https://ebookmass.com/product/aristotles-empiricism-marc-
Instant digital products (PDF, ePub, MOBI) ready for you
Download now and discover formats that fit your needs...
Logical Empiricism as Scientific Philosophy Alan Richardson
https://ebookmass.com/product/logical-empiricism-as-scientificphilosophy-alan-richardson/
ebookmass.com
Guardrails: Empowering Human Decisions in the Age of AI Gasser
https://ebookmass.com/product/guardrails-empowering-human-decisionsin-the-age-of-ai-gasser/
ebookmass.com
Professional C++ 5th Edition Marc Gregoire
https://ebookmass.com/product/professional-c-5th-edition-marcgregoire/
ebookmass.com
Self-Made Boss Jackie Reses
https://ebookmass.com/product/self-made-boss-jackie-reses/
ebookmass.com
978-0133507331 Quantitative Analysis for Management (12th Edition) https://ebookmass.com/product/978-0133507331-quantitative-analysisfor-management-12th-edition/
ebookmass.com
Intellectual Disability and Social Policies of Inclusion: Invading Consciousness without Permeability 1st ed.
Edition David P. Treanor
https://ebookmass.com/product/intellectual-disability-and-socialpolicies-of-inclusion-invading-consciousness-without-permeability-1sted-edition-david-p-treanor/
ebookmass.com
Experiments Manual To Accompany Digital Electronics: Principles and Applications, 9th Edition Tokheim
https://ebookmass.com/product/experiments-manual-to-accompany-digitalelectronics-principles-and-applications-9th-edition-tokheim/
ebookmass.com
A Strange Hymn Laura Thalassa https://ebookmass.com/product/a-strange-hymn-laura-thalassa-3/
ebookmass.com
Big Data Application in Power Systems Reza Arghandeh https://ebookmass.com/product/big-data-application-in-power-systemsreza-arghandeh/
ebookmass.com
Dr. Lincoln: A Clean Medical Romance (An Alpine Hospital Romance: Cardiology Book 3) Andrea Kate Pearson
https://ebookmass.com/product/dr-lincoln-a-clean-medical-romance-analpine-hospital-romance-cardiology-book-3-andrea-kate-pearson/
ebookmass.com
Aristotle’sEmpiricism Aristotle’s Empiricism MARCGASSER-WINGATE OxfordUniversityPressisadepartmentoftheUniversityofOxford.Itfurthers theUniversity’sobjectiveofexcellenceinresearch,scholarship,andeducation bypublishingworldwide.Oxfordisareg isteredtrademarkofOxfordUniversity PressintheUKandincertainothercountries.
PublishedintheUnitedStatesofAmericabyOxfordUniversityPress 198MadisonAvenue,NewYork,NY10016,UnitedStatesofAmerica.
©OxfordUniversityPress2021
Allrightsreserved.Nopartofthispublicationmaybereproduced,storedin aretrievalsystem,ortransmitted,inanyformorbyanymeans,withoutthe priorpermissioninwritingofOxfordUniversityPress,orasexpresslypermitted bylaw,bylicense,orundertermsagreedwiththeappropriatereproduction rightsorganization.Inquiriesconcerningreproductionoutsidethescopeofthe aboveshouldbesenttotheRightsDepartment,OxfordUniversityPress,atthe addressabove.
Youmustnotcirculatethisworkinanyotherform andyoumustimposethissameconditiononanyacquirer.
LibraryofCongressControlNumber:2021932106
ISBN978–0–19–756745–6
DOI:10.1093/oso/9780197567487.001.0001
135798642
PrintedbyIntegratedBooksInternational,UnitedStatesofAmerica
Acknowledgments ThequestionsanimatingthisprojectareonesIbecameinterestedin whileworkingonmydoctoralthesisatHarvard.Iamdeeplygrateful tomyadvisors,MaryLouiseGill,RustyJones,AlisonSimmons,and GiselaStriker,fortheircommentsandguidanceduringthoseyears.
Papersdiscussingvariousargumentsdevelopedinthebookwere presentedatconferencesandworkshopsatBrownUniversity,DartmouthCollege,HarvardUniversity,NorthwesternUniversity,the UniversityofPennsylvania,ProvidenceCollege,theUniversityof Sheffield,TempleUniversity,theUniversityofToronto,andatthe 2018EasternAPA,the2018SAGPAnnualMeeting,the2018Boston SummerWorkshop,theTahoeAncientPhilosophyWorkshop,and theOrangeBeachEpistemologyWorkshop.Iamthankfultothe audiencesattheseeventsfortheircriticismandfeedback.Thanks arealsoduetomywonderfulcolleaguesatBostonUniversity,who generouslycommentedonearly,unpolisheddraftsofthismaterial; totheBostonUniversityCenterfortheHumanities,whoseJunior FacultyFellowshipprovidedmewithacriticalsemesterofresearch leave;andtotheCABALanditsassociatesforthesupportiveand stimulatingphilosophicalcommunitytheyhelpcreate.AsAristotle remindsus,wecanalldophilosophyalone,butwedoitbetterwith friends.
Forhelpfuldiscussionsaboutthecentralideasdevelopedinthis book,Ithank:DavidCharles,SukainaHirji,RobertHowton,DhananjayJagannathan,ColinKing,EmilyKress,JimLennox,Hendrik Lorenz,KeithMcPartland,KatyMeadows,BenjaminMorison,Jessica Moss,ChristianaOlfert,EvanRodriguez,DavidRoochnik,Susan SauvéMeyer,andJacobStump.IalsowishtothankKatjaVogtand ananonymousreaderforOxfordUniversityPress,whoprovidedvery detailedandcarefulcommentsoneverypartofthemanuscript.The bookowesmuchtotheirefforts.Finally,Iamespeciallythankfulto DavidBronstein,whosethoughtfulandalwaysgenerouscomments
greatlyimprovedthecentralargumentsinthebook;toDougKremm, forhissteadyphilosophicalspiritandforalwaysbeingawilling interlocutor;toPatriciaMarechal,forcountless,invariablyhelpful conversationsaboutnearlyeveryaspectofthebook;andtoWhitney Schwab,forlookingatearlychapterdraftsandkeepingmehonest duringresearchleave.
Somepartsofchapters1and2drawon“Conviction,Priority,and RationalisminAristotle’sEpistemology,”copyright©2020Journalof theHistoryofPhilosophy,Inc.Thisarticlewasfirstpublishedin JournaloftheHistoryofPhilosophy 58(1),2020,1–27.Theexcerptsare reprintedwithpermissionbyJohnsHopkinsUniversityPress.Chapter 3isarevisedandexpandedversionof“AristotleonInductionand FirstPrinciples,” Philosophers’Imprint 16(4),2016,1–20.Someparts ofchapter4drawon“AristotleonthePerceptionofUniversals,”copyright©BSHP.Thisarticlewasfirstpublishedin BritishJournalforthe HistoryofPhilosophy 27(3),2019,446–67.Theexcerptsarereprinted bypermissionofTaylor&FrancisLtd,http://www.tandfonline.com onbehalfofBSHP.Iwishtothankthesepressesforpermittingthe inclusionoftherelevantmaterials.IalsowishtothankPeterOhlin, myeditoratOxfordUniversityPress,forhissoundadviceandhis interestinthisproject,andZaraAmdur,forherhelpcompilingthe book’sindex.Anyerrorsareofcoursemyownresponsibility.
AboveallIthankmypartner,Amanda,forherpatient,loving,and insightfulsupport.IowehermorethanIcanexpress.
Introduction Aristotleisfamousforthinkingthatallourknowledgecomesfrom perception.Ofcourse,weacquireknowledgeinmanydifferentways— welearnfromourteachers,parents,colleagues,andfriends,or fromaccumulatedpersonalexperience,orthroughvariousformsof reasoning,orindeedoutofmereforceofhabit.Butalltheseformsof learning,Aristotlethinks,restonsomesortofpriorknowledge—the knowledgeofourteachers,say,orofthepremisesinourarguments, orofourownpastexperiences.Andgenerallythispriorknowledge willitselfdependonfurtherpriorknowledge.Wheredoestheregress end?With perceptual knowledge—theonlysortofknowledgethatis notitselfderivedfromanythingprior.Perceptionisthussupposedto provideabedrockforourlearning,bysupplyinguswithknowledge inawaythatdoesn’tdependonouralreadyhavingsomeknowledge atourdisposal.
ThisviewofourcognitivebeginningsplaysacentralroleinAristotle’sepistemology.He’ssometimesconsideredthefirstempiricistfor havingheldit.Butit’snotimmediatelyclearwhattheviewismeant toentail.It’snotclear,forinstance,whattheknowledgeperception supplieswouldcontributetothemoreadvancedstatesthatAristotletakestoderivefromit—stateslikeexperience(ἐμπειρία),craft knowledge(τέχνη),orscientificunderstanding(ἐπιστήμη).Noris itclearhowweshouldunderstandthenatureofitscontribution— whatitmight mean forvariousformsofknowledgetobe“basedon” or“derivedfrom”perceptualknowledge.Indeed,tosomeearsitwill alreadysoundpuzzlingthattherecouldbesuchathingasperceptual knowledge,orthatanyformofknowledgewouldemergedirectlyfrom perception,withoutanysortofinference,orany(perhapsimplicit) relianceonourexistingbeliefsorconceptualresources.Myaiminthis bookistoshedsomelightonthesequestions:togiveanaccountofthe
sortofempiricismAristotleespoused,andofhisbroaderconception ofperception’splaceinourcognitivelives.
Ofcourseempiricismisaslipperylabel.Tosaythatallourknowledgeissuesfromperceptionisnotyettosaythatperceptionitselfplays amorethanincidentalroleinourlearning.Plato’smiddledialogues provideagoodillustrationofthispoint:perceptionispresented asanecessarystartingpointforourlearningbecauseitprompts aformofrecollectionthatsuppliesuswithexpertknowledge.But perceptionisn’titselfcharacterizedasanepistemicallyvaluablestate: itpromptsrecollectionbyperplexingus,“summoningthought”and “awakeningtheintellect”intorecoveringknowledgealreadypresent inoursouls,andwhichmightservetosortoutwhatperceptionalone couldnotmakeclear.1 Onthiskindofviewperceptiondoesserveas astartingpointforallourlearning,insofarasit occasions therest ofourcognitivedevelopment.Butitdoesn’tprovidean edifice for moreadvancedformsofknowledge,orsupplythebasicmaterialsfrom whichtheseformsofknowledgewouldbederived.
Evenifperceptiondidserveassomesortofedifice,onemight wonderjusthowmuchitcouldteachusonitsown.ForAristotle oftenemphasizesthatourcapacitytoperceiveisacapacityweshare withallanimals,andwhichyieldsaformofknowledgefarremoved fromthesortoftheoreticalunderstandinghetakesashiscognitive ideal—asheputsitinthe Metaphysics,“toperceiveiscommontoall, andthereforeeasy,andnothingwise”(A2982a11–12).Inparticular perceptiondoesn’ttellus why anythingisthecase,orwhyit must be so,orevenhowthingsarewhenwearenotactivelyperceivingthem. Giventheselimitations,it’snaturaltothinkthatAristotlemusthave attributedanysignificantepistemicachievementtosomeother,nonperceptualcognitivecapacity—a rational capacity,perhaps,oratleast acapacitythatwouldallowformoreadvancedmodesofthought.Thus perceptionmightdomorethanawakenourintellect,andyetstillbe thoughttonothavemuchtocontributeofitsown.
Norindeedwouldtakingabroadviewofperception’scontributions necessarilymakeAristotleanempiricist.Forsomeempiricistswould
1 Republic 524d1–5;seealso Phaedo 75a–b.Moreonthesepassagesbelow.
rejecttheverytermsinwhichthesecontributionsarebeingassessed— termsthatsuggestadivisionbetweenbasicformsofperceptual cognition,ontheonehand,andmoreadvancedformsofrational thought,ontheother.Onemightinsist,forinstance,thatrational thoughtisinfactnothingabovetheoperationofperception,memory, andsomeperceptuallybasedassociativecapacities,oratleastthat rationalthought,whateverexactlyitsnature,doesnotmakeusany morereliableortrustworthythanperceptionalone,andsoshouldnot betakentomarkanysubstantiveadvanceinourepistemiccondition. Thisisarguablytheviewofthefirstempiriciststobearthename.2 Onthisviewthereissimplynoquestionofperception’s contribution tomoreadvancedformsofthought.Forallworthwhilethoughtis,at bottom,perceptual.
Now,it’squiteplainthatAristotlewouldrejectthesemoreradical versionsofempiricism.3 Hetakesperceptionandperceptuallybased formsofexperiencetobefundamentallydistinctfromtheoretical knowledge,andthinkshispredecessorswereconfusedprecisely becausetheyconflatedthetwo.4 Weperceiveparticulars,ashe oftenremindsus,whiletheoreticalknowledgedealswithuniversal causesandessences,andtheinferentialandexplanatoryconnections betweenthem.Butit’salsoplainthatheisinfluencedbythebasic insightthatperceptionandmemoryarepowerfulpracticalguides, evenwithoutthecorrectiveassistanceoftheintellect:herecognizes thatnonrationalanimalscanengageinhighlycomplex,reasonable behavior,andtakesthemtoshareinthesameformofexperience werelyoninmanyofourhumanpractices.Whilereasonremainsa distinctanddistinctivelyhumancapacity,itisnotacapacityweneed todevelopsophisticatedandreliableformsofknowledge,orinterms ofwhichwewouldunderstandthevalueoftheknowledgewedevelop bynonrationalmeans.OrsoIwillbearguing.
2 Ihaveinmindheretheancientempiricistdoctors.InportrayingthemthiswayI amfollowingtheinterpretivelinedefendedinFrede(1987,243–60)andFrede(1990, 227–34).
3 It’sunlikelythatAristotlehadaccesstotheformulationofempiricismsketched above,whichismostlybasedonGalen’sreportofcertainHellenisticdebatesofunclear origins.Butheprobablydidhaveaccesstoviewsthatweresimilarinspirit—see e.g.Lorenz(2006,4–7)andthepassagescitedbelowforsomeevidencetothiseffect.
4 See Met 51009b12–13, An I.2404a25ff,or An III.3427b6ff.
WhetherthismakesAristotleanempiricistwill,unsurprisingly, dependonwhatwetakeempiricismtorequire.Myaimisnottodefend thelabel.WhatIhopetoshowisthatAristotlehadaninteresting conceptionofperception’sroleasastartingpointforourlearning, andofitsrelationtovariousmoreadvancedformsofpracticaland theoreticalknowledge.Thisisalreadyapointworthdefending—a numberofprominentcommentatorshavethoughtthatAristotlehad nothing(ornothinggood)tosayaboutwhatperceptionteachesus, orhowmuchwecanlearnonitsbasis.5 It’seasyenoughtomotivate thissortofcriticism,sinceasfarasweknowAristotleneverexplicitlydiscussedperception’sepistemiccontributions,atleastassome philosophersmightconceiveofthemtoday.Concernsaboutwhich beliefsperceptionmightjustify,andhow,forinstance,aresimply neveraddressed.Butitwouldbeamistaketoconcludefromthis thatAristotlehadasimplisticornaïveconceptionofourperceptual beginnings.Tothecontrary:I’llbearguingthatacoherentand philosophicallyrichviewofperceptualknowledgecanbefoundinthe varioustextsinwhichAristotledescribesperception’sroleinanimal life,itsinteractionwithmoreadvancedmodesofapprehension,and thecognitiveresourcesonwhichitdoesanddoesnotdepend.AsI understandit,theviewisdrivenbythreemaininsights.
Thefirstinsightisthatperceptioncansolicitbehavior:wedon’tjust perceivesomestateofaffairs,butratherperceivethingsasthingsto beavoided,orpursued,orreactedtoinsomeother,morecomplex manner.Thuswhenattackedbyabearyoumightperceiveabrown ursineshapemovingtowardyou,butitwillalsobepartofyour experiencethatthebearissomething tobeavoided or toberunaway from.Perceptionsuggestsaresponsetoyoursituation,anditsaffective profilemotivatesyoutoactasitsuggests—anditdoesthiswithout yourdeliberatingorengaginginanysortofrationalthought,andthus inamanneravailabletohumansandnonrationalanimalsalike.
5 Barnes,forinstance,tellsusthat“weneedanaccount,whichAristotlenowhere gives,ofhowsuchconceptsas man arederivedfromthedataofperception”(1993,266), BurnyeatcharacterizesAristotle’streatmentofperceptuallearningas“perfunctoryin theextreme”(1981,133),andIrwintakesAristotletodisplaya“rathernaïveattitudeto observationandinduction”(1988,137;seealso32–36).
Thesecondinsightisthatperceptioncanbe trained:wecanexpand therangeofthingswerecognizeperceptuallyandtherangeofactions perceptionsolicitsfromus.Andagain,wecandothiswithoutrelying onrationalmodesofthought,usingonlyperceptionandsomeauxiliaryperceptualcapacities—inparticular,ourcapacitytoretainwhat weperceiveandassociatesomeoccurrentperceptionwiththeperceptionswe’veretained.6 Together,theseperceptualcapacitiesprovidethe basisforacomplexformofpracticalexperience(ἐμπειρία).Tohave suchpracticalexperienceistobedisposedtorespondappropriately toarangeofsituationswithoutunderstandingtheappropriateness ofone’sresponse—torecognizewhatsomesituationcallsforwithout knowingwhyitcallsforit.Thusanexperienceddoctormight reliablycuremalarialpatientswithoutbeingabletoexplainwhyher treatmentswork,orhowshecomestoherdiagnoses.OnAristotle’s viewsuchadoctorwould perceive somegivenmalarialpatientas someonetobeleeched,despitenotrecognizingthatthepatientis malarial,andtobeleechedforthatreason.Trainedperceptionofthis sortservesasareliablenonrationalguideinourvariouspractices, medicalorotherwise,andinthepursuitsandgoal-directedbehaviors ofallbutthemostbasicofanimals.
Thethirdinsightisthatthisperceptuallygroundedformofpractical experienceprovidesabasisformoreadvanced, causal formsof knowledge—and,ultimately,forthecognitiveidealthatisscientific understanding.Itprovidesthisbasisbecausethedomainsinwhich wecanbecomeexperiencedarepreciselythosethatcanbeinvestigatedintheoreticalterms:thefactthatsomedoctorsarereliably successfulatcuringmalarialsisasignthatmalariaisadiseasethat admitsoftheoreticaltreatment,thatis,adiseasewhoseeffectsand relationtootherdiseaseswemightcometounderstandscientifically. Aristotletakesthistofollowfromthefactthatwecanbeperceptually responsivetocertainuniversalcausesevenifwecannotperceivethe universalsthemselves.Wedosobyreliablydiscriminatingthefeatures perceptibleentitiespossessbecausetheyinstantiatetheuniversals
6 TheauxiliarycapacitiesIhaveinmindarethoseprovidedforbyperceptual phantasia.BelowIwillexamineinmoredetailtheroleAristotletakesperceptual phantasia toplayinanimalcognition,andthesenseinwhichitservesasanauxiliaryto perception.
inquestion—e.g.bydiscriminatingthesymptomsofmalariainour patients,thoughwecannotperceivetheuniversal malaria itself.
Aristotlethereforethinkswecanbeperceptuallyresponsiveto certainuniversalswithoutrecognizingthemassuch:perceptible particularspossesscertainfeaturesbecausetheyinstantiatecertain universals,andperceptionallowsustodiscriminatethesefeatures andexperiencethemasaction-guidingaspectsofourenvironment. Perception,sounderstood,servesasabasisforthedevelopmentof aperceptuallygroundedformofpracticalexperience.Andpractical experience,inturn,servesasabasisforadvancedcausalknowledge— knowledgethattreatsinexplicittermstheexplanatorystructurethat underpinsthereliablysuccessfulpracticeofthosewithexperience, andwhoseobjectsarethereforeuniversals,andrecognizedassuch.7 Aristotlecallsthecognitivedevelopmentleadingfromperception tosuchadvancedknowledge induction,andintendshisinductive accountofourlearningasanalternativetocertaininnatistviewshe findsabsurd—thecontrastdependinginlargepartontherolehe assignsperceptualknowledgeinourinductiveprogress.
Takentogether,thesethreeinsightsyieldaninteresting,moderate kindofempiricism.Thisisakindofempiricisminthatourperceptual powers(unassistedbytheintellect)affordusanepistemicallyvaluable formofknowledge,whichitselfservesasabasisforthedevelopmentofadvancedcausalknowledge.Thisempiricismismoderatein thatrationalreflectiononourperceptualknowledgeisrequiredto developanysuchadvancedcausalknowledge.Andthisempiricism isinterestingbothonpurelyphilosophicalgroundsandbecauseofits interpretiveimplications.Indeed,Aristotleoftenseemsratherquickto dismissinnatistviewsofourlearning.Understandinghisempiricism willhelpilluminatewhatexactlyhisalternativetotheseviewsis meanttobe,andgiveusabettersenseofthebackgroundzoological andpsychologicalconsiderationsmotivatingthisalternative.Itwill alsohelpilluminatesomecriticalpartsofhismoralpsychology— inparticular,theviewthatperceptioniscentraltothedevelopment
7 Wheretheresultingknowledgeiseithercraftknowledge(ifitaimsatproducing something)orscientificunderstanding(ifitaimsatworkingoutwhatisthecase).Iwill alsobeconsideringthecaseofpracticalwisdom,andargueweshouldthinkofitonthe modelofcraftknowledge.
anddeploymentofpracticalwisdom,andthatvirtuousagentsmust ultimately perceive whattodointhevarioussituationstheyface. Finally,itwillhelpilluminateAristotle’sconceptionofourhumanity: articulatingthedifferencebetweenrationalandperceptualformsof knowledgetellsuswhichpartsofourcognitivelifewesharewith otheranimals,andwhatisdistinctiveaboutthemodesofthoughtthat separateusfromthem.
Here,inoutline,ishowallthiswillbedefended.Ibeginchapter 1withsomegroundclearing:Iarguethattalkofperceptualknowledgeisnotacompletenonstarter,andexplainhowIwillbeusing variousknowledgetermstocapturethedifferentcognitivestates thatfeatureinAristotle’sepistemology.Ithenofferanaccountof scientificunderstanding(Aristotle’sepistemicideal)asaformof theoreticalexpertiserequiringasynoptic,reflectiveappreciationof theexplanatorystructureofsomedomain,andinparticularofthe rolethetruthscomprisingthatdomainmightplayinaspecialkind ofdemonstration.Iofferthisaccountearlyoninordertoresistviews thatmakescientificunderstandingthesolelocusofjustification,and concludefromthisthatperceptioncouldnotitselfplayanysignificant epistemicrole.Iarguethat,forallhisremarksaboutdemonstrative principlesbeing“mostconvincing”and“primitive,”Aristotledidnot considerthemanultimateoruniquesourceofjustification.However Ialsoraisesomeconcernsaboutinvokingtalkofjustificationinthis context,andsuggestanalternative,pluralisticconceptionofepistemic valuethatIthinkbetterfitsAristotle’sepistemology.IndoingsoIhope toexplainwhyweshouldresistdismissivetakesonourperceptual accomplishmentsbasedonAristotle’sdescriptionsofhisepistemic ideal.ButIalsohopetoarticulatethetermsinwhichIthinkwe shouldunderstandAristotle’saccountofourcognitivedevelopment, andperception’splacewithinit.
Inchapter2IconsiderthePlatonicbackgroundforAristotle’s accountofourcognitivedevelopment,suchaswefinditinsome ofPlato’smiddledialogues.Iarguethat,onthesurface,thePlatonic viewagreeswithAristotle’sclaimthatalladvancedknowledgemust come“from”perception:it’sfromperceptionthatweareprompted tothinkofourperceptualshortcomings,andreferourperceptionsto theknowledgeofFormswegoontorecollect.Buttheknowledgewe
recollectisnotmeanttobederivedonthebasisofwhatweperceive. ForourknowledgeofFormsissomethingwe’realreadymeantto possess,insomelatentform,beforeweperceiveanythingatall— andperceptioncontributestoourrecoveringthisknowledgeonly bycompellingustoreflectonitsowndeficiencies.Sothefactthat wemustbeginfromperceptionisalamentableconsequenceofour embodiedexistence,notsomethingthatwouldsupplyuswithany valuableformofknowledge,orbeusefulexceptasameanstorecollect. Aristotleexplicitlypresentshisaccountoftheperceptualoriginsof ourlearningasanalternativetothissortofview.Hedoesso,Iargue, preciselybecausehetakesperceptiontobeanepistemicallyvaluable capacity—acapacitywhoseexercisesuppliesuswithsophisticated formsofknowledgeevenwithouttheinterventionoftheintellect. WhatthisshowsisthatAristotledidnottakeperceptiontobe foundationalinanetiologicalsenseonly,asacausalprecursortothe restofourlearning.Perceptionisindeedsuchacausalprecursor,butit isalsoasourceofknowledgethatwecansafelyrelyoninourlearning andpracticalpursuits.
Inchapter3Iarguethatwelearnscientificfirstprinciplesby induction.ThisiswhatAristotlesays,butfewtakehimathisword. Iofferareadingofthelastchapterofthe Analytics onwhichthis claimmakessense.Itinvolvesconceivingofinductionasasomewhat morerobustprocessthaniscommonlythought—roughly,asaformof cognitiveprogressfromarangeofparticulartruthstosomeuniversal explanationwhyallthesetruthshold.Iarguetherearegoodgrounds tothinkthisiswhatAristotlehasinmindinthe Analytics,andIgo throughsomeexampleswhereheseemstobedisplayingjustthissort ofprogressinhisownscientificworks.Oneupshotofthisviewis thatwedon’tneedtothinkthatAristotlegaveuponinductionasa pathtowardprinciples,invokingsomekindofrationalintuitionor someothersetofmethodstoremedyitsdeficiencies.Anotherupshot isthattheknowledgeperceptionprovidesplaysacentralroleinour inductivelearningofprinciples,supplyingthebasicbodyoffactsthese principlesaremeanttoexplain.
Inchapter4Iconsideringreaterdetailtheroleperceptionis meanttoplayintheearlystagesofthisinductiveprocess.Ibeginby defendinganexpansivereadingofperceptualobjectsandcontents,
onwhichweperceivenotjustcolors,sounds,andsoon,butalso Callias,lyres,andloavesofbread,andwhetherCalliasisnear,the lyrewell-tuned,andtheloafbaked.Iconsiderwhatrelationship thisbroadperceptionhastothecharacterizationofsense-perception wefindin DeAnima.IarguethatforAristotle, perse perceptibles (theobjectsofoneormoresensemodality)arefundamentalina psychologicalsense:theyservetodefinethesensorypowerswhose operationunderliesallperceptualactivity.ButIfurtherarguethatthey arenotfundamentalinanepistemologicalsense:ourknowledgeof these perse perceptiblesshouldnotbetakentoprovidea“given”that wouldserveasanexclusivebasisforallourlearning.Ialsoconsider whetherornotbroadperceptionwouldimplicitlydependonour conceptualresourcesorsomesortof“cognitivepenetration”fromthe intellect,andargueagainstviewsonwhichourperceptualpowers areinherentlytransformedbytherationalpartofoursoul.Finally, IconsiderinmoredetailAristotle’stakeonperceptuallearning, focusingonhisclaimthatourperceptionsare“ofuniversals”even thoughweperceiveparticulars,hisbroaderconceptionofparticular anduniversalcognitivestates,andhisportrayalofourpretheoretical apprehensionof“compound”universalswemightseektodefinemore precisely.IarguethatAristotlethoughtwecouldbeperceptually responsivetouniversalswedonotyetrecognizeassuch,andthatwe canbesoresponsiveeventhoughweonlyperceiveparticulars—and thatthisviewinformshisgenerousconceptionofthesortofknowledgepossessedbythosewithnonrational,perceptuallygrounded formsofexperience.
Inchapter5Iexaminethezoologicalandpsychologicalviewsthat accountfortheperceptuallearningdiscussedinchapter4.These views,Iargue,aremotivatedbytwokeyideasdevelopedinAristotle’s accountofanimallocomotion.Thefirstistheideathatperception cansolicitsomebehaviorfromaperceivingsubject:weandother animalsperceivehowthingsare,butalsoperceivehowtoactinthe varioussituationsweface.Thisispossiblebecauseweperceivethings aspleasantandpainful,andthusasobjectsofsomeofourappetites, andalsoperceivethemeansnecessarytofulfilltheseappetites— sothatperceptionhasbothcognitiveandaffectivedimensions,the combinationofwhichallowsittoplayakeymotivationalrole.The
secondideaisthatperceptual phantasia makespossibletheretention ofpastperceptionsasmemories,andtheassociationofpastmemorieswithsomeoccurrentperception.Thismechanismallowspast perceptionstoinformwhatwerecognizeperceptually,andaccounts forthedevelopmentofthecomplex,goal-directeddispositionsthat characterizethestateofexperience—aswefinditinbothhumans andnonrationalanimals.Iendbycontrastingthestateofexperience withformsofknowledgethatdodependonourrationalpowers,and consideringthelimitsofAristotle’sempiricism.
Inchapter6IexaminewhatimplicationsthisaccounthasforAristotle’smoralepistemology.MymainquestionhereishowAristotle’s empiricistviewsbearonourunderstandingoftheroleperception playsforthevirtuous—andwhetherornot,asisoftenthought,they pointtowardacertainformofethicalparticularism,accordingto whichuniversalrulescouldneveradequatelycodifyvirtuousconduct. Iarguetheydonot.Whileitmightbeinexpedientforvirtuousagents toattempttoarticulateuniversalrulesofconductwithanysortof precision,thisisnotbecauseitisinprincipleimpossibletodoso, orbecausethesubjectmatterofethicsdoesnotadmitoftheoreticaltreatment.Still,whileethicaltheorizingispossible,perception andperceptuallygroundedexperienceplayanindispensablerolein thedevelopmentanddeploymentofpracticalwisdom.Forethical theorizing,whilepossibleinprinciple,doesnotitselfprovideany guidancehowtoact:ourlearningtobevirtuousdependsonfirsthand,personalexperiencethattheoreticalmodesofthoughtcould notprovide.Iarguethatthisissimplyaconsequenceofthefactthat ethicsisa practical subjectmatter:ananalogouspointcouldbemade (andisarguablymadebyAristotle)aboutadoctor’sknowledgeof medicine.Ifthisisright,ethicsisnotanexceptionalsubjectmatter, differentinkindfromthosethatadmitofuniversaltreatment—even ifpracticalwisdomdoesrequireaccumulatedpersonalexperiencein awayourknowledgeofsomepurelytheoreticaldisciplinedoesnot.I endbyconsideringwhatapracticallyoriented,universaltreatmentof virtuousconductwouldlooklike,andhowwemightconceiveofits ethicalsignificance.
1 Perception,Knowledge,and UnderstandinginAristotle’s Epistemology ThetopicofthisbookisAristotle’sconceptionofperceptual knowledge—whatthisknowledgeamountsto,howweacquireit,and howweshouldunderstandtherelationshipbetweenthisknowledge andtheother,moreadvancedformsofknowledgewemightdevelop onitsbasis.Tosomethiswillsoundlikeanonstarter:thereissimply nosuchthingasperceptualknowledgeinAristotle,andsothereisno questionwhatsortofknowledgeitis,orhowthisknowledgemight contributetoourlearning.Onemightofcourseallowthatperceiving thingswould,insomecases,resultinourdevelopingsomekindof knowledgeaboutthethingsweperceive.Buttheresultingknowledge insuchcases,theobjectiongoes,isnotsuppliedbyperceptionalone. Forsomesortofmediationisneededtoturnourperceptionsinto knowledge,andthismediationissomethingperceptionalonecould notprovide.
Objectionsofthissortcanbemotivatedinanumberofdifferent ways.Somewillarguethatperception,forAristotle,doesn’titself justifyanything,orprovideanygroundsforrationalconviction—and thatitthereforecouldnotbetakentoprovideuswithanyknowledge onitsown.Otherswillpointoutthatperception,forAristotle, hasparticular,contingentobjects,whileknowledgeparadigmatically concernsitselfwithnecessary,universaltruths,andtheexplanatory relationsbetweenthem.Othersyetwillnotethatperceptualmodes ofcognition,forAristotle,aresharedwitheventhemostprimitive nonrationalanimals,andarguethattheseprimitiveanimalscouldn’t rightlybesaidtoknowanything,becausetheircognitivelivesare tooimpoverishedtoformthesortsofstatesthatwouldconstitute
Aristotle’sEmpiricism.MarcGasser-Wingate, OxfordUniversityPress(2021).©OxfordUniversityPress. DOI:10.1093/oso/9780197567487.003.0001
aristotle’sempiricism
knowledge.Inallthesecasesthethoughtisthatknowledgeissupposed tobesomesortofepistemicaccomplishment,andthatperception,as Aristotledescribesit,isjusttoolimitedtoyieldanythingthatwould qualifyassuch.
Myaiminthischapteristodispelsomeoftheseinitialconcernsand toconsider,morebroadly,howweshouldapproachquestionsabout therelationshipbetweenperceptionandthemoreadvancedcognitive statesthatappearinAristotle’sdescriptionsofourlearning.I’llbegin withapreliminaryargumentagainsttheviewthattalkofperceptual knowledgeisanonstarter.Theargumentissimple:perceptionis (oryields,whensuccessfullyexercised)akindof γνῶσις; γνῶσις, asAristotleusestheterminhisaccountofourlearning,isfactive cognition;andfactivecognitionissomethingwemightplausiblytake toconstituteknowledge.Ifthisisright,itestablishesthatperception yieldsakindofknowledge—atleastononephilosophicallydefensible characterizationofwhatknowledgeis.
Now,thisisamodestclaim,aimedprimarilyatthosewhowould putdownthebookonesentencein.Wemightofcoursedisputethat knowledgeisfactivecognition,orinsistthatknowledgerequiresa kindofreflectionaboutthecontentsofourexperiencesorbeliefs whichperceptioncouldnotprovide.Forthepurposesofmybroader argument,however,itwon’tmattermuchwhetherornotweareright tocallperceptual γνῶσις aformofknowledge.Thatwillinevitably dependonwhatwetakeknowledgetorequire,andthereisn’tmuch togainbyterminologicalintransigence.1 Whatwillmatteristhat perceptionissufficienttoyieldepistemicallyvaluablecognitivestates, ratherthanmerelyservingasacausalprecursortosuchstates,which
1 Ifindoverlyrestrictiveconceptionsofknowledgeimplausible,andespeciallyoutof placeinthecontextofAristotle’sepistemology.SopartofthereasonIfinditappropriate torender γνῶσις as“knowledge,”andpresentperceptual γνῶσις aspecifickindof knowledge,isthatIthinkthiscapturesAristotle’sgenerousconceptionofthecognitive accomplishmentsattributabletotheperceptualpartofoursoul,andthusavailable(asI willbearguing)tobothrationalandnonrationalanimals.Butthischoiceoftranslation isnotmeanttobeprejudicial:thecentralquestioniswhattheseaccomplishments amountto,andhowweshouldunderstandtheirvalueandtherelationshiptheybear tomoresophisticatedformsofunderstanding.Ifyouagreewithmeonthesepointsbut nonethelesswishtocallperceptual γνῶσις somethingelse,andreserve“knowledge”for somemoreexaltedcognitivestate,Iamhappytogiveupthelabel.
perception,knowledge,andunderstanding3 wouldprompttheirdevelopmentbyother,intellectuallyinvolved means.
Andthisispreciselywhatanumberofcommentatorswould deny(“rationalists,”asI’llbecallingthem).Ontherationalistview, ἐπιστήμη is,forAristotle,thesolelocusofepistemicvalue:anything thatfallsshortof ἐπιστήμη doesnotstrictlyspeakingconstitute anyformofknowledge,oranysortofepistemicallyvaluablestate.
SinceAristotleisclearthatperceptionaloneneveryields ἐπιστήμη,it followsthatperceptionaloneneveryieldsanyepistemicallyvaluable state.Perceptionmightofcourseremainanecessaryprerequisite tothedevelopmentofsuchstates,andperhapshelpexplainthe psychologicalunderpinningsofthisdevelopment—butitwouldnot itselfsupplythebasicknowledgewemightuseasafoundationfor thedevelopmentoffurther,moreadvancedformsofknowledgeand understanding.For,asthepointistypicallyput,perceptioncannot providethesortof justification Aristotle’sconceptionofknowledge requires:thatsortofjustificationmustissuefromourrationalintuitionofcertainprinciples,andnothingelse,andisthereforeavailable onlytosomeonewith
AsIunderstandthem,rationalistinterpretationsalongtheselines aremotivatedbybackgroundviewsabouttherolejustificationand convictionplayinAristotle’sepistemology,andwhatitwouldtakefor somecognitivestatetocountasanepistemicallyvaluableone.Inthis chapterandthenext,Iwillbearguingthattheseviewsaremistaken— oratleastthattheyarenotsupportedbyAristotle’sdiscussionsof thenatureanddevelopmentof ἐπιστήμη.Iwillalsobedefending analternativeconceptionofepistemicvalue,whichIthinkbetterfits Aristotle’sdescriptionofourlearning,andshedssomelightonthe differentepistemicaccomplishmentshedescribes,andinparticular thoseaccomplishmentsheattributestotheperceptualpartofthesoul.
2 ForsuchviewsseeinparticularFrede(1996,172)andIrwin(1988,132–36).Some rationalistviewsaremorenuanced,andallowthatstatesotherthan ἐπιστήμη might countasepistemicallyvaluableinsofarastheyapproximate ἐπιστήμη toasufficient degree(seee.g.Fine(2010,152–55)).Buteventhesemorenuancedviewsassumethat ἐπιστήμη providesthesolestandardagainstwhichepistemicvalueistobemeasured. InwhatfollowsIwillargueagainstthisassumption,anddefendamoreexpansive, pluralisticconceptionofepistemicvalue.
IndoingsoIhopetoexplainwhyweshouldresistthedismissive takeonperceptionendorsedbyrationalistinterpreters.ButIalso hopetoarticulatethetermsinwhichwewouldbestunderstand Aristotle’saccountofourlearning,perception’sroleinthisaccount, andthecontrastbetweenvariousperceptualandintellectualformsof knowledge.
1.1Aristotle’sEpistemicTerminology BeforetakingonrationalistreadingsofAristotle’sepistemology,let mereturntoamodestpoint:talkofperceptualknowledgeisnot acompletenonstarterinthiscontext.It’snotacompletenonstarter becauseperception,asAristotlefrequentlyemphasizes,isakindof γνῶσις,and γνῶσις,asAristotleusestheterminhisdiscussions ofourlearning,canreasonablybetakentoconstituteknowledge— tohave γνῶσις of X (orto γιγνώσκειν or γνωρίζειν X )isto know X,onatleastsomephilosophicallyplausibleconceptionsof knowledge.
Thatperceptionisa γνῶσις is,Itakeit,uncontroversial.Hereare thetextswhereAristotlearticulatesthepointmostclearly:3
[1]Toseeissomethingwepreferoveralmosteverythingelse—not justwithaviewtowardaction,butevenwhenwearen’tsettingout todoanything.Thereasonisthatseeingmakesus γνωρίζειν most ofallthesenses,andrevealsmanydifferencesbetweenthings.(Met A1980a24–27)
[2]Wedonotregardanyofthesensesaswisdom;yetsurelythese givethemostauthoritative γνώσεις ofparticulars.Buttheydonot tellusthe“why”ofanything—e.g.whyfireishot;theyonlysaythat itishot.(Met A1981b10–13)
3 Seealso Top V.3131b23–28andthefurtherreferencesinBurnyeat(1981,114).Here andhenceforth,unattributedtranslationsaremyown.
perception,knowledge,andunderstanding5
[3]Thefunctionofananimalisnotjusttogenerate:thatiscommon toalllivingthings.Allanimalspartakeinakindof γνῶσις;some more,someless,someverylittleindeed.Fortheyhaveperception, andperceptionisakindof γνῶσις.(GA I.23731a30–34)
[4]Ofwhatispresentthereisnomemory,butratherperception.For bythelatterwedonot γνωρίζειν whatisfutureorpast,butwhatis presentonly,whereasmemoryisofthepast.(Mem 449b13–15)
[5]Wemust γνωρίζειν magnitudeandmotionbymeansofthesame capacitybywhichwe γνωρίζειν time.Soit’sclearthatit’sbythe primaryfacultyofperceptionthatthereis γνῶσις ofthese.(Mem 450a9–11)
Perception,inthesepassages,isadvancedassomethingthateither yields somesortof γνῶσις (asin[1]and[2]),orsomething bywhich we γνωρίζειν,orbywhichthereis γνῶσις (asin[4]and[5]),or somethingthatsimply is akindof γνῶσις (asin[3]).Itakeitthese differentformulationsreflectthedifferentwayswemightconceiveof perceptionitself:asapowerwhoseuseresultsinsomesortofcognitive state,orasthecognitivestateresultingfromthesuccessfuluseofthis power.Whenconceivedofasapower,perceptionissomethingthat yieldsakindof γνῶσις,orthroughtheuseofwhichwe γνωρίζειν. Whenconceivedofasthestateresultingfromthesuccessfuluseof thispower,perceptionjustisakindof γνῶσις.Infact γνῶσις itself isusedinbothoftheseways:in[2]and[5]itpicksoutacognitive state,whilein[3]itpicksoutthemeansbywhichthisstateisnormally broughtabout.Soperceptioncanbetakentocountasa γνῶσις both qua poweror qua state.4
Now, γνῶσις anditscognatesaretermsthatresisteasytranslation. Aristotleusesthemtocoveranextremelybroadrangeofstatesand modesofapprehension,inanumberofdifferentcontexts—sothat sometranslatorsoptforthe(presumablymoreneutral)“cognition”
4 IwillbefollowingAristotleinusing“perception”todenoteboththepowerto perceiveandtheperceptualexperienceorstateresultingfromtheuseofthispower—in contextitwillbeclearwhichisinvoked.FormoreonAristotle’sterminologyinthese cases,seepp.48–49.
aristotle’sempiricism
ratherthan“knowledge.”Thussomemightagreethatperception isakindof γνῶσις,butdenythatperceptionyieldsanykindof knowledge:ourperceptualpowersmightbetakentoallowforsome sortofawarenessoftheworldunavailabletoplants,yetstillfallshort ofsupplyingus,ontheirown,withanysortofknowledge.
Inhisdiscussionsofourlearning,however,Aristotledoesnotuse γνῶσις topickoutmereawareness.Forsomethingtocountasa γνῶσις itmustbeacaseof factive cognition,thatis,cognitionthat linksagentstotruths.5 ThisisapointBronsteinhasrecentlydefended ingreatdetail,andItakehisargumentstothiseffecttobedecisive.6 Tomymind,it’sapointthatshouldcomeasnosurprise.ForAristotle’s mainaccountofourlearningisframedasananswertothequestion howwemightcometo γνωρίζειν certainprinciples,andwhatprior formsof γνῶσις mustbethereforour γνῶσις ofprinciplestobe broughtabout(APo II.1999b17–30).Andwhatheisattemptingto explainisnothowsomepotentiallymistakencognitionorawareness ofprincipleswouldbepossible,butratherhowwemightcometo know theseprinciples,andunderstandthemforwhattheyare.Anythingless wouldnotreallyconstituteanexplanationhowwelearnprinciples: tolearnwemustendupwiththetruth.7 So γνῶσις mightbeused morebroadlyinothercontexts,butwhenitisinvokedtodescribeour learning,itpicksoutfactivecognition,specifically.
5 Nowadaysfactivityisoftenunderstoodinpropositionalterms—acognitivestateis factivewhenanagent’sbearingittoward p entailsthat p istrue.ForAristotle,however, thingscanbetrueorfalsewhicharenotpropositions:hetellsustheperceptionof special perceptiblesisalwaystrue(An III.3427b11–12),or,somewhatmorecautiously,thatit admitsoffalsehoodtheleast(428b18–19),andthissortofperceptiondoesnotinvolve anysortofpredicativecombinationorpropositionalstructure(cf.4.3n37).Iintend factivitytocoverthosecasesaswell.Formoreonperceptionanditscontents,see4.1.
6 SeeBronstein(2016a,16–21).Theviewthat γνῶσις isfactiveinthecontextof Aristotle’sdiscussionsofourlearning(thoughperhapsnotingeneral)isalsoimplicit inagooddealofscholarshiponAristotle’sepistemology,wherethetermisrenderedas “knowledge.”See,forinstance,Ackrill(1981),Barnes(1993),BoltonandCode(2012), Burnyeat(1981,2011),Karbowski(2016),Kosman(1973),orTaylor(1990).Fora dissentingview,seeFine(2010),andalsoFerejohn(1991,68–69)andLaBarge(2006, 38).
7 Noteonthispointthephrasingat APo II.1999b28–29,whereAristotleasks aboutprinciples πῶςἂνγνωρίζοιμενκαὶμανθάνοιμενἐκμὴπροϋπαρχούσης γνώσεως.Seealso Met A1980a27–b25,wheredifferentformsofperceptual γνῶσις are saidtoallowfordifferentformsoflearning(μανθάνειν,980b23–24).Itakeit μανθάνειν isfactiveinthesepassages,andthereforesois γνῶσις.
perception,knowledge,andunderstanding7
Onemightobjectherethat γνῶσις cannotbefactivecognition becauseperceptionisa γνῶσις,andnotallourperceptionsaretrue. Butwecanholdthatperceptionisawayto γνωρίζειν thingswithout holdingthateverysingleperceptualstatecountsasa γνῶσις.To callperceptiona γνῶσις,asIseeit,istosaythatperceptionis acapacitywhoseexercise reliably yieldsfactivecognition—aview Aristotlethinkswarrantedonbroadlyzoologicalgrounds,asIwill bearguingbelow(seeinparticularchapter5).It’scompatiblewith thisviewthatperceptionmightincertaincasesfailtoyieldfactive cognitivestates,orpresentthingsinaninaccurateway,solongasit doesnotnormallymislead.8 It’salsocompatiblewiththisviewthat perceptionwouldrevealtousonlyalimitedportionofreality—and inparticularthatitwouldnotrevealtousanyofthecausesthat wouldexplainwhatweperceive,andwhichAristotleconsiders“more true”thanwhattheyservetoexplain(cf. EE I.61216b26–35, Met α 1 993b19–31,andsection4.6).Perceptual γνῶσις,afterall,ismeantto belongtoeventhemostbasicanimals(cf.[3]).
Sotheexerciseofourperceptualpowersreliablyyieldsakindof γνῶσις,and γνῶσις,inthiscontextatleast,isfactivecognition.This doesnotyetestablishthatperceptionyieldsakindofknowledge.For onemightdenythatfactivecognitionconstitutesknowledge—one mightthinkthatknowledgerequiresjustification,oratleastsomesort ofreflectiveappreciationoftheepistemicstandingofourbeliefsor othercognitivestates.Note,however,thatatthispointIamarguing onlythattalkofperceptualknowledgeisn’tanonstarter.Andthe viewthatunreflectivefactivecognitionconstitutesknowledge,while ofcourseupfordispute,isn’tanonstarter.Consider,forinstance,the sortofanimalknowledgeSosadistinguishesfromitsmorereflective counterpart:
8 Aristotledoesofcourseseektoexplainperceptualerrorinhispsychologicalworks. Buthedoesn’tdothisinordertoexplaintheroleperceptionmightplayasastarting pointforourlearning,nordoeshetakeasepistemicallybasicthesortofperception heconsidersleastfallible(onwhichpointseepp.124–25).Itakeitthisisbecausehis accountofourlearningisnotprimarilyintendedasarecipeforthesecureaggregation oftruths,andtakesforgrantedthereliabilityofsomeofourcognitivepowers—for instance,in[2],thereliabilityofperceptioninreportingthatfireishot.
