CartesianLinguistics
AChapterintheHistoryofRationalistThought
NoamChomsky
ProfessorofLinguistics
MassachusettsInstituteofTechnology
ThirdEdition
editedwithanewintroductionby
JamesMcGilvray
CAMBRIDGE UNIVERSITY PRESS
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo
Cambridge University Press
The Edinburgh Building, Cambridge CB2 8RU, UK
Published in the United States of America by Cambridge University Press, New York www.cambridge.org
Information on this title: www.cambridge.org/9780521881760
This publication is in copyright. Subject to statutory exception and to the provision of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press.
First published in print format
© James McGilvray 2009 2009
ISBN-13 978-0-511-50685-7
ISBN-13 978-0-521-88176-0
ISBN-13 978-0-521-70817-3
paperback eBook (EBL) hardback
Cambridge University Press has no responsibility for the persistence or accuracy of urls for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.
JamesMcGilvray
IAnoverview
CartesianLinguistics (CL)beganasamanuscriptwrittenwhileNoam Chomskywasa35-year-oldfellowoftheAmericanCouncilofLearned Societies.AnearlyversionofitwaspreparedforpresentationasaChristian GausslectureonCriticismatPrincetonUniversityearlyin1964.Perhaps becauseitprovedbeyondtheaudience,itwasnotdelivered,andChomsky presentedagenerallectureonlinguisticsasunderstoodatthetime.Themanuscript,however,wasrevisedandpublishedin1966.Anintellectualtourde force, CL isnotaneasytexttoread,butitiscertainlyarewardingone.Itisan unprecedentedand – sofar – unequalledlinguistic–philosophicalstudyof linguisticcreativityandthenatureofthemindthatisabletoproduceit.
CL beginsbydescribingthesortoflinguisticcreativitythatisfoundwith virtuallyeverysentenceproducedbyanyperson,includingyoungchildren.As itssubtitle(“AChapterintheHistoryofRationalistThought”)suggestsitwill, though, CL soonturnstofocusonthekindofmindthatisrequiredtomakethis sortofcreativitypossible,andonthebestwaytostudysuchamind,and languageinit.Theseventeenth-centuryphilosopherRenéDescartesfigures prominentlyinthediscussionandthebook’stitle.Thisisbecausehewasamong thefirsttorecognizetheimportanceofthis ‘ordinary’ formoflinguistic creativity – creativityexhibitedbyeveryone,notjustpoets – forthestudyof thehumanmind.1 Becauseofthis,andbecauseagroupoflinguistsand philosopherswhocameafterDescartessharedhisinsight(eventhoughthey mighthavedisagreedwithDescartesinotherways)anddirectedtheirstudyof languageandthemindtowardsdealingwiththeissuesitposed,Chomskytitled hisstudyoftheirworkandoftheiropponents’ CartesianLinguistics.Those whoChomskythinkscanplausiblydealwiththeissuesthatlinguisticcreativity posesforthestudyofmindandlanguagehecalls “rationalists”;thosewho cannot,hecalls “empiricists.” InthisintroductionIadd ‘romantic’ to Chomsky’slabel ‘rationalist’ toemphasizewhatisimplicitinastudyof ordinarylinguisticcreativityanditsroleinhumanthoughtandaction:thatthe rationalistsofinteresttohim,liketheromanticshefocuseson,recognizethe
centralityineverydaylifeoffreedomofthoughtandaction,andtheytrywith theirviewofthehumanmindtospeaktohowthiscreativityispossible.For manyofthem – andcertainlyforChomskyinparticular – thenatureof languageitselfasacomponentofthemind/brainplaysacentralroleinthe explanation.
CartesianLinguistics hasmanyassets.OneisthatitplacesChomsky’s efforttoconstructascienceoflanguageinabroadhistoricalcontext.Itdoes notpretendtobeaworkinintellectualhistory;itistoobriefandtooselective intheindividualsitdiscussesforthat.2 Butitdoesofferimportantinsights intotheworksofhistoricalfigures,anduncoversanddiscussesoften-ignored butclearlyrelevanthistoricaltexts.Italsorevitalizesarivalrythathaslasted forcenturiesandthat – in1966andstillnow – continuesinthecognitive sciences.
Anotherassetistheunderstandingitgivesofthebasicobservationsthatlie behindChomsky’s – andotherrationalist –romantics’– researchstrategyor fundamentalmethodologyforthestudyoflanguageandmind.Therearetwo setsofobservations.One – the “povertyofthestimulus” facts – focuseson thegapbetweenwhatmindsobtainwhentheyacquirearichandstructured cognitivecapacitysuchasvisionorlanguageandthesmalland ‘impoverished’ inputthatthemindreceivesasitdevelopsthecapacity.Another – the “creativeaspectoflanguageuse” observations – focusesonthefactthat people,evensmallchildren,uselanguageinwaysthatareuncausedand innovative,whilestillappropriate.Becauseofitsextensivediscussionof linguisticcreativity, CartesianLinguistics focusesmorethananyoftherest ofChomsky’sworksonthecreativityfacts,andexplorestheirimplicationsfor thescienceofmindandtheexplanationofbehavior – andittouchesontheir broaderimplicationsforpoliticsandeducation,andevenart – especially poetry.Bydescribingaformofcreativitythateveryoneexercisesintheir useoflanguage – acreativitythatfiguresinvirtuallyallthoughtandaction wherelanguagefigures – ithighlightsacommonphenomenonthatseemsto defyscientificexplanation.Humansuselanguagecreativelyroutinely,yetthis routineuseseemstobeanexerciseoffreewill.Ifitis,itwouldhardlybe surprisingifthetoolsofscience,whichworkwellwithdeterminationor randomness,failtodescribeorexplaintheuseoflanguage.Freeactionsare uncaused,hencenotdetermined,yettheyareneverthelesstypicallyappropriate,hencenotrandom.ToChomsky,astootherrationalist –romantics,this suggeststhatifyouwanttoconstructascienceofmindandlanguage,you shouldavoidtryingtoconstructascienceofhowpeopleusetheirminds,and especiallytheirlanguage.Donottrytoconstructascienceoflinguistic behavior.Perhaps,infact,giventhedegreetowhichlanguageinfusesand shapessomuchofhowweunderstandandact,donottrytoconstruct sciencesofactionandbehavioringeneral.
Thisis not tosaythatoneshouldnottrytoconstructascience – infact,many sciences – ofthemind.Anditdidnotstopanyoftherationalist–romantics –withthepartialbutpuzzlingexceptionofDescartes3 – fromtryingtoconstruct sciencesofmindandlanguage.
Afterall,thepovertyofthestimulusfactsforlanguageandotherdomains, suchasvisionandfacialrecognition,suggestthattherecanbesciencesofat leastsomecomponentsofthemind.Theyseemtoindicatethatthemindismade upofinnatesystemsthatgrowautomatically,muchasdothehumanheartand liver.Adoptingsucharesearchstrategyiscalled “nativism”;rationalist–romanticsare nativists.Adoptingthisstrategy,Descartestoanextentaside, rationalist–romanticstrategistsaimedinonewayoranother,andwithdifferent degreesofsuccess,toofferspecies-universal,objectivetheoriesofvarious componentsofthemind,componentsofthemindthatthetraditioncalled “faculties.” Mentalfaculties – wewouldnowsay “modules”– doseemto workdeterminately.Oratleast,rationalist–romanticefforts,suchasDescartes’s venturesintoacomputationaltheoryofvision,thePort-Royalgrammarians’ effortstoproduceaUniversal(“philosophical”)Grammar,Cudworth’sspeculationsaboutthenatureof “aninnatecognoscitivepower,” andvon Humboldt’sefforttodealwiththementalmachineryneededtoprovidefor creativity,havemanagedwithvaryingsuccesstodealwithaspectsofdeterministicsciencesofvariousfaculties.Progresshasbeenmuchgreatersincethe mid-1950s.DavidMarrandChomskyandtheirco-workershaveproduced advancednativistsciencesofvisionandlanguage.Theirobvioussuccess –andthemorelimitedsuccessesofearlierrationalist –romantics – seemsto indicatethatitispossibletoconstructsciencesofvariouspartsofthemind/ brain,althoughnotofthewaysinwhichhumansusewhatthesecomponents providethemtodealwiththeworldandotherwisesolvevariousproblems.
Theveryrealriskoffailurewhenonetriestogooutsidetheheadtodealwith thecomplexitiesofhumanactionandbehaviorsuggeststhatthescientistof mindshouldfocusonwhatis ‘inthehead’ andhowwhatisinaparticular person’sheadcomestohavetheshapeand ‘content’ thatitdoes – howit ‘grows’.Chomsky’stermforthisstrategyforthestudyofmindis “internalist. ” Inadditiontoadoptingnativistassumptions,rationalist–romanticsadoptan internalist approachtothesciencesofmind.Linguisticcreativityobservations seemtosuggestthatthisistheonlyonelikelytoprovefullysuccessful.Of course,someoftheevidenceforascienceofwhatisinthehead(althoughbyno meansall)comesfromobservinghowapersonbehaves – inthecaseof language,howapersonpronouncesasentence,andwhenandwheres/he usesit,amongotherthings.But,obviously,aninternalisttheoryofwhatisin theheadisnotjustacompilationofthisoranyotherkindofevidence;the theoryisconcernedwithwhatreallyisintheheadandhowitworks.Itis concernedwiththeprinciplesofoperationofafaculty/module,withitsinternal
inputsandoutputs,andwithhowthisfacultydevelopsandgrowsasthe organismdevelops.Itdoesnotfollow,bytheway,thattherationalist –romantic theorist’snativistandinternalistapproachtothemindhasnothingtosay aboutcreativelinguisticbehaviorandaction.Forassuggestedaboveitcan –anddoes – accountbothforwhatinthehumanmindmakeslinguisticcreativity possible,andforwhylinguisticallyinformedcreativebehaviorisavailableonly tohumans.
InrecentyearsChomsky’slabelforhisapproachtomindandlanguagehas changedfrom “rationalistic” to “biolinguistic.” Heandothersworkinginthe fieldarenowcalled “biolinguists.” Thelabelchangehighlightsacharacteristic ofChomsky’seffortstoconstructsciencesoflanguagefromthestartofhis work;theaimhasalwaysbeentotrytoaccommodatethescienceoflanguageto somenaturalscience,thusbiology – forbiologyalonecanexplainhowlanguageisinnate,whyitisuniquetohumans,andhowitgrows.Nevertheless, biolinguisticresearchstrategyisjusttherationalist–romantics’ nativistand internaliststrategyupdated.Thesamepovertyandcreativityobservations continuetobehonored.Rationalist–romantic(RR)researchstrategyisalive andwellinthepracticesofbiolinguists.
Athirdassetof CartesianLinguistics (CL)isthatitpointstothecentralrole oflinguisticcreativityinalmostallhumanaffairs.Unlikeorganismsthatlack language,wecananddothinkandtalkaboutanything,anywhere;wespeculate andwonder,questionanddoubt,organizeourselvesinnon-kinandnon-contact communities,cooperatetocarryoutprojects,liveandthriveinmanyenvironments,engageinfantasyandplay,andsoon.Ourcognitivecapacitiesingeneral aremuchmoreflexiblethanthoseofothercreatures.Wecanadapttovarious environmentsandsolve(andcreate)problemswelloutoftherangeofanyother kindofcreature.Wecanmakeandinterpretart,developvariousformsof religionandthekindsofexplanationtheyoffer,developourselvesandour cultures.Linguisticcreativitysurelyhasacentralroletoplayinallthis,andthe operationsofthelanguagefacultyinmakingpossiblethiscentralfeatureof whatmakesushuman – givingusourdistinctivehumannatures.TheimplicationsofthisgiftwerenotlostonA.W.SchlegelorvonHumboldtor,following them,Chomsky.Somearepolitical.Idiscusssomeofthesebrieflyinanother sectionofthisintroduction.
Theintroductionhasfourparts.Inthenext,PartII,Idiscussinmoredetailthe placeoftherationalist–romanticviewofthemindanditsstudyinChomsky’s work,andexplainhowthisview,andalongwithithisscienceofmind,have developedsince1966’s CartesianLinguistics tobecomecontemporarybiolinguistics.Ialsocontrastittotheempiricistviewoftherightstrategytousein investigatingthemind,focusingonsomeofempiricism ’scontemporaryversions.Oneofmyaimsinthispartistoemphasizetheimportanceoftaking Descartes’screativityobservationsseriously.Doingthiswhilealsotaking
seriouslythepovertyofthestimulusobservationsleads – Isuggest – tothe remarkableprogressseeninrecentyearsinthescientificstudyofmindand language.PartIIIfocusesonDescartesandhiscontributionsandfailures.The other,PartIV,takesupbrieflysomeoftheimplicationsofbiolinguisticstudyof languageandmindforpoliticsandeducation.
ReadersmightwanttoreadChomsky’srichtextrightnow,andreturntoPart IItolearnmoreabouttheRRstrategyforthestudyofmind,andaboutprogress inChomsky’sapproachtolanguagesince1966.Alternatively,theymightwant toreadPartIInowtogetanoverviewoftheRRresearchstrategyandwhyit seemstoleadtoprogressinthescienceofmindwhereempiriciststrategyseems tofail.PartsIIIandIVareforthosewhobecomecuriousaboutDescartes’s specificcontributions,andabouttheimplicationsofaninternalistandnativist researchstrategyforpoliticsandeducation.
Thisthirdeditionof CL,likeanearliersecond,isinEnglishalone.Inthe original1966text,Chomskyleftmanyquotationsdrawnfromtheworksof thosehediscussesinFrenchorGerman;forthemostpart,heusedtranslations onlyiftheywereavailableatthetime.
Tomakethesecondeditionmoreaccessibletothewideraudience CL deserves,Susan-JudithHoffmanntranslatedthetextsinGermanthat remained,andRobertStoothoffthepartsthatremainedinFrench.Mostof theFrenchtranslationsofDescartes’sworksthatChomskyhadincludedin theoriginalhavebeenreplacedwithimprovedonesfromlatertranslations –specifically,thoseavailableinvolumes1–3ofJohnCottingham,Robert Stoothoff,DugaldMurdoch,and(forthethirdvolumealone)Anthony Kenny’sCambridgeeditionsof ThePhilosophicalWritingsofDescartes. Withalltranslations,aneffortwasmadetosuitboththeoriginaltextand Chomsky’sterminology.Sometimesthisrequiredminormodificationsin availabletranslationstomakeclearwhatChomskyseemstohaveseenin theuntranslatedoriginals.Icannot – noonecan – guaranteethatthetranslationsorchangesexactlycapturewhattheoriginaltextsintended,ofcourse, buttheoverallresultmeets,Ibelieve,thedemandsofbothscholarand student,thankstoProfessorsStoothoffandHoffmann’sadmirablework. Finally,Ithankaformergraduatestudent(nowprofessor),SteveMcKay, forhisworkinthepreparationofanindexforthesecondedition;theoriginal hadnoindexatall.ThecurrentindexmodifiesMcKay’stoaccommodatethis introduction.
Notethefollowingconventions:Chomsky’sendnotenumberscontinuefrom thoseofthenewintroduction.EditorialadditionstoChomsky’snotesappearin squarebrackets([ ]);forthemostpart,theseadditionsoffersuggestionsfor furtherstudy.Chomsky’sreferencestotextsandpagesremainastheywerein theoriginal;alladditionshavetheform(Author,dateofpublication:page). Iaddedbibliographicalitemsdatedafter1966.
NoamChomskyreadthisintroductionearlyin2008andsentmemanyuseful comments.Iamverygrateful;hiscommentsledtoseveralchangesand improvements.Intwoplaces,Isimplyquotedwhathehadtosay.Iamalso verygratefultoCedricBoeckx,OranMagal,andJuhaniYli-Vakkuriforreading draftsofthisintroduction.Idon’tdoubtthaterrorsremain.Theyare,however, entirelymine.
IIThescienceofmindandlanguage
II.1Creativityandpoverty:internalists,nativists,andtheiropponents
Rationalist–romantics(RRs)andempiricistsdifferagreatdealintheirviewsof themindand – notsurprisingly – intheirviewsofhowthemindshouldbe studied.Theydifferbothinhowtheyconceiveofthemindhavingthe ‘shape’ andcontentitdoes,andinhowtheyconceiveoftheroleoftheworldoutsidethe headinshapingandgivingcontent.Empiricistsclaimthatwe learn muchof whatweget – atleast,whenitcomesto ‘higher ’ conceptsandcognitive processes.RRsdisagree;thesearemostlyinnate.4 Comparingtheseviews highlightsthefeaturesofeach,andallowsustoaskwhichview,andwhich researchstrategybasedonthatview,hasthebestprospectsofsuccess.
Toillustratetheirdifferences,letuslookathoweachcampconceivesoftwo kindsofmentalentitiesandhowtheycometobeinthemind – howtheyare acquiredorlearned.Oneclassconsistsof ‘atomic’ conceptssuchasWATER, DRINK,COLD,andthousandsofothersthatweuseinvariouswaystocarry outvariouscognitivetasks,suchasdescribing,speculating,reminiscing,telling stories,etc.Theotherclassof ‘entities’ consistsoftherulesorprinciplesthat governhowthemindputstheelementalconceptsthatwordsexpresstogetherto assemblethecomplexconceptsexpressedbyphrasesandsentences.Phrases include “drinkcoldwater,” andendlessothers;sentencesinclude “Janewill onlydrinkcoldwater” andendlessothers.Humans – orbetter,humanminds –routinelyputtogethercomplexessuchasthese.5 TheRRsholdthatthemind’s conceptsandthewaysofputtingthemtogetherinlanguageandthoughtare largelyinnatelyconfigured;theyalsohold,then,thattherightwaytostudythe mindistoconstructtheoriesofthevarioussortsofinnermentalmachinerythat putconceptsinplaceor ‘activates ’ them,configurestheminformsthatthe machineryallowsorrequires,anddoesthesamefortherulesorprinciplesthat governhowtoputconceptstogetherinthecomplexformsexpressedby sentences.TheRRtheoristisa nativist (someonewhomaintainsthatboth conceptsandthewaystoputthemtogethertomakecomplexessuchasthose expressedbysentencesaresomehowinnate,implicitinthemind).Andbecause theRRresearcherisanativistandtriestosaywhatconceptsandcombinatory mechanismsareandhowtheydevelopinachild’sautomaticprocessof
maturationbyconstructingtheoriesoftheinnatemechanismsandtheiroperationswithouttryingtoincludeanyobjectsoutsidetheheadinthesubject mattersoftheirtheories,RRtheoristsalsoadoptan internalist researchstrategy. RRs(seeinthisregardespecially CL’sdiscussionsofvonHumboldtand HerbertofCherbury)pointtowhattheyseeasastrongconnectionbetween nativismandthephenomenaofeverydaylinguisticcreativity. ‘Ordinary’ linguisticcreativityalongwithitsimportantconsequences – thecapacitytoengage infantasy,speculation,play,planning,thoughtunconnectedtocurrentcircumstance,plusthecapacitytoconstruct ‘theories’ oftheworld,suchasspeculating whoisgoingtowinthenextelectionorthenextgameoffootball – isreadily availabletoeveryoneatanearlyage,RRshold,onlybecausehundredsof thousandsofrichlyendowedlinguisticallyexpressedconceptsandthemeansof puttingthemtogetherareinnateandthusreadilyavailableatanearlyage. Becausetheyare,children’smindsreadilyprovideinnovativesentences,which thechildcanuseinmultipleways.Anyonecanobservementalcreativityin youngchildren – itisfoundinusingoften-novelsentencesinunderstandingand doinginvariousways.Itisexhibitednotjustinspeech,butinturningcardboard boxesintohouses,inachild’sfantasies,inwonderingabouthowsomething works,inchildren’sestimatesofwhattheirparentsandotherchildrenintend,in theirexperimentationwithvarioustoolsandtoys,andsoon.Theissueishow youngchildrencanmanagetobesocreativeatayoungage – certainlybythe timetheyarefourorso,oftenbeforethen.Sinceonemustassumethatwitha childaswithanyoneelse,theconceptualtoolsoneneedstoclassifyandthink, andthecombinatorymechanismsthatallowonetoputconceptstogetherin variouskindsofarrangementsmustbeinplacebeforetheycanbeassembledin complexes,theonlywaytoexplaintheearlyappearanceofcreativityisto assumeinnatenessofbothconceptsandcombinatoryprinciples.Anditisonly becausetheseconceptsandprinciplesofassemblyandthewaystoactivatethem withminimalexperiencearebuiltintochildren’smind – presumablylodgedin theirgenomeandthewaysitdevelopsorgrows – thatwecanquicklyunderstandtheircreativeefforts,andtheyours.Innatenessprovidesabasisfor understandingoneanother,evenatayoungage.Forinnateconceptscanbe thoughtofasthemeaningsofwords(lexicalitems,intechnicalterms);they constitutewords’‘internalcontent’ (orperhaps ‘intrinsiccontent’).
AssuggestedinPartI,RRsalsoemphasizeaconnectionbetweencreativity andtheirdecisiontoadoptinternalismasaresearchpolicyforthescientific studyofthemind.Considerwhathappensifonedecidestoconstructatheory(a sciencenow,notaguessabouttheoutcomeofafootballmatch)ofaninteresting andimportantaspectoftheuseoflanguageandconcepts – usinglanguageto refertothings.Attheveryleast,attemptingthisrequiresfocusingnotjuston wordsandhowtheyareassembledintophrasesandsentencesinasysteminthe head,butonrelationshipsbetweentheseinternalentitiesandthingsandclasses
ofthingsintheoutsideworld.Doingthisexpandsthesubjectmatterofone’s theorytoincludenotjustmentalobjects – conceptsandsuch – butthingsand classesofthingsintheworld,andperhapstheirpropertiestoo.Italsodemands thattherelationsbetweenwhatisinsidetheheadandwhatoutsidebe ‘natural’ anddeterminate,fixedperhapsbysomethinglikebiologicalgrowth.Thatisa dauntingand – ifthecreativityobservationsaretakenintoaccount – verylikely impossibletask.Onewillfindnodeterminatehead-worldrelationsofthesort requiredto ‘fix’ theusesofsentences.6
Yetmanycontemporaryphilosophers – Putnam,Kripke,Burge,Fodor,etc. –believethatinordertomakesenseofhowlanguageismeaningfulatall,andfor itswordstohavemeaning,onemustassumeadeterminateconnectionbetween somenouns,atleast,andthingsintheworld – asinglethingforapropername, oraclassofthingsforageneralterm.Therelationshipmustbedeterminate,or involveveryfewspecifiableoptions.Otherwisetoolsoftheory-construction fail.Proceedingonthisassumption,thesupposeddeterminaterelationshipis oftencalled “reference,” although “denotation” and “signification” arealso used.Itisoftenclaimedthatnouns,oratleastsomeofthem,refer “rigidly,” touseKripke’scolorfulterminology.Ordinarylinguisticcreativityposesa seriousproblemforanattempttoconstructatheoryofmeaningthatrequires determinatehead-worldrelationships.Ifyouholdthatmeaningdependson referenceandyouwantatheoryofmeaningforalanguage,youbetterhopethat foreachnoun,thereisadeterminatereferent.Orif,likeGottlobFrege(1892), youthinkthatareferentialrelationshiptothingsismorecomplicated,thata wordisfirstlinkedtoasense(forhim,anabstractobject),andasenseinturn fixesareference,youbetterhopethatforeverynounthereisasinglesense,and foreachsuchsense,asinglereferent.Otherwise,yourtheorywillhavetoallow forallofthecomplexandhighlyvariablefactorsthatfigureinaperson’suseof languageforvariouspurposes,andintheeffortspeoplemaketounderstand whatanotherperson’slinguisticactionsmean – whattheyintendbythem, includingwhattheyintend/meantoreferto,ifanything.Youwillhavetotake intoaccountchangesinspeakerintentions,inthekindofjobawordisbeing askedtodo(tellsomeonehowtogettoChicago,criticizeaworkofart…),inthe circumstancesofspeech,inironyasopposedtoflat-footeddescription,in fictionasopposedtofact,andsoon.Tospecifywhatthecontextofdiscussion is,youwillhavetosaywhatcountasthe “subjectswhichformtheimmediate focusofinterest” (toquotethephilosopherPeterStrawson);7 andthereislittle hopethatanyonecansaywhattheseareinawaythatallowsforanykindof population-wideuniformity,unlesspossibly – thelimitcase,andhardlyrelevantfortheconceptionoflanguage,meaning,andreferencethephilosophers underconsiderationhaveinmind – thepopulationconsistsofthespeakeralone, atatime,tryingtoaccomplishasingle,well-understoodtask.Moregenerally, thereisnoguaranteethatanything,evenwhendealingwithflat-footed
descriptionandsmallpopulations,canbefixeddeterminately.Tofixistofix languageuse.Unfortunatelyforyourprojectofconstructingatheorybasedon hopeslikethese,asDescarteslongagoandChomskyin CL andelsewhere(New HorizonsintheStudyofLanguageandMind – Chomsky 2000 – amongothers) pointout,peoplejustdonotcareaboutwhatyourtheoreticaleffortsdemand –theydonotwant,anddonotproduce,fixeduses,evenofnouns.8 Andyettoa degreethatseemstobeadequateforsolvingeverydaypracticalproblems,at least,peoplestillmanagetounderstandtheory-resistingfreeusesofexpressions.Resistingtheneedsofthosewhowouldliketohaveregularityandeven determination,peopleseemtobenefitfromtheircapacitytobecreative.They enjoyusingwordsinallsortsofways,allthewhilebeingadequately(forthe task(s)athand)understandableandspeakingappropriately.Apparently,usinga word – nounorotherword – inthesamewayallthetimeisastediousasputting awidgetinaslotonanassemblylineoverandover.Insum,innocasedoes anything determine howtheyoryoumustuseawordorunderstanditwhenused byanother,forwhateverpurpose,onwhicheveroccasion.Theuseoflanguage isaformofhumanaction,anditisonthefaceofitaparticularlyinnovativeand uncaused,yetcoherentandappropriatefreeformofit.
Nevertheless,someonedrawntothekindsofcasesKripkeandothersfocus ontoprovideamotivationfortakingpropernamesas “rigiddesignators” seriouslymightsuggestthatnothingelseexplainshowpeoplewithwidely differingviewsof,say,DickCheneycanstilluse “Cheney” andexpectothersto knowwhotheyintend.GivendifferentunderstandingsofCheney,onecannot relyonwhatthoseothershappentoknoworassumeaboutCheney.So – itis argued – theremustbe some referentialrelationshipthatdoesnotrelyatallon people’sknowledgeorunderstandingofCheney,oranyotherobjectoreventto whichonewantstorefer.ButthisattemptatconvincinganRRtheoristisbogus. Nothingoutsideofcontextofspeechorauthor-controlledcontextofwriting9 antecedentlyfixes areference – antecedently,thatis,tosomeone’susingaterm torefer,andsomeoneelseinterpretingwhatthespeakersays,usingwhatever resourcess/hehas.Ofcourse,theprocessofdeterminingwhatanotherperson “hasinmind” canfail,althoughourresourcesoftenprovesufficientlyreliable thatitdoesnotmatterforthepurposesofdiscourse.Theseresourcesinclude sharedbiologies,aswellasenvironments,communities,interests,choicesin lexicalpairingsofsoundsandthesemanticfeaturesofthehearer ’slexicon,and thelike.Theseusuallysuffice.Theymust:wordsdonotrefer,peopledo – and thosewhowouldunderstandthespeakermust,asbesttheycan,putthemselves intothepositionofthespeakerbyusingwhateverresourcestheyhavetofigure outwhatthespeakerhasinmind.
Twodifficultiesconfrontthosewhowanttoclaimthatthere ‘is’ areferential relationshipbetweennaturallanguagetermsandthings ‘outthere’.Oneisthat infewcases – perhapsnone – istherereasontothinkthattheworld ‘outthere’
actuallycontainsany ‘things’ ofthesortthefixedreferentialistshaveinmind. Londonisasetofbuildingsonaterritory,butit(thesame ‘thing’)couldbe movedupstreamtoavoidinundation;Chomskywrote FailedStates,which weighshalfakiloandit(thehalfkiloofwoodpulp)iscompelling(becauseit containsanargument);mypersonallibraryhas FailedStates andmyuniversity libraryhasittoo;Theseusbuiltashipandreplacedallofitsplankswhichwere thenreassembledinthesamepositions,butTheseus’sshipistherebuiltmodel, notthereassembledone.Thewaysweunderstandthingsarefixedbyour conceptualresources,andourconceptualresourcesclearlyallowforthingsto beabstractandconcreteatthesametime;theycontainwoodpulpandinformation;theyareoneyetmany;theyletownershipandresponsibilitytrump materialconstitution.Theseareonlyafewinnumerableillustrationsthatindicatethatwe ‘make’ thethingsofourworldtosuitourconceptualresources,and thattypicallythese ‘things’ areidentifiedintermsofourinterests,notsomekind ofobjectivestandards.Weroutinelynamepersons,butwhat ‘are’ personssuch asDickCheney?PERSONiswhatLockecalleda “forensic” concept,onethat suitsourneedtoassignresponsibilityforactionsandthatmaintainspsychic continuity.Thepointisgeneral,thethingsandclassesofthingsthatmakeupthe worldaswetypicallyunderstanditarenotthewell-definedentitiesofthe sciences.What,however,ofareferentialistfavorite,WATER?Surelywateris H2O?Chomsky(2000, 1995a)offersmanyexamplesthatindicatethatwe naturallanguageusershavenothinglikethescientist’sH2Oinmindwhenwe speakandthinkofwater.Wefindnodifficultyinsayingthatwaterbecomestea whenheatedandateabagisplacedinit.Ourwaterwashesusandour possessions;itmayormaynotbeclear;itiswhatisinariver,nomatterwhat itmaycontaininaddition,evenifpollutantsconstitutethemajority;watercan becalmordisturbed;andsoon.Mostoftheuniverse’swaterisinaglassystate (inasteroids,andthelike),yetifaglassismadeofthismaterial,itisnotoffered forchewingwhenoneasksforaglassofwater.Theseandotherexamples constitutethebackgroundforChomsky’sotherwiseenigmaticremarkthat “WaterisH2O” isnotasentenceofEnglish.ItisnotbecauseH2Obelongsto molecularchemistry,WATERiswhatournaturallanguageEnglish “water” expresses.Ifstillnotconvinced,Chomskypointstoaparallelinphonology.The syllable/ba/isinthehead.Itisnot ‘outthere’.Thepointisgeneral:linguistic soundsare ‘inthehead’.Theydonotissuefrompeople’smouths.Allthatissues frompeople’smouthswhentheyspeakisaseriesofcompressionsanddecompressionsintheair,not/ba/or/ta/.Justasthereisno/ba/or/ta/ ‘outthere’,so thereisnoLondon.10
Aseconddifficultyisthatnaturallanguagesdonotseemtohaveanything likewhatphilosophersandsomeotherscall “ proper names”– nounsthat ‘directly’ refertoasingleentity – orrigidlyreferringgeneraltermssuchas “water.” Languages(thelanguagesindividualshaveintheirheads)dohave
names,ofcourse;thatisasyntacticcategoryofexpression,onewhichmayor maynotbeaprimitiveofatheoryofalanguage.Andnamestendtohaveatleast somemeaning:mostpeoplewhenhearingwordssuchas Moses and Winchell willbydefaultassignthemsomethingliketheconceptualfeaturePERSON NAME.Theirspecificlexiconsmightassignspecificnamesmorethanthis.But whetherminimallyormoreheavilyspecified,namesdohavemeanings,or ‘expressconcepts’,understandingbythattheyhaveatleastsomesemantic features – andtheyhavemeaningsdistinctfromthepropernounsandrigidly referringtermspostulatedinphilosophicaldiscussion.Sincetheydo,itishard tounderstandwhyanyonewouldthinkthatatheoryofmeaningforanatural languagerequiresgoingoutsidethehead.
Perhaps,however,thereisanexplanationforthis:ananalogytoscienceand thepracticesofscientists,onethathasoftenmisledstudiesofnaturallanguage. Noticethatfamiliaritywithapersonandhisorhercircumstances,relianceupon folktheoriesandotherdefaultstrategies,andthelike,playnoroleinunderstandingtechnicalpresentationsinmathematicsandthenaturalsciences. Nevertheless,referenceforthegroupofparticipants(mathematiciansand scientists)isvirtuallydeterminate,andthetermstheyusereallydoseemto ‘referbythemselves’.Thisisnot,however,becausethesymbolsoftechnical workreally do refer ‘bythemselves’,butratherbecausealloftheparticipants canbeassumed – asFregeputit – to “graspthesamesense,” andthesenseis takenbyalltocharacterizeanentityorclassofentitiesdrawnfromthesubject matteroftheirjointproject,whetheritbemathematics,elementaryparticle physics,orformallinguistics.Thereisroomfordisagreementoverwhethera difficultproofsucceeds,orahypothesisiscorrect,butindoingtechnicalwork inascientificormathematicaldomain,itcanbeassumedeveryoneknowswhat aspeakeristalkingabout,whats/herefersto.Onephysicist’schiralanomalyis thesameasanother ’s,onemathematician’saleph-nullanother ’sbecausethey strivetobespeakingof ‘thesamething’,whateverthatmightbe.Thisis because,asChomskysuggests,inthedomainsofmathematicsandthenatural sciences,onefindsstrong ‘normative’ constraintsonsame-use,constraints not foundintheuseofnaturallanguage,wherepeopleemployandenjoylinguistic creativity.Everydayspeakersarenotengagedonaunifiedproject.Andas Chomskyalsopointsout,itisnosurprisethatFregeansemantictheories – those thatsupposeacommunitywithsharedthoughtsandshareduniformsymbolsfor expressingthesethoughts,andanassumedconstrainttobetalkingaboutthe samethingwhenevertheyuseaspecificsymbol – workquitewellwith mathematicsandthenaturalsciences(1996,ch.2).Buttheydonotworkwith naturallanguages,ahardlessonforthemanyphilosophersandsemanticians whotrytoadaptFregeansemanticstonaturallanguages.
Strongnormativeconstraintsonuse – the “conventions” ofDavidLewis (supposedlyneededinordertoallowforcommunicationandcooperationatall)
andthesupposedlydeterminate “practices” ofSellarsandcompany – donot exist.11 Theyarejustnotneededineverydayspeech.Wehavemanyresources availabletodealwithinterpretation,andspeakerandhearerfindattemptsto constrainfettering.Thatdoesnotmeanthatonecannothaveatheoryofmeaningforanaturallanguage.Butitmustbeinternalist.
Insum,thereisnoreferenceapartfromsomeonewhorefers;relationstothe outsideworld(andeveninasense ‘theworldoutside’ asunderstoodbythe conceptsexpressedinnaturallanguages),areestablishedbyandthroughactual uses.Thatistrueinthesciencesandeverydaydiscourse,althoughinthe sciencesandmath – asindicated – practicesare ‘normalized’ andcomeclose enoughtotheFregeanpictureofsemantictheorytoallowonetoidealizeand ignorethecontributionsofaperson.AllthisputsinternalistssuchasChomsky inwhatisthesedaysanunusualposition.Herejectstheverypopular(among linguistsandsomephilosophers)Fregeanmodelofsemantics(‘theoryofmeaning’),andalongwithitwhatJerryFodorcallsa “representationaltheoryof mind.” Ifyouholdthatnaturallanguagereference(thatwhichinvolvesuseof thetermsofnaturallanguagessuchas “London” bypeopleinvariablecircumstancesengagedondifferentprojectsandhavingvariableinterests)isnotanapt subjectmatterforscience,youmustalsoholdthatrepresentationofthingsinthe outsideworldbyuseofnaturallanguagetermsisnoteither.Indeed,youmust reject – orperhapsreinterpret – aconsiderablechunkofcontemporary “cognitivescience,” attheveryleast,thatchunkthatpurportstoofferasemanticsfor naturallanguagesthatassumesarelationbetweennaturallanguageentitiesand theworld.Perhaps,asFodor(1998)putit,arepresentational(essentially Fregean)theoryofsuchconceptsisthe “onlygameintown.” YetChomsky andothercontemporaryRRtheorists(thereareafew)seemtohavenoqualms aboutdoingcognitivescienceanddealingwithfullyinternallydetermined concepts/meanings.Isuspectthatisbecausetheyknowthatthereisanonre-presentationalnaturalisticscienceoflanguageandofwhatitprovidesthe mind(likelyintheformof “semanticfeatures”)inplace,andtheythinkthatthis sufficesforatheoryofnaturallanguagemeaningandmeaning-composition.If so,theycanlookatthelossofdeterminatemind–worldrelationshipssuchas referenceanddenotationwithequanimity.12 Indeed,theymightbequitewilling tomaintainthatnoscienceofvision,norothertheoryofthemind,needbe committedtoaFodorianrepresentationalview.13 Dothe “blobs” ofDavid Marr ’s Vision denoteanythingoutthere?Surelynot.His3-D ‘representations’ donot,either.Thepointsmadeaboveaboutthesyllables/ba/and/ta/in phonology(aninternalistscience)areworthconsideringagaininthisconnection.ThereisaninterestingwayinwhichChomskyagreeswiththephilosopher Wittgenstein(whoselaterworks,alongwithJ.L.Austin’s,hewasreadingwhen hewrotehismassive LogicalStructureofLinguisticTheory – aworkthattakes languageasa(natural)toolthatcanbeusedinvariousways).Wittgenstein
(1953)thoughtthatwordsandsentencesare ‘tools’ thatweusetocarryout variouseverydaytasks,andheheldthattheirmeaningsarethejobsthey perform.Sincetheyare,hethought,ifyouwanttoknowwhatexpressionE meansforpersonP,findouthows/heusesit – whatfunctionitservesin performingwhatevertasks/heiscarryingout.Thenhereasonedthatsince peopleusewordsandsentencesinallsortsofways,thebestonecandois describe howanotherusesawordonanoccasion.Youcannot,hesaid, constructa theory ofmeaningifyouthinkthatthemeaningsofwordsare foundinthewaystheyareused.InthisrespectheandChomskyagree:therejust isnotenoughuniformityinthewayspeopleuseexpressionstosupporttheory. Soifyouthinkofmeaningsintermsoftheiruses,yougetnoscience.Sofar, Chomskyandheagree.ButforChomsky,thatjustshowsthatyouarelookingin thewrongplaceforatheoryofmeaning;lookatwhatbiologyprovidesyouin thehead.Wittgenstein ’swarningwasgenerallyignored;philosopherssuchas LewisandSellarsandinnumerableotherssimplyassumed(perhapswiththe practicesofmathandscienceinmind)thattheremustbeagreatdealmore uniformityofusethanappears,andpostulatedconventionsanduniformpracticesthatjustdonotexist.Aswehaveseenwithreference,thatisnotagood strategy.InternalistssuchasChomskysuggestlookingatthematterfromthe otherdirection;donotthinkofmeaningintermsofuse,butthinkinsteadof internallysourcedandtheoreticallyspecifiablefeaturesofwordsandsentences. Thesestillprovide ‘tools’.Havingthenaturestheydoallowsthemtobeusedin thewaysthattheyobviouslyare.Inotherwords,explainnothowpeopleuse words,buthowtheircreativeuseofwordsispossible.Itispossible,theRR theoristholds,onlybecausewithlanguage,internalsystemsprovideconfigured andrich ‘perspectives’ fortheuseofpeople.Theseperspectiveshavetheshapes andcharacterstheydobecauseofthecontributionsofsyntax(whichputswords together)andthesemanticfeaturesofthewordsthatcomposethem,wherethese semanticfeaturesaretakenfrominternalresources.Andtheseshapesand charactershelpshapeand ‘givemeaningto’ experienceandthought.
FocusingnowmorecloselyonRRstrategyforinvestigatinglanguage, proceedbyassumingthatthelanguagefacultyisamodularinternalsystem thatoperatesinadeterminateway,wherethatwayisfixedasaresultof principlesoforganicgrowth;thesystemis,then,innateandbecauseitoperates whenmatureindeterminatewaysanddevelopsinaccordancewithinfixed constraintsongrowth,itispossibletoconstructatheoryofit.Thattheoryaims todescribeandexplaintheinternaloperationsofthefaculty,assumingtheseto beproceduresfortakingthevariouskindsof ‘information’ availableinlexical itemsstoredinaperson’smentaldictionaryand,byaprocedurethatjoins lexicalitemsandtheinformationthattheycontaintooneanother,yieldsa sententialexpression.Asententialexpressionconsistsofasoundandameaning;itcanbethoughtofasapairedformofinformation:soundinformationand
meaninginformation.Eachkindofinformationappearsinformsthatcanbe ‘read’ (‘understood’)bytherelevantkindsofotherinternalsystemsattwo interfaces,thesoundinterface(PHON(eticinterface))andthemeaninginterface (SEM(anticinterface)).Thesoundinformationisusedbyperceptualand articulatorysystemstoyieldcompressionsvaryinginfrequencyandamplitude inair(or – aswiththought – itstays ‘inside’)orto ‘decode’ thesesignals receivedattheear;themeaninginformationisusedby “conceptualandintentional” systemstodo – afterthecontributionsoftheseotherssystems – withthis informationwhatapersonwants,perhapstoasksomeoneaquestionortotryto figureoutwhyherwatchdoesn’twork.Aninternalisttheoryof linguistic meaningfocusesnotonhowthesemanticinformationinlexicalitemsis ‘read’ byothersystemsinthehead,butratheron(1)describingintheoretical termstherelevantkindsofinformationavailableinlexicalitemsand(2)saying howitgotthere(answeringPlato’sProblem – theacquisitionproblem – for lexicalsoundsandmeanings).Forchildrenacquire ‘words’ veryquickly indeed,offeringindoingsolargenumbersofpoverty-of-the-stimulusobservations.Giventherateatwhichtheyareacquired – aboutoneawakinghour betweenagetwoandeight – thespecificcharactersoflexical ‘sound’ information,andtheintricacyoflexicalmeanings(intricatebeyondanythingdescribed inthemostdetaileddictionary),wemustassumesomekindofinternalmechanisms.Perhapsthemostchallengingoneisthemechanismthatassembles lexical ‘meanings’– i.e.,relevantkindsofsemanticinformation.Whenthat mechanismisspecified,itwillconstituteanimportantpartofananswerto Plato’sProblemforlexicalacquisition.Then,(3)wemustsayhowthesemantic informationinlexicalitemsiscomposedbysyntacticprinciples(syntaxyieldsa theoryofcompositionality),and(4)indicatewhatisprovidedatthesemantic interfacetoothersystems,includingmodifications,ifany,duetocomposition. Allofthismustbedone,furthermore,whilemeetingstandardconditionson naturalisticscientificresearch,outlinedbelow.Accomplishingallofthesetasks isverydifficult,buttherehasbeensomeprogress.Atleastitisclearernowthan itwasfiftyyearsagohowsyntaxcontributes,andhowtoconstruetheoverall contributionsoflanguageattherelevantinterfacestotherestofthemind.There hasalsobeenatleastsomeprogressmadeininvestigatingsemanticfeatures. Butalotremainstobedone.Nothinglookstobeimpossible,though,asdoesa naturalistictheoryofreferenceorasociallydeterminateaccountofuse.
Informally,onecanthinkofthesemanticinformationprovidedinalexical itemasalexicalconcept,andofthesemanticinformationcombinedatthe semanticinterfaceasasententialconcept.Thiscapturestheideathatconcepts aretheinternaltoolswehumansuseto – asabove – asksomeoneaquestion orspeculateaboutwhyone’swatchdoesnotwork(andwhattodoabout it).Internalconceptualtoolsarewhatweusetocategorize,tothink,tospeculate,andsoon.Aninternalistlinguistictheoryofmeaningdetailsthe
contributionsofthelanguagefacultytoaperson’sconceptualtools.That contribution,itseems,issubstantial.Andthereislittledoubtthatsyntaxallows ustobeasflexibleinourcognitiveexercisesasweseemtobe.Agreatdeal morecouldbesaid,butthiswillsufficeforthisintroduction.Aninternalist theoryoflinguisticmeaningcangoalongwaytowardsmakingsenseofwhy wethinkofourwordsandsentencesas “meaningful.” Theyaremeaningful becausetheyofferusthetoolstodowhatwedo.
Butifsemanticrelationshipstotheoutsideworlddonotfigureininternalist theoriesoflanguageandtheirmeanings,surelytheoutsideworld – itsthingsand events,itssocialinstitutionsandthepracticesofpeople – mustfigureintheRR pictureofthemindanditsstudyin some way?Theworldoutsidetheheaddoes figure,butinwaysthatdonotchallengetheRRassumptionsandtheinternalist researchstrategytheybaseonthem.First,asmentionedbefore,theaudibleand visible ‘externalized’ wayspeopleuseconceptstothinkandact
tocategorize, describe,assess,complain,convince,etc. – providesomeevidenceforand againsttheoriesoftheinternalmechanisms.But,theypointout,atheorydoes notconsistofitsevidence.Operationalismandinstrumentalism,epistemological versionsofbehaviorism,andsoon,arenottheories;theyareempiricist-motivated (andverypoor)methodologicalrecommendations.Amentaltheoryisatheoryof aninternalsystem – itsalgorithmsorrulesofoperation,itsinputsandoutputs,and themeansbywhichitcomestohavetheserules,etc.
Second,theoutsideworldisnodoubtthesourceofsomeoftheinputordata –the ‘experience’– neededforthelanguagesystem/organtobegintodevelop, andcontinuedevelopmenttoreachasteadystate(vocabularyadditionsand subtractionsaside).Justasvisiondoesnotdevelopnormallyifitisnotgiventhe ‘right’ kindsofinputduringcriticalstagesofdevelopment,soconceptsand theircombinatorymechanismsdonotdevelopinthechildunlesstheyreceiveat leastsomeexternalinputofthe ‘right’ kind.Inthecaseoflanguage,thechild doesnotdeveloplanguagenormallyunlesshisorhermindisprovidedwithat leastsomelinguisticinputthathastheformofonenaturallanguageoranother. Theinputrequiredis,however,remarkablysmall – notinabsolutescale,but relativetothespecificityofwhatisacquired – andtheinputmaybecorrupt. Developmentisrobust;itseemstobe “channeled.” And,itisimportantto recognize,withneitherconceptsnorcombinatoryprinciplesdoestheshapeor characterofwhatisactivatedresultfromtheinput.Perhapstheneedfora conceptarisesasaresultofpromptingorstimulationfromtheoutsideworld, buttheshapeandcharacterofaconceptorcombinatorysystemisdetermined bytheminditself,nottheworldorcommunity.Indeed,themind’sdevelopmentaloperations ‘say’ whatkindsofdataareneeded.Generally,internal developmentalmechanisms – nottheexternalworldoraperson’scommunity –specifythekindsofcausesorexternalprompts/patternsrequiredforactivation andmaturation.14
Givensuchbeliefsaboutthemindanditsstudy,theRRadvocateislikelyto maintain – assuggestedabove – thattheviewoftheworldthatonegetsthrough thelensofourinnateconceptsandcombinatoryprinciplesowesmoretothe charactersofourconceptsandcombinatoryprinciplesthanitdoestohowthe worldmightbe ‘initself ’.Toputalabelonthiskindofview,Iwillcallit(aform of)constructivism:ourminds ‘make’ theworld,ratherthantheotherway around.ImentionitheretounderscorethedifferencebetweentheRRcamp andtheempiricistone.Forempiricistsbelievethatinthecaseofmostconcepts (perhapsnot ‘pure’ sensoryones)andcombinatoryprinciples,theworldshapes themind.
TheRRs’ opponents,theempiricists,holdthatmostoftheconcepts expressedinnaturallanguages(DOG,HOUSE,WASH…)andthecombinatory principlesthatplacetheminunderstandablesentencesarenotinnate,butrather learned.Perhapstheyareassembliesofperceptual ‘features’ (Locke,Prinz (2002)),specifickindsofrolesconstitutedbythe ‘moves’ (inferences)people engageinwhenexercisingsocialpractices, ‘connectionweights’ inneuralnets, andsoon.Howeverconstrued,conceptsand “rules” arelearnedbyengaging somekindofgeneralizedlearningprocedure(hypothesisformationandtesting, association,trainingprocedures,behavioristconditioning…)thatafterrepetitionand ‘feedback’ intheformofpositiveandnegative ‘evidence’ comesto convergeonwhatsociety,theexperimenter, ‘theworld’,orsomeotherassumed judgeofmeetingacriterionaccepts.Howeverconstrued,theempiricistbelieves thattheenvironment,includingsociety,makesandshapesconceptsandthe principles(“rules”)oftheircombinationthroughsomesortofgeneralized learningprocedure,aprocedurethatusuallyinvolvesnotjust(alotof)sensory orotherlow-levelinputand/ordata,butatrial-and-errorprocedureofsomesort, whereerroriscorrectedbysomekindof ‘negativeevidence’ [“that’snotright;” pain/punishmentonabehavioristversionofempiricism],perhapsprovidedby parentsorinstructors,perhapseven(itisclaimed)bythelackofdata.Because theempiricistholdsthatsuchproceduresaresufficienttolearnthethousandsof conceptsthatfour-year-oldshaveavailable, and tolearnthecombinatory principlesandstructuralconstraintsofalocallanguage,theempiricistsmust assumethatmuchofthechild’searlylifeanduseoflanguageisdevotedto focuseddata-gatheringandtrainingsessionsthatconsistingettingthechildto conformtothe “speechhabits,”‘proper ’ (epistemicallyappropriate,etc.)applications,orusesofconceptsthechild’strainerswantittoexhibit.Forotherwise one wouldfind – contrarytofact – childrenacquiringlanguageandmany thousandsofconceptsatverydifferenttimes(dependingontraining,the resourcesoftrainers,nativeintelligence,interestanddevotiontoduty ), goingthroughverydifferentstagesofdevelopment,andsoon.Howeverthe storyistold,empiricistsare antinativist and externalist:theymaintainthat conceptssuchasthosementionedaboveandthecombinatoryprinciplesor
“rules” thatlanguagesoffertoproducecomplexesofconcepts(whatsentences express)are ‘learned’ byexperienceofthingsandevents ‘outside’.Becauseof this,theymustholdthatstudyofthecontentsofthemindcannotbedivorced fromtheenvironment(s)inwhichthemindissituated,whereanenvironment includesforlanguagecruciallythelinguisticbehaviors(presumablyrespecting thelinguisticpractices)ofa “community.” Inatleastthissense,theempiricistis committedtoanexternalistprogram.
Thereisamassiveamountofquestion-begginghere.What,exactly,isa generalizedlearningprocedure?Howarehypothesesconcerningtheapplicationsofconceptsformedifonedoesnothavetheconceptsinthefirstplace? Whatcountsassimilar,whatdissimilar?How,specifically,doesanalogywork intryingtoextendacquiredknowledgetounacquired?Wheredoeslinguistic structurecomefrom?WhatcountsassufficienttoshowthatchildNhas acquiredanyorallconcepts{c1…cn},orrules{R1…Rn}?Whereareallthe trainersthelearningprocedurerequires,andevenassumingthattherearesome, wheredotheyfindthetime?Whydochildrenataparticularstageofdevelopmentcompletelyignore – virtuallynothear – theirparents’ admonitionstosay “went,” not “goed?” Whataboutthefactthatourconceptshavelittletodowith whatisreallyoutthere?Whydoesonlythehumanmindseemtoacquire languageinthenormalway?Whydointelligentapesfailtoacquirealanguage likeanyhumansignlanguage,nomatterhowmuchtheyaretrained?Wheredo thehierarchiesfoundinnaturallanguageclausalconstructionscomefrom? Whydolanguagesappearonlyincertainforms?Howdoesachildmanageto developthenotionofaphoneme?Andsoon,andon.Chomskyhasnotedmany ofthesegaps,andothers,inempiricistviewssinceatleast1957with Syntactic Structures (andbeforethat, LogicalStructureofLinguisticTheory,althoughit wasnotpublisheduntillater)and1959withhisreviewofSkinner ’s Verbal Behavior.Someofhiscriticismsreappearin CL inwhathehastosayabout “modernlinguistics.” Afewarehighlightedbelow.
Thereisanotherproblem,relatedbutlessnoted.Onthefaceofit,the empiricisthasadifficulttimeaccountingforearlycreativity.Giventheenormousamountthechildmustbeassumedtohaveacquiredinordertodisplay whatisinfactobservedtobeintheirrepertoires(thousandsofconcepts,sounds, andthecombinatoryprinciplesofalanguage)inordertogainthekindof “mastery” ofconceptsandthewaysthattheycanbecombinedneededfor everydaylinguisticcreativity,andgiventheenormousamountoftimesurely neededtogetwhatacommunityinsistsis “therightwaytospeak” (andclassify, describe,explain,speculate…)outofa “generalizedlearningmechanism,” itis hardtoexplainwhatseemstobeeffortlesscreativeuseoflanguageontheparts of all normalthree-and-a-halforfour-year-oldchildren.Anditbecomes extremelydifficulttounderstandhowallchildrenmanagetobecreativein the ‘ordinary’ wayataboutthesametime;surelyintensityoftraining,
differencesinnativeintelligence,parentalresources,anddifferentvarietiesof experienceswouldallleadtodifferentratesatwhichachildlearnsto “master” whatisneeded.Ordinarycreativityseemstoposeaseriouschallengetothe empiricistcamp’sassumptionsaboutthenatureofmindandtheirresearch strategy.
Asmyremarkssofaremphasize,thechoicebetweenRRandempiricist assumptionsaboutthemindandscientificresearchstrategiesturnonarather simplesetofobservationsthatanyonecanmakeconcerninglinguisticcreativity (particularlywithchildren),andanothersetconcerningrateandtimingof learning/activation,andthekind(s)andamountofinputreceived.RR approachestomindandlanguageseemtorespectthesesimpleobservations, andtaketheirtasktobeexplainingthemor – wherescientificexplanationis impossible,aswithcreativity – toseektoshowhowwhatisinthemindmakes thephenomenonpossible.Totheextentthatempiricistapproachesdonottake theseobservationsandthetaskstheysetforthescienceofmindseriously,they seemtobeignoringfactsthatstarethemintheface.Empiriciststrategyforthe studyofminddoesnotseemtoberesponsibletoempiricalfacts.
II.2TheChomskydifference:naturalizinglanguageandmind
Whilehisbasicassumptionsaboutthemindandthestrategytousetostudyit aremuchthesameasthoseofhisRRpredecessors,andwhilehispredecessors –manyofthemthescientistsoftheirdays – wouldverylikelyhavewelcomedhis effortsandtheirresult,Chomsky,unliketheothers,hastriedthroughouthis careertoturnthescientificstudyoflanguageinthemind(andofmindin general)intoanaturalscience,ultimatelyabranchofbiologyandothernatural sciences.(Afterhisearliestefforts,ofcourse,hehadthehelpofmanyothers workingwithinhisassumptionsabouthowtoproceed.)Thisprojectisimplicit inRRassumptionsaboutthemind.Ifmuchofthementalmachineryneededto developconceptsandtheircombinatoryprinciplesisinnateandoneisgoingto trytoexplainhowitcomestobeinthemindatbirth,itwon’tdotosaythatGod putitthere(Descartes)ortoconstructmythsofreincarnation(Plato).Theonly courseopentousistolooktobiologyandthoseothernaturalsciencesthatcan saywhataninfanthumanbeginswithatbirthandhowwhats/heisbornwith develops.Andtakingthattackalsomakesitpossibletoatleastbegintospeakto thequestionofhowhumanbeingscametohaveapparentlyuniquemachineryin thefirstplace – toaddresstheissueofevolution.
Placingthestudyofconceptsandtheircombinatoryprinciplesinnatural sciencealsoimposesspecificandwell-knownmethodologicalconstraintson thetaskofthescientistofthemind.Adoptingnativismamountstobeginninga researcheffortthatassumesthatconceptsandlanguagearesomehowimplicitis somekindofnatural ‘mechanism’ ofthehumanbody-mind,under(partial)
controlofthegenomeandthecourseofdevelopmentitcontrols.Constructinga theoryofconceptsandlanguageis,then,constructingatheoryoftherelevant mechanisms,andofhowtheydevelop.Tryingtodothatplacesone’sefforts firmlywithinthenaturalsciences,anddemandsthatonemeetthecriteriaof successfulsciencethatothernaturalsciencesaimtomeet.Ofcourse,one’s theoriesaretheoriesofaninternalsystemandhowitdevelops;but ‘going inside’ makesnodifference.One’stheorymustmeetthestandardsofsuccess thatanynaturalsciencemustmeet,regardlessofsubjectmatter:thelinguistor lexicalconcept-theoristmustconstructtheoriesthatsatisfythesamedemands thatthetheoriesofthephysicistaimtomeet.Would-beinternalistmental scientistsmustproducetheoriesthatare descriptivelyadequate inthatthey fullyandaccuratelydescribealltheelementsandpropertiesofthe ‘things’ the theorydealswith – inthecaseoflanguage,words,sentences,andhowwords cometobeputtogethertomakesentences.Further,atheorymustbe explana torilyadequate;inthecaseoflanguage,itmustsaywhyachildhasjustthese elementsinhisorherlanguageataspecificstageofdevelopment,and – once onehasexplainedthis – onemustdealwithfurtherexplanatoryissues,suchas howlanguagecametobeintroducedinthespecies.Further,atheorymust provideanexplicit,formalspecificationofeverythingthatthetheoryaimsto describeandexplain – itmustbe formalized with,ifrelevant,quantification accordingtoanappropriatemeasurementschemeoftheelementsandtheir ‘powers’ orfeatures.Inaconnectedvein,anaturalizedformofmentalscience mustaimtowards simplicity:thetheorymustofferacompactyetfullaccountof itssubjectmatter.Withlanguage,forexample,thiscanamounttoconstructing theorieswithasfewprinciplesorrulesofcombinationofwordsaspossible,and statingtheserulesaseconomicallyaspossible.Further,atheorymustaim towards objectivity.Asthehistoryofscienceindicates,thatrequiresabandoning theanthropocentricallyorientedconceptsofcommonsense(theonesthatwe arebornwith,includingthecommon-senseconceptLANGUAGE)andinventingconceptsthatcanadequatelydescribeandexplain – forlanguage – any languagewhatsoever,plussayhowandwhyaparticularlanguagedevelopedin aspecificindividual.Andatheorymustaimtowards accommodating the mentalscienceinquestiontoanotherscience – inthecaseoflanguage,certainly biology – amongothernaturalsciences.Finally,aconditionthatreflectseverythingsaidsofar.Theinternalisttheorymust makeprogress – successiveorat leasttemporallynearbytheoriesoftherelevantdomain(theoriesoflanguage, vision,facialconfiguration…)thatshowimprovementinoneormoreofthe dimensionsjustmentioned.Progressismeasuredbythestandardsofthe methodologyitself.
Chomsky’seffortsatconstructingsciencesoflanguagehaveimprovedinall thesedimensions – thatis,histheorieshavemadeconsiderableprogress.Ipoint tosomeofthesignsofprogresslater.Thattherehasbeenprogresssuggeststhat
hiseffortsandthoseofthemanywho,likehim,adoptanRRviewofthemind, areontherighttrackandsomehowseemtocutnature – inhiscase,thelanguage organ – attherightjoints.Theprogressinturnmakesittemptingtoholdthatthe targetofthesetheories – thelanguageorgan – is ‘real’.Anditmakesittempting toadoptapolicyofsayingthatwhatthebesttheory(bytherelevantstandards) saysabouttheorganistrue,andthatitdoesdescribeandexplain ‘howthings are’.Thetemptationshouldnotberesisted.Werelyonbothscienceand commonsensetogetanunderstandingofourselvesandoftheworld.Each approachhasitsmerits.Common-senseconceptsnotoriouslyfailinscience, whilescientificconceptssuchasMU-MESONarehopelessinsolvingthekinds ofpracticalproblemsthatcommonsensedealswith.Eachkindofcapacityto solveproblemshasitsplace,andneitherdealswithallproblems.Wheretheaim isobjectivityandtheprecisionofexplicitandformalstatement,however,there isnochoice:themethodsofthenaturalsciencesyieldtheonlyanswersweare likelytobeabletoget.Andifatatimeatheoryisthebestavailable,ittellsus whatlanguage – orvision,etc. – is.
CertainlyChomsky’seffortshaveprogressedwellbeyondthoseofhisRR predecessors.Wherehispredecessorscountenancedunexplainedpowersand hadtobesatisfiedwithpointinginadirection,orwheretheyhadlittleideaof howtomakesenseofinnateness,Chomskycannotanddoesnot.Forexample, DescartesandthePort-Royalgrammariansinthesixteenthandseventeenth centuriesinthefinalanalysisattributedthearticulationandproductivityof languagenottoanorganicinternalsystem,buttoanunexplainedpowercalled “reason” or “thought.” Butthought,forexample,isbestunderstoodasconstituted,inpart,bylanguage:weuselanguagetothink.Soweshouldtrytosay whatlanguageiswithoutappealingtothought.Chomskyclearlydistinguishes languageanditsstudyfromthewaysinwhichlanguagecanbeused,including thoughtandreasoning.Todothat,heconstructsatheoryofabiophysically basedorganicsystem.Andtomakesenseofinnateness,hispredecessorsmight havepointedtodivinity(it’sagiftfromGod)ortheymightholdthatithas somethingtodowithNature,butsaylittleabouthowGodmanagedtoget conceptsandcombinatoryprinciplesintoourheads.HisRRpredecessorsdid, though,makeatleastsomeprogress.Incontrast,empiricistsseemtohaveadded littletoLocke.And,likeLocke’sefforts,theirsgenerallyfailtomeetthe conditionsonadequacyofa naturalistic theory.
Manyempiricistswouldnotbebotheredtobetoldthattheireffortsarefar removedfromnaturalistictheory-construction.Thatisbecausemanycurrent empiricists,atleast,thinkoflanguageasacomplexformofsocialpractice learnedfromothers,withpracticesvaryingfromlanguagetolanguageand environmenttoenvironment,andofconceptslearnedasapartofacquiring linguisticandotherpractices.Theymight,asWittgensteindid,thinkofconceptsasepistemicallygoverned ‘roles’ inlinguisticpracticesor ‘language
games’.Forthem,languagesaresocialinstitutionsandartifacts,notstatesofa biologicallybasedmental ‘organ’ . 15 Still,theywanttocallthemselvesscientists.Tomakethisseemcredible,theymightarguethattheyareofferingafar simpler ‘hypothesis’ thanChomsky’s(once)ratherdaunting-lookingtheoriesof UniversalGrammar(UG).Theirhypothesisisthatlanguageandconceptsare learnedbysomekindofgeneralizedlearningprocedure.Theytaketheir generalizedlearningproceduretobethesimplesthypothesisconcerningconceptsandcombinatoryprinciplesbecauseitistheleastcommittedtothe existenceofanykindofdomain-specificmentalmachinery – anythingdevoted tolanguage,forexample,forthatwouldmakelanguageinnate.Soiftheywant tospeaktowhatisintheheadatall,theystipulatethatwhatisintheheadmust besomekindofneuralnetthathasthepropertythatitcanbemodifiedby ‘experience’ andcorrectionofitsoutputs.Perhapsthereareinnateorganic mentalmodulesforvision,audition,andthelike.Andnodoubtthesehave somequiteremarkableinnateproperties.Buttheremustbealotofthemind/ brainthatisplasticandcanbemodifiedbyexperienceandtrainingprocedures, forthatiswherelanguagemustgo.Toshowthemeritsoftheir ‘hypothesis’ (whichisrarely,ifever,explicitlystated),theymightintroducecomputer modelsintheformof ‘neuralnets’ ofwhattheybelievearetheplasticregions ofthemind/brainandsubjectthesecomputermodelstoinputthatissupposedto simulatetheirviewsofhumanlinguisticexperienceandthedataoflanguage learning.Theyconsidertheireffortssuccessfulifthecomputermodel ‘learns’ to performthe ‘task’ setittothesatisfactionofthe ‘experimenter ’ .
Thisstorymakeslanguageacquisitionamiracle:standardnaturalistic explanationsroutinelyemployedelsewherearesimplyrejected,andfor humanlanguagealone.Theempiricistishappyenoughtosaythatachild’s petsongbirdacquiresthesongitdoesfornaturalisticreasons;inputmaybe requiredfordevelopmenttoproceed,butdevelopment,andthesongpatterns thatdevelop,arelargelyundergeneticcontrol.Buthumanlanguage?It must be explainedsomeotherway.Refusingtoseelanguagesandconceptsasnatural objectsforwhichoneneedsnaturalistictheoriesisaformofwhatChomsky calls “methodologicaldualism”:whenitcomestocrucialfeaturesofthemind, theempiricistsabandonnotjustinternalismandnativism,butthemethodsof thenaturalsciences.Theylodgelanguagesoutsidetheheadandmight (e.g.Prinz 2002)treattheeverydayconceptswehave,suchasCURRY,as builtoutofprimitivementalitemslikesensoryfeatures,butbuiltunder exogenouscontrol,notthecontrolofinternaldevelopmentalmachinerythat demandsmanynon-sensoryfeatures(suchasABSTRACT).Ortheymight (Sellarsandmanyothers)speakinsteadofthe ‘token’“curry” intermsof learnedgamesandsocialpractices,construingitsmeaningintermsofits contributionstosociallysanctionedtruthorcorrectnessconditionsforthe ‘token’“curry.” Howeverglossed,thecharacterandcertainlythe(supposed)
referentialpowersoftheconceptCURRYdependonhowacommunityof speakersusetheword curry – thecontexts,whatgetcountedastherightor correctortrueorappropriateuses,andsoon.Theonlyinternalandpresumably ‘natural’ machineryrelieduponaresomesensorycapacitiesandsomeformof generalizedlearningprocedure,cashedoutperhapsinaversionofassociationismand/orbehaviorismembeddedinaplasticneuralnet.Theproceduremight consistinsomekindofstatisticalsamplingprocedurelodgedinwhatstartsout asanundifferentiatedneuralnet,amodernempiricist’sversionofLocke’sblank slate.
Notallempiricistsaremethodologicaldualists.Butthatisnotbecausethey thinkthatlanguagesandconceptsaresomehownaturalobjects.Some,suchas HerderandFoucault(whocountasempiricistsbyvirtueofbeing anti-nativistandexternalist16)wereunabletounderstandwhatanaturalistic scientificmethodologyis,orappreciatetheverydifferentobjectiveviewof theworldityieldswhencomparedtotheanthropocentricviewgotfrom commonsense. 17 Onlythoseempiricistswithanappreciationofthedifferences betweenscienceandcommonsense(“folkphysics,”“folkpsychology,” andthe like)areapttoadoptmethodologicaldualism.Thetwentieth-centuryAmerican philosopherSellarsandhisintellectualprogeny(e.g.Churchland,Brandom), andthephilosophersQuineandPutnamandtheirs,mighthaveanythingfroma reasonablycleartoaverygood(e.g.Putnam)ideaofwhatnaturalisticscientific methodologyis,18 buttheyclearlyrefusetoholdthatlanguageandconcepts couldbeinvestigatedusingthismethodology,ratherthansomeversionofan empiricistone.Atthisstageinthedevelopmentofthescienceofmind,their refusalistotheRRstrategistpuzzling,atbest.Withconcepts,perhapsthereis someroomforsympathywiththeempiriciststrategist.Whilethereisnodoubt thatthepovertyandcreativityobservationsmakeaRRstrategyreasonable, naturalistictheoriesofconceptsarestillintheirinfancy.Amongotherthings, eventhosesympathetictothestrategy,suchasFodor(1982, 1998),aredrawnto views(suchasthatconceptsbythemselvesdenote)that – asnotedbefore –makeastraightforwardinternalistandnativistapproachimpossible.But,RR strategistssuchasChomskymustfeel,surelyitisunreasonabletoadopta non-naturalisticmethodologyforthestudyoflanguage.Herewefindarticulatedtheories,andveryconsiderableprogressonallfronts.Quine’sviewthatin thestudyoflanguage, “behaviorismismandatory” appearstobenothingbut dogmaticrefusaltofacetheobviousfacts.
Butisnotwhatempiricistsproposenaturalisticinquiryafterall,especiallyin theformthatsomeoftheconnectionistshavedevised,wherewefindingenious effortstoget “simplerecurrentnetworks ”19 andtheliketo(say) “recognize” classesofthings,oraftermassivetrainingproduceoutputsthatseemtoindicate thatthenetwork’sconnectionweightshavesomehowembodieda ‘rule’– that is,yieldedapairingofinputstooutputsthatsuittheexperimenter ’scriteriafor
Another random document with no related content on Scribd:
The Project Gutenberg eBook of Mabel Wynn
This ebook is for the use of anyone anywhere in the United States and most other parts of the world at no cost and with almost no restrictions whatsoever. You may copy it, give it away or re-use it under the terms of the Project Gutenberg License included with this ebook or online at www.gutenberg.org. If you are not located in the United States, you will have to check the laws of the country where you are located before using this eBook.
Title: Mabel Wynn or, "Those boys"
Author: Faye Huntington
Release date: June 6, 2024 [eBook #73786]
Language: English
Original publication: London: George Routledge and Sons, Limited, 1889
*** START OF THE PROJECT GUTENBERG EBOOK MABEL WYNN ***
Transcriber's note: Unusual and inconsistent spelling is as printed.
MABEL WYNN.
CHAPTER I.
THE CONSECRATION.
"My all is on the altar."
WESTVILLE had a new sensation. Deacon Griffin, earnest Christian man as he was, ever ready to use tongue, hand and money for the cause of Christ, shook his head, and said he didn't know about these things. There was so much danger of falling into the ways of the world; the Church must be wary of innovations.
Then good Mrs. Culver had opposed the plan, and determined to stand aloof. What was the use in making a stir! They had gone on for years without such an affair, and for her part, she didn't believe in new-fangled notions.
However, in spite of strong opposition, a few brave hearts persevered and carried the matter forward, and Westville had its first "Sunday-School Institute." These few feeling the want of more labourers, sensible of their own want of knowledge as to the best mode of carrying forward the work, invited one whose life is devoted to the cause to give a course of instruction upon topics connected with the work, hoping to be themselves better fitted for their duties, and trusting that others might be led to enter the same field of labour.
Many, like Deacon Griffin, looked upon Sunday-schools as especially for children; some, like Mrs. Culver, clung to the old ways. As a child she had recited her seven verses Sabbath after Sabbath, without question or comment on the part of her teacher; brought her penny for the mission
fund, and carried home a library book, or perhaps a copy of the "Dayspring" or "Messenger," and had about as much idea as a kitten what all or any part of it meant; and she taught her class in very much the same way. True, in maturer years, the precepts and promises stored in her mind had an influence upon her heart and life. As if Christ's precious words were not for little children to carry in their hearts as well as in their memories!
In those three days, much work was done for the Master. With instruction, discussion and earnest talk were mingled many prayers and much praise. The interest growing, the audience increasing in numbers, until at the last session, the great church was filled to its utmost capacity, and many hearts were throbbing with emotion. New thoughts had found their way into the minds of gray-headed old men, and to some of them it now seemed a grand thing this idea of a whole congregation coming together for the study of God's word. And old men and women almost through with church-going, the middle-aged, busy fathers and mothers, and the young, enthusiastic and eager, pledged themselves to the work and joined in the consecrating hymn. It was grand and solemn, that chorus of hundreds of voices. Many of feeble faith were encouraged, slothful souls were quickened, while to some, it was the hour of a new and entire consecration to Christ. As the song died away, the voice of the leader, smooth and clear, filled the house.
"Father in heaven," he prayed, "we thank thee for this evidence that thou hast been with us, directing and guiding us by thy Spirit. We know not what vast results may grew out of our efforts to advance thy cause, and we are content to leave the knowledge with thee. Thou hast heard our solemn consecration. Wilt thou enable us to keep the vow? May we all feel its solemnity and importance, and may each of us be faithful to our promise. Wilt thou bless these thy children, who work together here? May they work hand to hand in Christian love and forbearance, illustrating in their lives the truths they teach. May they be patient and persevering, untiring and full of the spirit of devotion. May they have all needed wisdom and every Christian grace, and may thy blessing so crown their efforts that multitudes shall throng the streets of the beautiful city led into the upward way by these thy servants. And though we no more meet together here, at last, dear Christ, we shall be together and always with thee. Amen."
CHAPTER II.
WORK.
"O thou whose call our hearts has stirr'd!
To do thy will we come."
"I'm here," began Mabel Wynn, in her bright way, presenting herself promptly the next Sabbath at the superintendent's desk. "You know I am a fresh recruit. I shall make a poor show, but I'm ready for drill. Where shall I go?"
Mr. Clarke looked at her earnestly for a moment. He knew her only as the daughter of the richest man in Westville, a man who was prompt to serve Christ with his money, but whose time, strength and influence were given to the world. He knew that Mabel came to church regularly, dressed handsomely, and was a sort of a leader among the young ladies. He knew, too, that she was a professed friend of Christ, but he did not know the earnest and unselfish purpose of her heart to work for the Master. Mrs. Wynn called her daughter odd. "One of Mabel's freaks," or "just one of her odd whims," she would explain to her friends.
If to live for a purpose, if to do one's work with earnestness, if to seek out opportunities for doing good, if to carry the spirit of Christ into every day's living—in short, if to be a sincere, consistent Christian, if this is to be odd, then Mabel was odd. As a child she had been a Sunday-school scholar, but as she grew up, she had yielded to her mother's persuasions and discontinued her attendance.
"It was such an inconvenience," explained Mrs. Wynn. "So tiresome to remain after service, and such a trouble to send back the carriage. Thomas did not like it at all."
But now Mabel determined that circumstances must yield; fortunately Thomas had departed, and a very obliging Michael reigned over the stables.
Thus she stood ready and willing, while Mr. Clarke deliberated a moment. His first impulse was to send her to Dr. Eaton's class of young ladies, and this was just what she expected—indeed she had at first intended to take her place there at once, but some of her mother's teachings in regard to observing the proprieties under all circumstances, impelled her to report herself at headquarters. But they were short of
teachers; so many of those who had come in for the first time, had asked for a scholar's place, that old classes were overflowing and new ones had to be formed.
While Mr. Clarke deliberated, a scene of confusion across the room attracted his attention. A thought struck him. "Who knows but she may be able to do something with those boys? Anyway they must have a teacher to-day." And with a smile for the eager face before him, he nodded in the direction of the five unruly boys.
"Mr. Clarke," began Mabel, then stopped suddenly.
"Well?"
"Nothing. I forget that it is my business to obey a superior officer. I'll go."
As the opening exercises went on, Mabel had time to still her throbbing heart and lift her soul to God for help. She had not thought of becoming a teacher at once, and what to say or do first, was a puzzling question. She tried to remember what they did in those days so long ago when she went to Sunday-school. Then she tried to recall some of the instructions of the past week, but she could think of nothing that would exactly fit the case. Dissimulation was no part of her nature, and when the time came, she turned toward the class; but what a class it was!
Five bright-looking boys, combed and brushed and neatly dressed, but the element of quietness seemed entirely wanting. There they were—five bobbing heads, five pairs of elbows pushing this way and that, five pairs of shuffling feet, five pairs of eyes and ears, open and eager to take in any sight or sound that might be turned into fun. Evidently they understood all the boy-tricks in the catalogue. Now and then a head would be thrown back with a peculiar spasmodic jerk, and when the cause was looked for, another boy's arm would be found quietly resting upon the back of the seat, thrown lovingly around his neighbour's. Of course the owner of the arm had nothing to do with it. Suddenly the small boy at the end of the seat found himself on the floor, a concerted pushing along the line having brought about the catastrophe. There were pinching and nudging and treading upon toes, and winking and laughing, and much whispering. Altogether a scene of disorder. And these five active, eager minds were to be fed—those five souls to be led to Christ! What a work for some earnest disciple! No wonder that Mabel's heart almost quailed. But she said, simply and frankly,—
