908b-e, 909b). T he penalty for the honest atheists is to be confined for at least five years in a special reform atory. T h eir only visitors will be m em bers o f the nocturnal council who will come to adm onish them . Those reform ed by this experience will be allowed to retu rn to a norm al life, b u t anyone reconvicted w ith be p u t to death. T he penalty for the corrupt atheists is solitary confinem ent for life. T h e proposed treatm ent of the honest atheist has been widely and rightly condem ned (though one eighteenth-century w riter did suggest th at it w ould be an appropriate way of dealing w ith D avid H um e). No d o u b t P lato expects the m em bers o f the n o ctu rn al council to convert the atheist through philosophical argum ent. B ut since the atheist’s



o n ly hope o f freedom (and indeed of survival) lies in agreeing w ith his in stru c to rs, th e conditions are hardly those o f a fair and free discussion. ‘B rainw ashing’ m ight be a m ore adequate description. I t is com m onplace to co n trast P la to ’s intolerance w ith the freedom o f th o u g h t exercised by his m aster Socrates. All th a t can be said in m itig atio n is th a t he does not req u ire adherence to the detailed doctrines o f a dogm atic religion.


The Closed Society of the Laws 1 P la to an d P opper Many readers will be familiar with the bitter onslaught launched against Plato by Sir K arl Popper in The O pen S o c ie ty a n d its E nem ies. The basis of P o p p er’s attack rests on his distinction betw een open and closed forms of society. Popper sees Plato as the first and m ost brilliant exponent of a philosophical tradition th at rejects the open and seeks to return to the closed society. The closed society, in its archetypal form at least, is characterized by ‘its subm ission to magical forces’ (1966, p. 1). It lives in a charm ed circle of unchanging taboos, o f laws and custom s w hich are felt to be as inevitable as the rising o f the sun, or the cycle of the seasons, or sim ilar obvious regularities of n atu re’ (p. 57). Its characteristic intellectual attitude is thus a naive m onism that fails to distinguish the m an-m ade and changeable laws o f society from the natural laws o f the universe (p. 59). Closed societies are collectivist in the sense th at they emphasize the significance o f some group or collective (e.g. the tribe), without which the individual is nothing at all (p. 9). T his collectivism finds strong expressions when w riters like Plato present an ‘organic’ view of the state treating the individual citizens as though they stood in the same relation to the state as do the parts o f a hum an or anim al body to the whole (pp. 79-81 ). T h e open society, by contrast ‘sets free the critical pow ers of m an ’ (p. 1) by recognizing the distinction betw een norm ative and natural laws. Its characteristic intellectual attitude is thus a critical dualism which appreciates that the laws and custom s of a com m unity, unlike the natural laws o f the universe, are open to criticism and can be deliberately changed. In this kind of society, individuals are genuinely confronted by personal decisions (pp. 60-1). T he collectivism o f the closed society is replaced by individualism , i.e. people are im portant as individuals, not only as tribal heroes or saviours (pp. 100-2,190). T he aim of the law is to protect the freedom of individuals, provided that



th is does not involve harm ing others (pp. 110, 114-15). P o p p e r’s criticism s o f Plato are based largely on the R ep u b lic, w hich P o p p e r sees as a b lu e p rin t for-political action. T h e L a w s , in his view, w as w ritten only w hen P lato had ‘given up hope o f realizing his p olitical ideal in all its glory’ (p. 102), b u t since he believes that it is still a c tu a te d by th e sam e principles, P op p er quotes from it freely, though selectively, in su p p o rt o f his conclusions. T his m istaken view o f the re la tio n sh ip betw een the R ep u b lic and the L a w s enables Popper to m ak e some lu rid charges against Plato. In the city o f the R ep u b lic, p h ilo so p h ers rule w ithout check and there is a strict class system based o n selective breeding. T h e whole system could only be im posed by fo rce. So P op p er accuses Plato o f advocating dictatorship based on violence, o f su pporting a caste society and o f being a racialist. O nce it is ap p re c ia ted th a t th c R ep u b lic is not a political program m e and th at one m u s t look to the L a w s for Plato ’s practical proposals, these charges lose m o st of th e ir force. B ut this does noth in g to underm ine P o p p e r’s fu n d a m e n ta l contention th a t Plato is an enem y of the open society, and a read in g of the L a w s may even stren g th en P o p p er’s case. By now no arg u m en t should be needed to show th a t the city o f the L a w s will be a closed society in th at it severely lim its personal freedom a n d initiative. As evidence, one could cite th e tight restrictions on the m o d e of life of th e citizens, including the pro h ib itio n o f trade and com m erce, th e rigid control o f m usic and th e other arts, and the law a g a in st atheism . T h e city will be alm ost literally a closed society in the sen se th at foreign contracts will be cu rtailed so far as possible (704d70 5 b , 950d-953e). T h e closed society, as described by P opper, is a traditional tribal society con tro lled by a system of law and custom th o u g h t to be as p e rm a n e n t and unchallengeable as th e laws o f the universe. Sim ilarly in th e city o f laws every p art o f life will be controlled by law, w ritten a n d u n w ritten . T h e intellectual m echanism s will be geared to d e m o n strate th a t these laws instantiate the same rational principles as th e m ovem ents of the heavens. C riticism o f the system will be p ro h ib ite d (except to th e old) and change will be nearly im possible. As in m ost trad itio n al societies, th e in terp retatio n of the law will lie w ith a b o d y o f elders (the guardians o f the laws, and the nocturnal council). I n all these respects the city o f th e L a w s will correspond closely to the k in d of trib al society envisaged by Pop p er. Indeed it w ould be m uch m o re like a traditional society than w ould the city of the R epublic. (T rib a l rulers do n o t have th e unchecked pow er gran ted to the R e p u b lic ’s philosopher kings, so P o p p e r’s charges o f tribalism and d ictato rsh ip do not go well together.)



It should perhaps be said here that P op p er’s description of tribal societies does not in all respects correspond to those offered by historians and anthropologists. In particular, custom ary law is generally more flexible than Popper allows and prim itive peoples are generally aware of the difference betw een their laws and the laws of the universe; like the Cretans and Spartans they generally ascribe their customs to a remote ancestor. But these points strengthen rather than weaken the case against Plato; he is not just reverting to traditional pattern but going beyond it. Of course, neither Plato him self nor his imaginary A thenian share what Popper would regard as the prim itive cast of m ind. T hey are themselves only too well aware that laws can be changed, b u t they want the citizens of the new state to be educated so as to find change unthinkable. T his contrast between the A thenian’s (or P lato’s) attitude and that required in the new citizens expresses itself in an ambivalence already noted in the treatm ent of the nocturnal council. The m em bers of the council will study law, b u t it is not clear w hether their qualifications to do so will rest on experience, like genuine tribal elders, or on philosophical training. For his charge of collectivism, Popper relies heavily on two key passages from th e L avis. In the first the A thenian has been describing the ideal b u t unattainable state in w hich there is com plete com m unity of property. H e quotes the dictum ‘friends have all things in com m on’ and goes on to argue th at, so far as possible, nothing should be private, wives, children and property should all be com m on. Ideally we should even see, hear and act in com m on, we should be unanim ous in our praise and blame, and be happy and sad at the same things. In short the laws should create the greatest possible unity (739c-d). This is an explicit attack on any theory that attaches im portance to the individual. So, although Popper is w rong to suppose th at Plato ever considered the com m unity o f family and property an attainable ideal, he is justified in citing it as evidence o f P lato’s collectivism. The second passage is not only used in support o f the charge of collectivism b u t is also placed at the head o f the first chapter as a motto to illustrate Plato’s antipathy to the open society. As cited there it reads T he greatest principle of all is th at nobody w hether male or female, should be w ithout a leader. N o r should the m ind of anybody be habituated to letting him do anything at all on his own initiative; neither out of zeal nor playfully. But in war and in the m idst of peace to his leader he shall direct his eye and follow him faithfully. And even in the sm allest m atter he should stand



u n d e r leadership. F o r exam ple, he should get up, or move, or w ash, or take his m eals . . . only if he has been told to do so. In a w ord, he should teach his soul by long h ab it, never to dream of acting in d ep en d en tly and to becom e u tterly incapable of it. (942c-d) I n context, this passage is a little less horrifying th an it seems by itself. T h e A thenian is describing m ilitary service and the references to g ettin g up, m oving, w ashing and taking meals clearly apply to these activities as perform ed in cam paigns or exercises. T here is also a p ro b lem about P o p p er’s use of the w ord ‘leader’ in his translation. T he G re e k w ord archon certainly can m ean ‘leader’ b u t it is also the sta n d a rd w ord for any kind o f ruler, including dem ocratically ap p o in te d m agistrates. Im m ediately after the passage quoted, the A th en ian goes on to say th a t the whole experience o f the citizen m ust be one of ‘rulin g and being ru le d ’ (or ‘leading and being led ’). T he tro u b le w ith the translation ‘leader’ is th at it could suggest the dem and for unflinching loyalty to a pow erful personality (a F ührer), whereas w h a t th e passage really asks is th a t the citizen should obey those who fo r th e tim e being are d u ly appointed as m agistrates or com m anders. (A nyone who thinks the passage is evidence of m ilitarism should re m e m b e r th a t th e dialogue began w ith an attack on the m ilitaristic system s o f C rete and Sparta; see S ilverthorne, 1973). T h ese qualifications may soften th e im pact of the passage quoted b u t they do not significantly affect P o p p e r’s basic point. T h e A thenian is arguing th a t th e entire experience of citizens from earliest childhood sh o u ld train th em to accept discipline and the unthinking obedience th a t this involves. I t explicitly suggests th at any expression of in d iv id u ality ' is undesirable. So P o p p er is justified in using it as evidence of P lato’s opposition to th e open society.


P la to ’s prin cip les

A lthough P o p p er is entirely correct to claim that the kind of state advocated by Plato would be far rem oved from the ideal of the open society, the general style o f his argum ent could still be m isleading in tw o ways. In th e first place P o p p er’s talk of the open society suggests th a t all societies are open or closed and th a t we have a straight choice b etw een the tw o models. But, o f course, there is in fact no such thing as a com pletely open or com pletely closed society. It may not even be possible to rank societies in order of openness and closedness — society A could be m ore open th an society B in one respect and more



closed in another. In the second place, Popper's m ethod could lead to the obviously questionable assum ption that the m ore open of two societies is always to be preferred. B ut there are other values w hich a reasonable man might seek and these may conflict with the dem ands of openness; for example if we set a high value on order we may prefer the less open but more orderly o f two societies. In short, the unw ary reader o f Popper’s book could get the im pression that we have a sim ple choice between the open and the closed society and th at one of these alternatives (the open m odel) is obviously th e right one to choose. In this perspective P lato’s preference for a relatively closed society may seem perverse, the product perhaps of a pathological personality. Popper explains P lato’s position in term s of a supposed ‘historicist’ methodology which led him to adopt the m axim 'A rrest all change’. But if we are to take Plato seriously as a philosopher we should consider what he actually says. In particular, we w ould expect to explain his proposals in term s of his own explicit account o f the aims of legislation (see above, chapter 4). We saw that the first object o f the legislator is to secure internal and external peace. From very early in the L a w s , the A thenian m aintains that this is to be achieved through law. He believes th at a code o f law must therefore be accepted as a perm anent fixture and th at the citizens m ust be so indoctrinated as to accept this code unthinkingly. T his argum ent rests on some basically sound assum ptions. T h e A thenian is right to assum e that law m ust be relatively stable and th at it m ust command the respect of the citizens. Even in a relatively open society like those of m odern Britain or Am erica, the constitution can survive only so long as those with political power accept that they m ust obey its fundam ental provisions. T hey adopt the habits of m ind o f a closed society b u t only w ith respect to th e central features of the constitution. The trouble is th at the A thenian pushes these points too far. He treats the creation of order as though it were an overriding aim to be pursued at all costs. But even if we accept th at order is a fundam ental requirem ent of any society we need not suppose that order m ust always be maximized at the expense o f any other values whatever. Indeed there comes a point at which the pursuit of order becomes selfdefeating since the constraints it imposes on the citizens will lead to discontent. So the question is not ‘How can we achieve the greatest degree o f order?’ b u t ‘H ow can we produce sufficient order to perm it the pursuit of other values?’ P lato’s assum ption that the chief aim of legislation is to make the citizens virtuous and th at virtue is objectively discoverable would also lead naturally to a preference for a relatively closed society. It becomes



th e function of th e state to ensure th a t the citizens conform to a certain p a tte rn of character and behaviour: to give them freedom is sim ply to allow them to do the w rong thing. B ut it is im portant to see exactly why th is argum ent is erroneous. Political thinkers in the liberal tradition som etim es talk m istakenly as though it is possible for the legal code to av o id a com m itm ent to any values. B ut some values, if only that of lib e rty , m ust be incorporated w ithin the legal code and some possible values m ust be excluded. I t is absurd to suggest that a legal system co u ld be im partial betw een, say, liberal dem ocrats and supporters of A yatollah K hom eini’s bran d o f Islam ic revolution. So P lato ’s mistake, i f he has m ade one, does n o t consist in his belief th a t there are values th a t a legal code m ust pro m o te, b u t ra th e r in the way in w hich he conceives those values. H e im plies th at the good life dem ands co n fo rm ity to established p a tte rn s o f th o u g h t and behaviour. I t is this assum ption th a t m ust be contested by supporters of the open society. P la to ’s conform ist conception of virtue is closely connected with w h a t P o p p er calls his ‘holism ’ — his wish to see hum an beings as parts o f larger wholes. A lthough th is picture of hum anity has dangerous im plications for m orals and politics and can lead to the collectivism characteristic o f th e closed society, it has attractions th at are no t purely aesthetic. M ost hu m an beings feel a need to ‘belong’, to feel th at they have a secure place as m em bers o f a social group. These sentim ents can u n d o u b te d ly encourage th e kinds o f a ttitu d e th at P opper would ascrib e to m em bers o f closed societies. M arx appeals to th em w hen he advocates co-operative rath er th an com petitive modes of economic activity. P opper scorns such feelings as relics of tribalism . It m ight be m ore accurate to suggest th a t they are p art o f our heritage as social anim als. As such they are inescapable. T hey may som etim es lead to m ilitarism an d intolerance b u t they are also the source o f some o f our deep est satisfactions. P lato shows his aw areness of this in his calls for frien d sh ip am ong th e citizens an d in some o f his proposals for the new city, notably th e com m on meals. It seems likely that any satisfactory form of society will need to take account o f them while trying not to suppress individual initiative. T h e choice betw een collectivism or individualism is not so clear c u t as Popper apparently supposes.


The a fterlife o f the L a w s

A lthough th e R epu blic has been m uch m ore widely read, it is approp riately the L a w s th at has had th e greater influence on practical politics. Som e o f its proposals seem to have been adopted by G reek



cities after P lato’s death, but its doctrine o f the sovereignty o f law and accom panying political theory of checks and balances has had a lasting effect. M ediated by w riters such as A ristotle, Polybius and M ontesquieu, it has had a profound influence on the developm ent of modern dem ocracies. O ne u n fortunate consequence o f P o p p er’s concentration on the R epublic is that he misses Plato’s advocacy of this theory altogether. This is the m ore surprising because P o pper him self advocates a theory o f checks and balances against w hat he takes to be Plato’s theory of (unchecked) sovereignty. T he im p o rta n t question, Popper m aintains, is not ‘W ho should ru le?’ but ‘Howr can we so organize political institutions that bad and incom petent rulers can be prevented from doing too m uch dam age?’ (121). In the L a w s Plato adopts a m ere optim istic view o f the capacities of h u m an rulers — he believes that political institutions can be arranged so th a t the better people hold office — b u t he knows that even these will not be infallible. So he w ould agree w holeheartedly w ith Popper that th e chief task is to design political institutions th at take account of hum an frailty. T hus Popper’s focus on the R epublic not only prevents him from acknowledging the chief debt that liberal political thinkers owe to Plato, b u t also blinds him to these elem ents in P lato’s thought which, on his own account, would be most positive and valuable. It is perhaps understandable that Popper, living u n d er the shadow of H itler, should have been obsessed by th e threat o f dictatorship and should have interpreted Plato accordingly. B ut civilized ways of life can also be threatened by a loss of confidence in the state and political institutions generally. This danger is perhaps more ap p aren t at a time when our attention is often focussed on political groups who denounce law and the state as irredeem ably co rru p t w ithout offering any clear account of what could replace them . In the L a w s, Plato, emphatically expresses his own belief in the value o f political society. So the message is not, as has so often been supposed, one of despair. He is telling us that a good life is possible, b u t only if we take the trouble to develop and defend sound political institutions.


The Constitution O FFIC E-H O LDER S

T h e A ssem b ly

M em bership:

A ttendance:

F unctions:

open to all who b ear or have borne arm s for the state, i.e. to all ad u lt male citizens (753b); it is not clear w h eth er w om en w ould be m em bers (805a-c may im ply th at they would), com pulsory for the top two property classes; optional for th e o th er tw o except on special occasions (764a). th e m ain function is to elect m em bers of the council and o th er officials; a few other functions are m entioned (767e-768a, 772c-d, 850b, 9 2 le, 943c); b u t the role o f the Assem bly is in general very unclear.

T h e C ouncil (756b- 758d)

M em b ersh ip : M o d e of election:

m e m b e r s fr o m each p ro p erty class; m em bership lasts one year. very com plex; the whole process lasts five days: D a y s 1 a n d 2: candidates from the first and second classes nom inated; all citizens required to take part. D a y 3: candidates from the th ird class nom inated; p articipation is optional for m em bers of the lowest class b u t com pulsory for others. D a y 4: candidates from the lowest class nom inated; participation is optional for m em bers o f th e th ird and fo u rth classes b u t com pulsory for m em bers of th e tw o highest classes.

n in e ty




D a y 5: an election is held to choose 180 o f those nom inated from each class; this n u m b er is then reduced by lot to ninety from each class, conducting the day to day business of the state, sum m oning th e assembly and guarding the city; for these purposes the Council is divided into twelve sections each of which functions for one m onth; its other responsibilities are left vague (758a-d).

G uardians o f the lai 'js (7 5 2 d -7 5 5 b )

M embership: Mode of election:

th irty-seven citizens aged at least fifty; they hold office for tw enty years or until they are seventy. (1) for elections occurring after the new city has been properly established, each citizen is required to w rite dow n th e nam e of the person he thinks most suitable; the names of those thus nom inated are exhibited for th irty days during which objections may be made; the 300 who have received m ost nom inations are then selected; a further selection (apparently by election) reduces these to 100; a final election produces the required n um ber of guardians. (2) special arrangem ents are to be m ade for the transitional period when the new city is being established; the m ost im p o rtant suggestions are (a) th at, to begin w ith, the guardians should include eighteen from Cnossos (the most im p o rtan t city o f Crete) and nineteen of the new colonists; (b) that the first elections should be supervised by a board of 100 colonists and 100 Cnossians; m ost scholars (e.g. M orrow , 1960, pp. 204-6 ,2 3 8 -4 0 ) have supposed th at these sug­ gestions are alternatives linked together by an editor, b u t Saunders (1970b) has suggested that we could make sense o f the text as it stands if (a) is the procedure used at the first foundation o f the new city, (b) is used when it has been established for a short tim e, w ith the procedure described in (1) above being used for subsequent ‘norm al’ elections.



F u n ctio n s:

in ten d ed to be the m ost influential organ of the state; at first th ree functions are m entioned — g uarding th e laws, keeping the p roperty registers, and hearing th e cases of those charged w ith not registering property; m any other specific func­ tions are also m entioned.

M i li ta r y officers (7 5 5 b -7 5 6 b )

M o d e o f election:

candidates for th e post of general are nom inated by th e guardians and three of these are th en elected by th e soldiers; the procedure is the same for cavalry com m anders except th at the election is th e n m ade by the cavalry men; com pany com m anders are nom inated by the generals, b u t if these nom inations are disputed there is an election.

R eligiou s officials (7 5 8 d -7 6 0 a )

P riests:

chosen by lot b u t those chosen will be subject to exam ination to ensure that they are of suitable character. E xpounders: candidates are elected and the D elphic O racle is th en asked to choose three o f these. T em p le treasurers: elected by th e highest property class.

C o u n try w arden s (7 6 0 a -7 6 3 c )

M ode of election:

F unctions:

five m em bers o f each of the twelve tribes are elected; each o f these (or possibly each group of five) th e n chooses twelve young assistants; they hold office for two years. each group o f country w ardens supervises a d ifferen t area o f the country each m onth; they are responsible for fortifications, public works and police d u ties; they act as judges in less im portant m atters.



C ity w ardens (763c-e)

Mode o f election:


six m em bers of the highest p ro p e rty class are elected; this nu m b er is then re d u ce d by lot to three; they hold office for one year. supervision o f public works a n d police duties within the city.

M ark et w arden s ( 7 6 3 c-7 6 4 c)

Mode o f election:


ten m em bers of th e two highest pro p erty classes are elected; this num b er is reduced by lot to five; they hold office for one year. supervision of the markets.

The m inister o f education (7 6 4 c -7 6 6 b )


Mode of election: Functions:

the m inister m ust be at least fifty years old, the father of legitim ate children and a guardian of the laws. elected by all the officials except m em bers of the Council; term of office five years. overall supervision of education; said to be the most im portant office o f state (766e).

Scrutineers (945b -9 4 8 b )

M ode of election:


each citizen m ust propose the nam e of one man over fifty years o f age; o f those th u s nom inated, half (i.e. those who have received most nom inations) go forw ard; the process is repeated until the num ber o f candidates left corresponds to the num ber of places to be filled; twelve candidates are to be chosen in the first year and three in each subsequent year; they hold office until they are seventy-five. to scrutinize the conduct in office of all officials and to impose penalties w hen appropriate.



T h e nocturnal council (9 5 lc - 9 5 2 d , 9 6 la - 9 6 8 e )

M em bership:

F unctions:

there are slight discrepancies betw een the two accounts of th e m em b ership o f this council (95 le, 961 a); it looks as though it would consist of (a) the te n oldest guardians o f the laws together w ith the present m inister of education and those o f his predecessors who are still alive, (b) all who have w on aw ards o f hono u r, and (c) those citizens who have been sent abroad to discover w hether anything can be learned from other states; each m em ber is also req u ired to bring at least one suitable young m an aged at least thirty. to study th e natu re of virtue (the true end o f the state) and to investigate w hat laws are best calculated to achieve it; m ore specific functions include re-educating atheists and hearing the reports o f those sent abroad to study foreign custom s.

TH E A D M IN ISTR A TIO N OF JUSTICE (766d-768c, 855c-856a, 875e-876e, 956b-958c)

P lato follows the A thenian practice o f dividing cases into two categories, public and private. T h e private category includes not only w h at we w ould call civil cases b u t also all those cases where an injury has been done to an individual in his private capacity. P rosecution in su ch cases is norm ally th e responsibility o f th e injured party. Public cases are those w here h arm has been done to the city as a whole. P ro secu tio n s in these cases m ay be u n d e rta k en either by m em bers of th e public or by th e appropriate board of officials. Procedures are as follows: P r i v a t e cases

(1) In the first case disputes are to be b ro u g h t to arbitrators whom th e parties choose from am ong th eir neighbours. (2) T h ere is an appeal from the decision of arbitrators to a tribal court — a cou rt consisting o f jurors chosen by lot, probably from the whole citizen body b u t organized in tribes. (3) T h ere is appeal from the trib al courts to a suprem e court



consisting o f one m em ber nom inated from am ong th e ir ow n num ber by each board of officials whose period o f office lasts for m ore th an one year. Public cases

The opening and closing stages are before a court of the w hole people, but there is an intervening stage in w hich a detailed exam ination is carrieci out by three judges chosen from th e highest officials by the prosecutor and defendant, or by the council if they cannot agree on a choice. Capital cases come before a special court, apparently consisting of the m em bers o f th e court of select judges (see u n d er ‘Private cases’ 3) sitting together w ith the guardians o f th e laws. M orrow (1960, pp. 264-70) points out (a) th at the precise demarcation betw een the court of the people and th e court o f capital cases is left unclear, and (b) th a t in sum m arizing the judicial arrangem ents of th e new city, th e A thenian suggests th at it should be left to the guardians of the laws to decide an ap p ropriate procedure for trying public cases (957a-b). It may therefore be th at the arrangem ents just described do not em body P lato’s final thoughts on this topic. M a g isteria l ju stice

Almost all the boards of officials have power to act as courts in cases which come w ithin the sphere which they norm ally supervise. This com petence is normally restricted to cases where the subject of the dispute is below a certain value, but this limit is set quite high com pared w ith contem porary A thenian practice. So the officials have extensive judicial powers. T h e precise relation of these to the courts described above is in many respects unclear.

Further Reading G en era l S u m m a rie s o f the d ialogue w ith co m m en ts are to be foun d in G rote (1888), S h o re y (1 933), T aylor (1960a and b ), F riedlander (1969), Strauss (1975), and G u th rie (1978). I have foun d G rote and G uthrie m ost helpful. Sau nd ers (1970a) has a useful in troduction. M orrow (1960) has very full and detailed d iscu ssion s o f the institution al prop osals and relates these to w hat is k n ow n o f contem porary G reek practice. E nglan d (1921) is a fin e w ork o f classical scholarship bu t has little to offer readers w hose interests are prim arily ph ilosophical. M u lle r (1951) argues for a view o f th e L a w s as a chaotic produ ct o f P lato’s d e c lin e . T h e reply b y G orgem ann s (1960) is particularly helpful. For a su m m a ry o f the key issu es see the review s (in E n glish ) o f these tw o w orks by C h ern iss (1953) and O stw ald (1962). S au n d ers (1975b) is a very readable accou n t o f the problem s o f translating th e L aw s. Saunders (1975a) is a fu ll bib liograp hy o f the Law s.



T h e nature o f law M o rro w (1960), chs. X I and X I I , d iscu sses the relation betw een ph ilosophical and legal issues in the Law s. O n P lato as a foun der o f the natural-law tradition see M aguire (1 947), H all (1 9 5 6 ) and O stw ald (1977). For the d evelop m en t o f the natural-law tradition see A ristotle, N icom achean E th ic s , V. vii; X . ix; R hetoric, I, xiii, i3 7 3 b 1-8; cf. P hysics I I. viii; Cicero, R ep u b lic, II I. xxii. 33; L a w s, II. iv. 8 to v. 13; A quinas, S u m m a Theologiae, la 2ae. qq. 9 0 -7 . M od ern d iscu ssion s o f natural-law theory are su m m arized in Harris (1980), ch s. 2 and 3. H art (1 961), ch. X , criticizes natural-law theories. F u ller (1969) d e fe n d s a view o f law as purposive. F in n is (1980) gives a detailed defen ce o f natural law , taking som e account o f P lato’s contribution.



The aims o f legislation Aristotle’s view is close to th a to fP la to in the L a w s; see X ico m a ch ea ti E thics. II. i, 1103b 1-7; V. i; X . ix; Politics, V II. ii; V II. xiv, 1333a 3 0 -1 3 3 4 a 10. O n the problem o f legal m oralism see M ill, O n L ib e r ty , B erlin (1969) Hart (1962) and D e v lin (1965). T h e issues are sum m arized in H arris (1 980), ch. 10. On P lato’s doctrine o f persuasion and its relevance to the pream bles o f the laws see M orrow (1953). Plato as a legislator M orrow (1941) and Cairns (1949) assess P lato’s m erits as a legislator. Vol. xi o f des Places and D ie s (1 9 5 1 -6 ) con tain s a len g th y acco u n t by L. G ernet o f the relationship betw een Plato’s code and G reek law in P lato’s day. M acdow ell (1978) is a convenient source for the law o f A th en s. 3

M o ra l philosophy

Virtue M ost m odern accounts o f P lato’s moral philosophy neglect the L aw s. G ould (1955) and C rom bie (1 9 6 2 -3 ), vol. 1, ch. 6, are e x ce p tio n s to this rule. For the idea o f sophrosune (self-control) see N orth (1966). D over (1974) is a mine o f inform ation on the attitu d es o f P lato’s contem poraries to moral matters. G igon (1954) argues for the incoherence o f the o p en in g pages o f the Law s. For a reply see G orgetnanns (1960), pp. 113-54. A ristotle’s treatm ent o f moral virtue in N icom achean E th ic s, II, is in many respects very close to that o f Plato in the Laws. M oral psychology D od ds (1951), R ees (1957) and Vlastos (1957) discuss the basic principles of P lato’s moral p sych ology as it appears in the later dialogues. Saunders (1962) argues that the L aw s, like the R epublic, m aintains a parallelism b etw een the parts o f the soul and the parts o f the state. G orgem anns (1 960), pp. 155-61, considers the roles o f reason and passion in the Law s. T he possibility o f akrasia M ost m odern com m entators have concentrated on the treatm ent o f this problem in the P rotagoras, together with the M en o and Gorgias. See Santas (1964) and G ulley (1965). W alsh (1963), O’Brien (1967) and H ackforth (1946) pay more attention to later dialogues including the Lazvs. M ortim ore (1971) includes essays on the problem as it appears in both ancient and m odern philosophy. A ristotle’s celebrated account is in N icom achean E th ics, V II. i-ix .



P leasu re and happiness M o st m odern d iscu ssio n s o f these topics in P lato have concentrated on the P ro ta g o ra s, R epublic and P hilebus. G oslin g and T aylor (1982) have a rather b r ie f chapter on the Laws', see also G orgem ann s (1960), pp. 165-92. A ristotle deals w ith these topics i n N ico m a ch e a n E th ics, V II. x i-x iv , X . i-v . F o r B en th am ’s felicific calculus see his P rinciples o f M o ra ls and Legislation, ch . iv.


P olitics and society

G en era l accounts B arker (1960) is still the m ost h elp fu l accou nt o f P lato’s political p h ilosop h y as a w h o le and deals very fu lly w ith the L aw s. T h ere are briefer treatm ents in C ro m b ie (1 9 6 2 -3 ), vol. 1, ch. 4, and in H all (1981). M orrow (1962) includes an a c c o u n t o f P lato’s in v o lv em en t in Syracuse and o f the political im p lication s of th e L etters. A lm o st every page o f A ristotle’s P olitics has som e bearing on the L a w s. T h e n o te s o f Barker (1946) and esp ecially, Sau nd ers (1981a) are valuable in this resp e ct. O f particular in terest are B ook I I , chap. vi, w here A ristotle offers so m e lop -sid ed and u n gen erous criticism s o f the L aw s, and Books V II and V I I I , w h ich describe an ideal state not unlike that o f the Law s.

T h e op en society P o p p er (1966) is essential reading, th ou gh his treatm ent o f the L a w s is fragm en tary and o n e -sid e d . L ev in so n (1953) is a m assive reply to P opper, w h ile B am brough (1967) in clu d es con trib u tion s from b oth sides o f the argu m en t. Id ea l states and utopias A r isto tle, P olitics, IV . i, d iscu sses the relation b etw een ideal c on stitu tion s and w h at is practicable. H e d iscu sses variou s ideal states, in clu d in g those o f the R ep u b lic and the L aw s, in II. i-v iii. H is o w n ideal is describ ed in Books V II and V III. F or m odern d iscu ssion s o f u top ian ism see P opper (1966), ch. 9, and H o rsb u rg h (1 9 5 6 -7 ).

T h e m ixed con stitu tion T h e idea o f the m ixed co n stitu tio n as it appears in the L a w s is d iscu ssed in M o rro w (1 9 6 0 ), ch. X . A ristotle g iv es his version in P olitics, IV . ix -x i. T h e id ea has in flu en ced m odern th o u g h t c h iefly through the doctrin e o f the separation o f the pow ers; see M on tesq u ieu , L ’E sp rit des Lois, X I. 6.



The rule of law M orrow (1960), ch. X I, d iscusses P lato’s view . For A risto tle’s version see Politics, II I, xi, 1282a 4 1 - b l 1; II I. xvi; IV . iv -v i. For a m odern treatm ent o f the idea o f the rule o f law see L ucas (1966), pp. 106-43. See also the discussions o f the ‘inner m orality’ o f the law in Fuller (1969) and o f the different kinds o f rule that com prise a legal system in Hart (1961), chs IV and V. T here is a sum m ary o f m odern d iscu ssio n s in Harris (1980), ch. X I.

T he system of governm ent Barker (I9 6 0 ), ch. X V , and M orrow (1960), chs. V and V I d iscu ss the political and judicial institution s o f the L aw s. M orrow ’s d iscu ssion is very fu ll and detailed. O n dem ocracy and equality see A ristotle’s remarks m P olitics, II. vi; III. ix; IV . iv; V. i; V I. ii. For a m odern introduction to these problem s see Benn and Peters (1 9 5 9 ),ch s. 5 and 15. L ively (1975) and S ch u m p eter (1950) d istin gu ish different forms of dem ocratic theory. Harvey (1965) places P lato’s account o f equality in its historical context.

Social organization M orrow (1960), ch. IV , and Barker (1960), ch. X IV , contain general accounts o f the forms o f social organization proposed in the Lazes. O n slavery see M orrow (1939) and V lastos (1941'. A ristotle’s d iscu ssion in Politics, I. iii-v ii, makes valuable background reading. On class structure see Saunders (1961 and 1962) and W ood and W ood (1978), pp. 183-202; cf. A ristotle, Politics, IV , iii-iv ; IV . xi; V II, v iii-x . On social solidarity see D u rk heim (1972), chas. 5 and 6, and A ristotle, Politics, III. iv.


E ducation and the A rts

For general accounts o f these m atters as they appear in the L a w s, M orrow (1960), ch. V II, and Barker (1960), ch. vii. See also G orgem ann s (1960), ch. 1. Bury (1937) is a w ide-ranging discussion o f education in the L aw s. Sargeaunt (1 9 2 2 -3 ) concentrates on the role o f song and dance and on the idea o f man as G o d ’s playfellow. For the theory o f m usic and the other arts see T ate (1928 and 1932), V erdenius (1949), Schipper (1963) and Crom bie (1 9 6 2 -3 ), vol. 1, pp. 190-3. T he educational system o f A ristotle’s ideal city in m any respects resem bles that of the Law s; see P olitics, V I I, x iii-x v i, and V III.

196 6


P u n ish m e n t a n d R esponsibility

P la to ’s theory o f p u n ish m en t is d iscu ssed in Adkins (I9 6 0 ), pp. 2 9 9 -3 1 2 , S c h u ch m a n (1963), H ub y (1 972), ch. 7, Saunders (1973a, 1976, and 1981b), a n d M ack en zie (1981). T h e se all have som e bearing on P la to ’s con cep tion o f resp o n sib ility . W o o z le y (1972) and Saunders (1973b) d iscu ss the theory o f responsib ility as it appears in P lato’s treatm ent o f killings in anger. F o r m odern d iscu ssion s o f the problem o f p u n ish m en t see H on d erich (1 9 7 6 ) and A cton (1969). W ootton (1963) has a th eory o f p u n ish m en t and r e sp o n sib ility w h ich in som e w ays resem b les P lato’s. She is criticized by H art (1 9 6 8 ), esp . ch. V I I I , and F lew (1973). P la to ’s account o f the causes o f injustice is discussed in O ’Brien (1957) and Sa u n d ers (1968); see also the item s on akrasia cited in 3 above. A r isto tle ’s scattered remarks on p u n ish m en t som etim es im p ly agreem ent w ith P lato and som etim es su ggest other approaches; see N icom achean E th ic s, I I , iii, 1104b 14-18; III. v, 1113b 21 to 1114a 3; V. i v, 1132b 34; V. xi, 1138a 4 14; X . ix, 1179b 31 to 1180a 13; cf. Sorabji (1980), ch. 18. A r isto tle ’s d iscu ssio n o f r esp on sib ility in N icom achean E th ics, V. viii, has a c lo s e bearing on the Law s; see also N icom achean E th ics, I I I , i-v . F o r contem p orary G reek p ractices and attitudes see D over (1974), pp. 1446 0 and M acd ow ell (1963), chs. 1 and 12.


R eligion and Cosmology

F o r general accounts o f religion in the L a w s see Barker (1960), pp. 4 2 2 - and M o rro w (1960), ch. V III. K au fm an n (1 958), s. 4 4 , and C raig (1 980), ch. 1, offer critiques o f the argu m en t for the existen ce o f god(s). T h e fo llo w in g treat the th eology and c o sm o lo g y o f the L a w s in con n ection w ith P lato’s doctrin e as it appears in other dialogues: H ackforth (1936), S o lm se n (1942), esp. chs. V I I I - X , D o d d s (1951), ch. V I I, Skem p (1967), D e m o s (1968). O n the explanation o f evil see V lastos (1939), C herniss (1954) and M ohr (1 9 7 8 ). Saunders (1973a) con n ects the account o f the sou l’s progress after d e a th w ith the p enology o f the L aw s. V lastos (1975) gives an account o f P lato’s c osm ology, criticizing the relig io u s assem p tion s that underlie it. A r isto tle ’s argum ent to the first m over is in som e respects parallel to the a r g u m e n t o f L a w s, X , th ou gh it also ex h ib its im portant differences: see M e ta p h y sic s , X I I. v i-ix ; P hysics, V I I - V I I I .



M etaphysics and epistemology

O n the philosophical m ethod advocated in the later d ialogu es see Crom bie (1 9 6 2 -3 ), vol. II, ch. 3, iv and v, the articles by A ckrill and L lo y d in A llen (1965), and Sayre (1969). O n the Form s in the later dialogues see Crom bie (1 9 6 2 -3 ), vol. I I , pp. 257— 61, O w en (1953), C herniss (1957). C herniss (1953), G orgem an n s (1 960), pp. 21 8 -2 6 , and G uthrie (1978), pp. 3 7 8 -8 1 , d iscuss the ev id en ce for the Form s in the Laws.

Select Bibliography

T h is in c lu d e s all the works referred to in the text or m en tion ed in the s u g g e s tio n s for F urther R eading. For a full bib liograp hy o f the Law s see S a u n d er s (1975a). A c to n , H .B . (ed .) (1969), T he P hilosophy o f P unishm en t, L o n d o n : M acm illan. A d k in s, A .W .H . (1960), M e r it a n d R esponsibility, Oxford: C larendon Press. A lle n , R .E . (ed.) (1965), S tu d ie s in P la to ’s M etaphysics, L ondon: R outledge. B am brough,


(ed .)

(1 967), P lato,

P opper a n d P olitics,

Cam bridge:

C am b ridge U n iversity Press. B ark er, E. (1 946), T he P olitics o f A risto tle , translated w ith in troduction, notes and a p p en d ices, L ondon: O xford U n iversity Press. (1 9 6 0 ), G reek P olitical T heory, 5th edn, L ondon: M eth uen. B e n n , S .I. and P eters, R .S. (1 9 5 9 ), Social P rinciples a n d the D em ocratic S ta te , L on d on : A llen & U n w in . B e r lin , I. (1969), ‘John Stuart M ill and the en d s o f life ’, in F o u r essays on L ib e rty , L ondon: O xford U n iversity Press. B u r y , R .G . (1 9 2 6 ), P la to w ith an E nglish T ra n sla tio n , vols. IX and X , L aw s, L o e b C lassical L ibrary, C am bridge, M ass.: Harvard U n iversity P ress, L on d on : H einem ann . (1 9 3 7 ), ‘T h e theory o f education in P lato’s L aw s’, R evue des E tu d es Grecques, 50, pp. 3 0 4 -2 0 . C a ir n s, H . (1 9 4 9 ), ‘P la to ’s theory o f L aw ’, in L egal P hilosophy fro m P lato to H egel, Baltim ore: Johns H opkins U n iversity P ress, pp. 2 9 -7 6 ; also in F ried länder (1969). C h e r n iss , H .F . (1 953), R eview o f M üller (1 951), G nom on, 25, pp. 36 7 -8 9 ; reprinted in C herniss (1977). (1 9 5 4 ), ‘T h e sources o f evil according to P lato’, Proceedings o f the A m erican P hilological S o c ie ty, 98, pp. 2 3 -3 0 ; reprinted in V lastos (1972) and in C h ern iss (1977). (1 9 5 7 ), ‘T h e relation o f the T im aeu s to P la to ’s later d ia lo g u e s’, A m erica n J o u r n a l o f P hilology, 78, pp. 2 2 5 -6 6 ; also in A llen (1965) and C herniss (1 977).



(1977), Selected Papers, Leiden: Brill. Craig, W .L. (1980), T he Cosmological A rg u m en t fro m P lato to L e ib n iz, London: M acmillan. Crom bie, I.M . (1 9 6 2 -3 ), A n E xa m in a tio n o f P la to ’s D octrines, 2 vols., London : Routledge. D em os, R. (1968), ‘Plato’s doctrine o f the psyche as se ll-m o v in g m otion ’, Jo u rn a l o f the H isto ry o f Philosophy, 6, pp. 133-45. d esP laces, É .and D ies, A. (1951 - 6 ) , P laton, Oeuvres C om pletes, vols. X I - X I I , Budé edn, Paris: Société d ’Edition ‘Les Belles L ettres’. D evlin, P. (1965), The E nforcem ent o f M o ra lity , Oxford: C larendon Press. D odds, E.R. (1951), The Greeks and the Irrational, Berkeley and Los A ngeles: U niversity o f California Press. D over, K.J. (1974), G reek P opular M o ra lity in the tim e oj P la to a n d A risto tle , Oxford: Blackwell. Durkheim , E. (1972), Selected W ritings, ed. A. G id d en s, Cambridge: Cam bridge U n iversity Press. England, E.B. (1921), The L aw s o f P lato, 2 vols., M anchester: M anchester U niversity Press. Engels, F. (1972), ‘Socialism , utopian and scien tific’, in R .C . T ucker (ed.), The M a r x Engels R eader, N ew York: N orton , pp. 6 0 5 -3 9 . FinniSj J. (1980), N a tu ra l L a w and N a tu ra l R ights, Oxford: C larendon Press. Flew, A .G .N . (1973), C rim e or Disease? London: M acm illan. Friedlander, P. (1969), P lato, vol. I l l , trans, H . M eyerh off, Princeton: Princeton U niversity Press. Fuller, L. (1969), T k e M o ra lity o f L a w , revised edn, N ew H a v en a n d L ondon: Yale U n iversity Press. G igon. O. (1954), ‘Das E inleitungsgespràch der G esetze P laton s’, M useum H elveticum , 1 1 ,2 0 1 -3 . G ôrgem anns, H. (1960), B eitrage zu r Interpretation von P latons N om oi, Z etem ata series, no 25, M unich: Beck. G osling, J. and T aylor, C .C .W . (1982), The Greeks on Pleasure, Oxford: Clarendon Press. G ould, J. (1955), The D evelopm ent o f P la to ’s E thics, Cam bridge: Cam bridge U niversity Press. G rote, G. (1888), P lato and the O ther Companions o f Socrates, new edn, vol. IV, London: John Murray. G ulley, N . (1965), ‘T h e interpretation o f “ N o one does w rong w illin gly” in PLato’s dialogues’, Phronesis, X , pp. 8 2 -6 . G uthrie, W .K .C . (1978), H istory o f G reek P hilosophy, vol. V, Cambridge: Cambridge U niversity Press. H ackforth, R. (1936), ‘Plato’s th eism ’, Classical Q u a rte rly , 30, pp. 4 -9 , reprinted in Allen (1965).



(1 9 4 6 ), ‘M oral evil and ignorance in P lato’s e th ic s’, Classical Q u a rterly, 40, pp. 118-20. H a ll, J. (1 956), ‘Plato’s legal p h ilo so p h y ’, In d ia n a L a w J o u rn a l, 31, pp. 1 7 1 -2 0 6 . H a ll, R. (1981), P la to , London: A llen & U n w in . H arris, J.W . (1980), L egal Philosophies, L ond on: Butterw orth. H art, H .L .A . (1961), T he Concept o f L a w , O xford: C larendon Press. (1 9 6 8 ), P unishm ent and R esponsibility, O xford: Clarendon Press. H arvey, F .D . (1965), ‘T w o kinds o f eq u ality’, Classica el M ed ia eva lia , X X V I, pp. 101-46. H o n d er ic h , T . (1 976), P unishm ent: the supposed ju stifica tio n s, revised edn, H arm ondsw orth: P enguin. H o rsb u r g h , H .J .N . (1 9 5 6 -7 ), ‘T h e relevance o f the u top ian ’, E thics, L X V II, pp. 127-38. H u b y , P. (1972), P lato and M odern M o ra lity , L ondon: M acm illan. J o w e tt, B. (1 9 5 3 ), T h e D ialogues o f P la to , 4 th e d n , vol. IV , Oxford: Clarendon Press. K au fm an , W. (1 958), C ritique o f R eligion a n d P hilosophy, London: Faber. L e v in so n , R .B . (1953), In D efense o f P la to , C am bridge, M ass.: Harvard U n iv ersity Press. L iv e ly , J. (1975), D em ocracy, Oxford: Blackwell. L u c a s, J.R . (196 6), T he Principles o f P olitics, O xford: C larendon Press. M a c D o w e ll, D .M . (1 963), T h e A th e n ia n H om icide L a w in the A ge o f the O rators, M anchester: M anchester U n iv ersity Press. (1 9 7 8 ), The L a w in C lassical A th en s, L ondon: T ham es & H udson. M a ck en zie, M .A . (1 981), P la to on P unishm ent, B erkeley, L os A ngeles and L ondon: U n iversity o f California Press. M a g u ire, J .P. (1.947), ‘P lato’s theory o f natural law ’, Yale C lassical S tu d ies, 10, pp. 151-78. M o h r , R .D . (1 978), ‘P lato’s final th ou gh ts on e v il’, M in d , L X X X V I I, pp. 5 2 2 -5 . M o rro w , G .R . (1939), ‘Plato and Greek slavery’, M in d , 48, pp. 186-201. (1 9 4 1 ), ‘Plato and the rule o f law ’, P hilosophical R ev ie w , 50, pp. 105-26; reprinted in V lastos (1972). (1 9 5 3 ), ‘P lato’s con cep tion o f p ersu asion ’, P hilosophical R eview , 62, pp. 2 3 4 -5 0 . (1 9 6 0 ), P la to ’s C retan C ity , Princeton: P rinceton U n iversity Press. (1 9 6 2 ), P la to ’s E pistles, Indianapolis: B obb s M errill. M o rtim o re, G .W . (ed.) (1971), W eakness o f W ill, London: M acm illan. M ü lle r , G . (1 951), S tu d ie n zu den platonischen N o m o i, Zetem ata series, no. 3, M unich: Beck, revised ed n , 1968. N o r th , H . (1966), Sophrosune, Ithaca: C ornell U n iversity Press.



O ’Brien, M .J. (1957), ‘Plato and the '‘good conscien ce". Lazes 8 6 3 e 5 - 8 6 4 b 7 \ Transactions o f the A m erican Philological Association, 88, pp. 8 1 -7 . (1967), The Socratic P aradoxes a nd the G reek M in d , C hapel H ill: U n iv ersity o f N orth Carolina Press. Ostwald, M . (1962), R eview o f G orgem anns (1960), Gnomon, 34, pp. 2 3 1 -4 1 . (1969), N om os and the Beginnings o f A th e n ia n D em ocracy. O xford: C larendon Press. (1977), ‘Plato on law and nature’, in N orth (ed.), In terp reta tio n s o f P lato, M em nosyne Supplem ent, 50. Owen, G .E .L . (1953), ‘T he place o f the Tim aeus in Plato's d ialogu es’, Classical Q uarterly, 3, pp. 79-95; reprinted in A llen (1965). Pangle, T. (1980), The Law s o f P lato, translated w ith notes and interpretive essay, N ew York: Basic Books. Popper, K .R . (1966), T he O pen Society and its Enem ies, vol. 1, Plato, 5th ed n , London: R outledge. (1976), U nended Q uest, London: Fontana. Q uinton, A. (ed .) (1967), P olitical P hilosophy, London: O xford U n iversity Press. Rees, D .A . (1957), ‘Bipartition o f the soul in the early A cad em y’, J o u rn a l o f Hellenic Studies, 77, pp. 112-18. Ritter, C. (1896), P latons G esetze, K om m entar zum G n e ch isc h e n T ex te, M unich: Beck. Santas, G. (1 9 6 4 ),‘The Socratic paradoxes’, P hilosophical R eview , 73, 1 47-64. Sargeaunt, G .M . (1 9 2 2 -3 ), ‘T w o stu dies in P lato’s Law s' ,H ib b e rt J o u rn a l, 21, pp. 4 9 3 -5 0 2 , 669-79. Saunders, T .J. (1961), ‘T h e property classes and the value o f the kleros in P lato’s L a w s', Eranos, 59, pp. 2 9 -3 9 . (1962), ‘T h e structure o f the soul and the state in P lato’s L a w s’, E ranos, 60, pp. 37-55. (1968), ‘T h e Socratic paradoxes in P lato’s L a w s ’, H erm es, 96, pp. 4 2 1 -3 4 . (1970a), P lato, the Law s, H arm ondsw orth: Penguin. (1970b), ‘T h e alleged double version in the sixth book o f the L a w s’, Classical Q uarterly, 20, pp. 2 3 0 -6 . (1972), ‘N o te s on the L aw s o f P lato’, B u lletin o f the In stitu te o f Classical Studies, supplem entary papers, 28. (1973a), ‘Penology and Eschatology in P lato’s Tim aeus and L a w s ’, Classical Q uarterly, 23, pp. 2 3 2 -4 4 . (1973b), ‘Plato on killing in anger: a reply to P rofessor W o o z le y ’, Philosophical Q uarterly, 23, pp. 350-6. (1975a), B ibliography on P la to ’s Laws, 1920-1970, w ith a d ditional citations through 1975, N e w York: Arno. N e w edn, 1979. (1975b), ‘The penguinification o f P lato’, Greece and R o m e, 22, pp. 19-28.



(L 976), ‘P lato’s clockw ork orange’, D urham U n iversity J o u rn a l, L X V I I I , pp. 113-17. (1981a), revised ed n o f A ristotle’s Politics, trans. Sinclair, H arm ondsw orth: P enguin. (1 9 8 1 b ), ‘Protagoras and P lato on p u n ish m en t’, in G .B . K erferd (ed .), The S o p h ists a n d their Legacy, H erm es, ein zelsch riften , 44, 1 29-41. Sayre, K. (1969), P la to ’s A n a ly tic M e th o d , Chicago: C hicago U n iv ersity Press. S ch ip p er, E .W . (1 963), ‘M im esis in the arts in P lato’s L a w s ’, J o u rn a l o f A esthetics a n d A r t C riticism , 32, pp. 199-202. S ch u ch m a n , P. (1 963), ‘C om m en ts on the crim inal code o f P lato’s L a w s ’, J o u rn a l o f the H isto ry o f Ideas, 24, pp. 2 5 -4 0 . S ch u m p eter, J. (1 950), C apitalism , Socialism a n d D em ocracy, 3rd edn, London: A llen & U n w in; a portion o f this is reprinted as ‘T w o con cep ts o f dem ocracy’ in Q uinton (1967). S h orey, P. (1928), review o f Bury (1926), Classical P hilology, 23, pp. 4 0 3 -5 . (1933), W h a t P la to S a id , Chicago: U n iversity o f C hicago Press. S ilverth orn e, M .J. (1973), ‘M ilitarism in the L a w s ? ’ S ym bolae Osloenses, 49, pp. 2 9 -3 8 . S k em p , J.B. (1 967), T he T heory o f M o tio n i n P la to ’s L a te r D ialogues, enlarged edn, Am sterdam : Hakkert. S o lm se n , F. (1942), P la to ’s Theology, Ithaca: C ornell U n iversity Press. S orab ji, R. (1 980), N ecessity, Cause a n d B la m e: perspectives on A r is to tle ’s theory, L ondon: D u ck w orth. S trau ss, L. (1 9 7 5 ), T h e A rg u m e n t and the A c tio n o f P la to ’s L a w s, Chicago: U n iversity o f C hicago Press. T a te , J. (1928), ‘Im itation in P lato’s R ep u b lic’, Classical Q u a rte rly , 22, pp. 16-23. (1932), ‘Plato and im itation’, Classical Q u a rte rly , 26, pp. 1 6 1 -9 . T a y lo r, A .E . (1960a), P lato, the L a w s, E verym an ed n , L ondon: D e n t. (1960b ), P la to the M a n and his W o rk , 7th ed n , L ondon: M eth uen. V e rd en iu s, W .J. (1 9 4 9 ), M im esis, P la to ’s doctrine o f artistic im ita tio n a n d its m eaning to us, L eiden: Brill. V la sto s, G. (1939), ‘T h e disorderly m otion in the T im a eu s’, Classical Q u a rterly, 33, pp. 71 -8 3 ; also in A llen (1965). (1 941), ‘Slavery in P la to ’s th o u g h t’, P hilosophical R eview , 50, pp. 28 9 -3 0 4 ; reprinted in V lastos (1981). (1957), ‘Socratic k n ow led ge and P latonic p essim ism ’, P hilosophical R eview , 66, pp. 2 2 6 -3 8 ; also in V lastos (1981). (1972) (ed .), P la to , vol. II , M odern Stu dies in P h ilosop h y, London: M acm illan. (1975), P la to ’s U niverse, O xford: C larendon Press. (1981), P latonic S tu d ies, revised edn, Princeton: P rinceton U n iv ersity Press.



W alsh, J. (1963), A risto tle 's Conception o f M o ra ! W eakness, N e w York and London: C olum bia U niversity Press. W ood, E .M . and W ood, N . (1978), Class Ideology and A n c ie n t P olitical Theory, Oxford: Blackwell. W ootton, B. (1963), Crime and the C rim inal L a w , London: Stevens. W oozley, A .D . (1972), ‘Plato on killing in anger’, P hilosophical Q u a rterly, 22, pp. 3 03-17.

Index agricu ltu re, 7, 1 0 1 -2 , 106, 110 akrasia, 5 0 -2 , 53, 151; see also injustice, involuntary a lie n s, resident, 102, 108-10 an ger a c ts don e in, 1 4 6 -7 , 1 6 0 -1 , 163 as cause o f injustice, 46, 157-9 a n o m ie and alienation, 9 8 -9 A q u in a s, 26, 27, 32, 33, 169 A r isto tle, 32, 33, 48, 50, 74, 87, 102, 117, 118, 119, 169 art corru p tin g effect of, 1 2 6 -7 , 129-30 criteria o f ex cellen ce, 53, 126-7 see also im itation; m usic and dance a sse m b ly , 8, 113, 118, 186 a th e ism case for, 168 c a u ses of, 167 d an gers of, 1 6 7-9 see also im piety A th e n s, 4, 5 - 6 , 7, 41, 4 2 , 7 6 -7 , 97, 113, 1 1 5 -1 6 , 127 A u g u stin e, St, 26 A u stin , John, 25

Barker, Sir E rnest, 1, 54, 117 -1 9 B e n th a m , Jerem y, 25, 68, 89, 165 B u ry, R .G ., 11

C h a rm id es, 9, 14, 54 c h e ck s and balances, 75, 1 1 5 -1 6 , 122; see also m ixed constitution C h ern iss, H .F ., 135

ch oice, 6 0-1 o f lives, 6 4 -5 , 6 7 -7 0 C icero, 26 c ity, proposed , see M agnesia c ity, w ardens, 189 claim s to rule, 7 3 -4 , 116 classes, social, 1 0 8 -1 1 , 118 closed society, 179-81 co llectivism , 181, 184; see also individual c o n stitu tio n , proposed for M agnesia, 7 -8 , 1 1 2 -2 2 , 186-91 relation to G reek practice, 7, 112, 115-16 cou n cil, 1 1 8 -2 0 , 186-7 country w ardens, 114, 132, 188 courage, 37, 45, 56, 58, 124 courts, see justice, adm inistration o f C ratylus, 128 C rete, 4, 6, 25, 8 9 -9 1 , 105, 181; see also D orian in stitution s C rito, 62 C rom bie, I.M ., 60, 63

death penalty, 137, 147-8 d em ocracy, 18, 73, 7 7 -8 , 97, 114, 116, 119-22 G reek con cep t of, 120-2 dem ocratic theory, 121-2 d eterm inism , 4 9 -5 0 deterrence, 1 3 9 -4 0 , 141, 146, 165; see also pu nish m en t des P laces, E. and D ies, A ., 11 D io g en es L aertius, 2 -3

IN D EX divine and hum an goods, 37, 39, 45, 66, 144 D odds, E.R ., 9 Dorian institutions, 4 - 5 , 25, 3 5 -6 , 46,72, 10 3 -4 , 123, 181; see also Crete, Sparta Dorian league, 5, 72 drinking parties, 5, 46, 51, 55, 124 Durkheim , E m ile, 99, 104 econom ic organization, 7 -8 , 102-3 education, 8. 9 -1 0 , 15, 4 2 -4 , 53, 123-36 officials in charge of, 132, 189 organization and curriculum , 132-3 see also literature; mathematics; m usic and dance elections, 8, 118-22, 186-7 Engels, Friedrich, 88 Epinomis, 2 -3 , 174 equality, 117-121 E uthydem us, 14 evil, sources of, 172-3 fam ily, 103, 105 form s, 2, 14, 2 1 -2 , 128, 133-6 G igon, O., 46 G od(s), 91, 166 concern o f for m an, 176-7 existence of, 169-73 man as plaything of, 130-1 as source o f law, 24, 2 8 -9 see also religion G orgem anns, H ., 9 Gorgias, 9, 14, 40, 43, 52, 54, 59, 62, 97, 142, 145, 156, 176 Grote, G ., 73 guardians o f the laws, 8, 113, 114, 116, 187-8 relation to nocturnal council, 134 Hall, J., 33 Hall, R .W ., 40, 82 happiness and justice, 59, 6 2 -4 , 144


harm, see injury hedonism , see pleasure historicism , 96, 183 H um e, D avid, 88, 1 77 ignorance, as cause o f vice, 4 6 - 7 , 50, 157-9; see also injustice, involuntary im itation, 126-9, 131 im piety law against, 150, 177-8 varieties of, 167, 177 see also atheism individual, P lato’s attitude to, 3 8 -4 2 , 4 4 ,1 0 3 - 4 ,1 0 6 ,1 1 6 ,1 7 9 , 181-2, 184 injury, 142, 144, 15 2 -4 injustice as disease, 14 1 -4 , 150, 160 involuntary, 151-6 sources of, 157-9 see also ignorance Jowett, B., 11 juries, see justice, adm inistration o f justice, 37, 45, 58, 8 5 -6 , 117 adm inistration of, 1 1 3 -1 4 , 14 6 -7 , 190-1 just punishm ent, paradox of, 152 knowledge and law, 28-31 and virtue, 4 1 -2 , 5 6 -7 see also reason; ph ilosopher king Laches, 9 law, 2 3 -3 4 as alternative to rule o f the w ise, 17-19, 3 1 -2 , 84, 147 constitutional, 6, 8 3 -4 , 112 Greek concept of, 2 3 -4 fear of, 76 and morality, 2 3 -4 , 4 0 -2 natural, 26, 3 3 -4 purpose of, 2 4 -8 , 35 -4 4 and reason, 2 4 -3 0 , 32, 124, 128



a n d religion , 4 , 24, 2 8 -9 , 166 so v e r e ig n ty of, 8, 21, 8 0 -6 , 185 sta b ility o f, 32, 8 1 -4 , 126, 183 u n w r itte n , 24 Law s c h aracter of, 9 -1 0 d a te , 23 E n g lish translations, 1 0 -12 im p o r ta n c e, 1-2 , 3 -4 in c o n siste n c ie s, 3 str u c tu r e, 4 - 7 sty le , 3 a s tex tb o o k , 9 -1 0 , 43, 133 see also R epublic le g isla tio n , conditions for, 6, 90 -4 ; see also law le g isla to r , role of, 3 1 -3 L e tte r s , 1 9 -2 0 , 31 lib e r ty , 4 0 -2 , 4 3 -4 , 7 6 -8 , 98, 106, 1 0 8 ,1 3 1 -2 , 183-4 lite r a tu r e , in education, 132-3 lo t, selectio n by, 74, 119; see also electio n s M a c d o w e ll, D .M ., 149 m a g istr a te s, 186-90 p o w er of, 113-15 lia b le to prosecution, 115 M a g n e sia c o n d itio n s presup posed , 8 9 -9 0 , 101

p u rp o se o f P lato’s accou nt, 90 M a g u ir e , J.P ., 33 M a r c u s A u reliu s, 33 m a rk et w ardens, 189 m a r x ism , 84, 9 8 -9 m a th e m a tic s, 133 m e a ls , com m on , 4 -5 , 1 0 3-4 m ilita r y officers, 188 M ill, J .S ., 40, 77 m ix e d c on stitu tion , 5 -6 , 8, 7 4 -5 , 7 7 -9 , 81, 115, 116-20; see also checks and balances m o n a r c h y , 17, 77, 116-17 M o n te sq u ie u , C harles, 81, 116 M o r e , Sir T hom as, 88

M orrow , G .R ., 7, 90, 103, 107, 112, 116, 187 m otion se lf-m o v in g , 169-72 varieties of, 169-70 M üller, G ., 9, 46 m usic and dance, 53, 76, 7 8 -9 , 97, 1 2 4 -3 0 , 1 3 4-5 criteria o f excellen ce, 1 2 6 -7 , 129 educational role, 1 2 5 -6 , 132-3 inn ovation in, 7 6 -7 , 96, 12 6 -7 , 150 see also art nature and con ven tion , 29, 41, 168 distin guished from chance and art, 168 nocturnal cou n cil, 7, 28, 30, 37, 57, 112, 133-6, 177, 189 -9 0 relation to guardians o f the laws, 134 objectivism , 4 1 -2 O ’Brien, M .J., 9 officials, see m agistrates old age, respect for, 2 9 -3 0 , 124-5 oligarchy, 91, 92, 11 7 -1 9 o p en society, 179, 182-3 O stw ald, M ., 23 P angle, T .L ., 11-12 P arm enides, 136 peace, as object o f law, 36, 3 8 -9 , 84, 183 P elop on n ese, history of, 5, 7 2 -3 , 7 4 -5 , 80; see also Sparta P ericles, 38, 39 Persia, 5 -6 , 7 6 -7 persuasion, 4 2 -4 P haedo, 27, 59 P haedrus, 52 P hilebus, 54, 59, 6 9 -7 0 , 78, 174 P hilip o f O pus, 2 -3 p h ilosopher king, 8, 14 -1 5 , 18 -1 9 , 21, 113-15 play, education as, 130-1

IND EX pleasure and art, 76, 126 and choice, 6 0 -1 , 6 6 -9 and the good life, 5 9 -7 0 and happiness, 6 2 -4 and virtue, 4 6 -7 , 50, 5 2 -4 , 55, 6 0 -1 , 66, 78, 124 poetry, control of, 43, 62 politics as a matter o f k now ledge, 14-15, 1 7 -1 8 ,2 1 purpose of, 22 and rationality, 9 2 -3 , 113-15 pollution, 148-9 Polybius, 116 Popper, Sir Karl, 11-12, 43, 9 5 -6 , 107, 179-85 population, 7, 99-101 preambles, 6, 10, 19, 4 2 -4 , 133 prem editation, 160-1, 163 Protagoras, 168 Protagoras, 9, 14, 52, 59, 71, 140 punishm ent, 137-50 aims of, 139-43 as cure, 140-8, 150, 153, 162, 164-5 as education, 148-9 penalties, 137-9 see also deterrence; retribution reason and desire, 9, 47, 5 0 -4 , 57, 6 0 -1 , 145, 157-9 role in politics, 21, 9 2 -3 , 94, 97, 113-15 and law, 26, 2 8 -3 0 , 32, 124, 128 religion, 7, 130-1, 166-78 officials responsible for, 188 and science, 175 see also G od(s) Republic, 15,27, 5 9 ,6 9 , 7 1 ,8 5 -6 ,8 7 , 9 7 -8 , 101, 128, 142, 145, 176 account of virtue, 39, 40, 45, 48, 55, 57, 64 account of the soul, 47, 157 arts in, 127-8


guardians, 1 1 3 -1 5 , 13 5 -6 ; see also philosopher king im practicably of, 1 6 -1 7 relation to La-zcs, 1 - 2 ,8 - 1 0 , 1 3 -1 4 , 1 8 ,2 1 - 2 ,9 4 - 6 , 1 1 4 -1 5 , 180 social institution s, 1 0 4 ,1 0 6 , 108-11 responsibility, 4 8 -5 0 , 144, 1 5 1 -6 5 retribution, 1 3 9 -4 0 , 176 Ritter, C., 158 Rom ans, Epistle to th e, 33 Saunders, T .J., 1 0 -1 1 , 1 4 ,8 7 , 90, 109, 118, 135, 142, 143, 152, 155, 159, 187 scrutineers, 112, 189 self-con trol, see sophrosune self-d ecep tion , 54 Shorey, P ., 155 slavery, 1 0 2 ,1 0 6 -8 , 109, 1 1 0 -1 1 , 120 social problem s, cau ses of, 9 7 -9 society, 97-111 solidarity, social. 99, 104, 1 0 9 -1 0 , 121-2 Sophist, 1 ,2 1 , 136 sophrosune, 5 -6 , 36, 46, 5 0 -2 , 5 4 -6 , 6 0 -1 , 8 4 -6 , 9 7 - 8 , 124 soul nature of, 4 6 -8 , 57, 157 priority of, 134, 1 6 9 -7 2 Sparta, 4, 5, 54, 7 4 -5 , 80, 105, 106, 116, 124, 181 ; see also D orian institutions; P elop on n ese state and individual, 39, 8 5 -6 S ta te sm a n , 14, 1 7 -1 9 , 21, 71, 77, 84, 88, 114, 128, 136 Strauss, L ., 73 Syracuse, 13-14, 1 9 -2 0 , 92 Taylor, A .E ., 11. 77, 106 Thales, 173 T heognis 37, 38 Tim aeus, 125, 128. 131, 136 responsibility in, 4 8 -5 0 , 52 the soul in, 4 7 ,4 8 - 9 ,6 6 - 7 ,1 4 5 - 6 , 171 theology and cosm ology 173-4



tim oria, 13 9 -4 0 trade, 7 -8 , 101, 102-3 tribalism , 180-1 tribes, 103 tyranny, 15, 17, 9 1 -2 T yrtaeus, 37 universe order of, 78, 131, 13 3 -5 , 136, 175 ruler o f, 174, 176; see also world soul utopia, 8 7 -9 , 9 2 -6 virtue, 37, 4 5 -5 8 as aim o f education, 12 3 -4 as goal o f legislation, 3 7 -4 2 , 184

3EX and know ledge, 9, 4 1 -2 , 45, 5 6 -7 as qualification for office, 117-18 un ity o f, 37, 45, 5 6 -8 , 134, 135 and happiness, 4 0 - 1 ,5 9 ,6 2 - 4 ,1 4 4 voluntary and involuntary, 151, 161-3 w eakness o f w ill, see akrasia w isdom , 3 7 ,4 5 ,4 8 ,5 6 ,5 7 ,6 6 ; see also know ledge; reason'' w om en, position of, 104-6, 120, 186 W ood, E .M . and W ood, N ., 110 W ootton, Lady Barbara, 143, 164-5 W oozley, A .D ., 161-4 w orld sou l, 172-4 w ou nd in g, 160-1

AN INTRODUCTION TO PLATO'S LAWS An introduction to Plato's Laws-Plato,R.F. Stalley (1983)  

MACEDONIA is GREECE and will always be GREECE- (if they are desperate to steal a name, Monkeydonkeys suits them just fine) ΚΑΤΩ Η ΣΥΓΚΥΒΕΡΝ...

AN INTRODUCTION TO PLATO'S LAWS An introduction to Plato's Laws-Plato,R.F. Stalley (1983)  

MACEDONIA is GREECE and will always be GREECE- (if they are desperate to steal a name, Monkeydonkeys suits them just fine) ΚΑΤΩ Η ΣΥΓΚΥΒΕΡΝ...

Profile for yandex963