Philosophical Review Volume 78 Issue 3 1969 (Doi 10.2307/2183829) Gregory Vlastos - Reasons and Causes in The Phaedo

Philosophical Review
Reasons and Causes in the Phaedo

Author(s): Gregory Vlastos
Reviewed work(s):
Source: The Philosophical Review, Vol. 78, No. 3 (Jul., 1969), pp. 291-325
Published by: Duke University Press on behalf of Philosophical Review
Stable URL: http://www.jstor.org/stable/2183829 .
Accessed: 18/11/2012 10:16
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
Duke University Press and Philosophical Review are collaborating with JSTOR to digitize, preserve and extend
access to The Philosophical Review.
http://www.jstor.org
This content downloaded by the authorized user from 192.168.72.225 on Sun, 18 Nov 2012 10:16:09 AM
All use subject to JSTOR Terms and Conditions
REASONS AND CAUSES IN THE PHAEDO1
THERE is a passagein thisdialoguewhichhas led many

scholars-thegreatmajority ofthosewhohavetranslated or
discussedit in detail-to thinkthatPlato'sFormsare meantto be
causes. This is the methodologicaland metaphysicalpreamble
(95E-Io5C) to thefinalargumentfortheimmortality ofthesoul.
The importanceof thispassage could hardlybe exaggerated:as
much is to be learnedfromit about Plato's metaphysics, episte-
mology,and philosophy ofscienceas fromany othertextofequal
lengthin his corpus.But it is also one of the mostperplexing.
Scholarswho have notconfessed itsdifficulty
have evidencedthis
difficulty
just the same in the wild diversity theinterpretations
of
theyhave put on it. Accordingto Eduard Zeller,Plato teaches
here that the Ideas are meantto be formal,efficient, and final
causes all rolledinto one.2 Paul Shorey,at the otherextreme,
maintainedthat when Plato speaksof the Ideas as aitiai he is
offering"only a tautological logic . .. a consistentand systematic
of the logical reasonforall otherformsof cause."3
substitution
1 An earlierversionofthispaper was deliveredat MichiganState University

as an ArnoldIsenbergLecture,and otherdraftshave been read elsewhere.I
acknowledgewithgratitudediversecriticisms whichhave helped me improve
the paper, and mostparticularlythosegivenme by Professors TerryPenner
and Richard Sorabji, whose detailed and penetrating querieshave prompted
revisionswhichhave clarifiedand strengthened theargument,thoughperhaps
withoutfullymeetingtheirobjections.A debt of anothersort I owe to the
CenterforAdvanced Studyin the BehavioralSciencesat Stanfordwherethe
presentdraftwas completed.
2 "In dieserganzenAuseinandersetzung [withthe physicists in our passage] wird
nunzwischen derbegrifflichen,
derwirkenden undderEndursache nichtloss nichtunter-
schieden,sondernalle dreiwerden deutlich
genug far Ein unddasselbeerkldrt."Zeller,
PhilosophiederGriechen, 5th ed. (Leipzig, 1922), II, i, 687, n. i. I shall not
criticizethisstatement directly,butmyreasonsforrejectingitwillbecomeclear
as I proceed. I shall followthe same policywithrespectto otherviewswith
whichI cannot agree. Limitationsofspace will preventme fromengagingin
controversy except in so far as I findthis essentialforthe elucidationand
supportof the interpretations I propose.
3 WhatPlato Said (Chicago, I933), p. I79. Zeller's contraryinterpretation
Shoreyhad alreadyrejectedin "The Interpretation of the Timaeus,"American
Journal ofPhilology, 9 (i888), 395 ff. (at p. 406) and stillearlierin his Munich
29I
GREGORT VLASTOS
More recently,commentatorshave not hesitatedto take theirown

perplexed reading of this text as evidence of unclear thinkingin
its writer. I. M. Crombie tells us thereis "a nest of confusions"
here, arising from Plato's "jumbl[ing] mathematical and non-
mathematical topics together,and fail[ing] into the bargain to
distinguishdifferentsenses of such notions as 'through' and 'in
virtueof.' " He says: "it would be a usefulelementaryexerciseto
make a list of such confusionsin this passage."4
The interpretationI shall offerhere is closer to Shorey's than
to Zeller's, and owes no small debt to Crombie's discussionof our
passage.5 But if my analysis is even approximatelycorrect,it will
show that neitheris Shorey's view acceptable in toto,and that the
"confusions" of which Crombie speaks are not in our text but in
misunderstandingsof it which he shares with many distinguished
scholars.Not that Plato's thoughthere will turnout to be entirely
clear or whollytrue. But forall its blemishes,both substantiveand
expository,it will appear, I trust,to be worthyof a philosopher
who was not only a pioneerofunsurpassedaudacity but also, when
full allowance is made for the difficultieswhich confrontedhim,
a remarkablysane and clear-headed thinker.
I. ON THE MEANING OF AITIA
Since so much will turn on the meaning of the word aitia,6I

must begin by calling attentionto the fact thatitsrange of signif-
dissertation:see his citationfromthe latterin "The Originofthe Syllogism,"

ClassicalPhilology,19 (1924), 7, n. 2.
4An Examination ofPlato'sDoctrine (London, I963), II, i69. I shall referto
thisworkhereafter solelyby the author'sname.
5 As well as to manyotherauthors,thefollowing mostofall:
J. Burnet,Plato'sPlhaedo(Oxford,I 9 I i).
A. E. Taylor,Plato,theMan andhis Work,4th (rev.) ed. (London, I937).
F. M. (iornford,PlatoandParmenides (London, 1939), pp. 74-80.
J. Moreau, La Construction de l'IdIalismePlatonicien(Paris, 1939) pp. 378 ff.
N. R. Murphy,TheInterpretation ofPlato'sRepublic(Oxford, I951) pp. 145-
148.
To each oftheseworksI shall referhereafter solelyby theauthor'sname.
6 Liddell and Scott,Greek-English Lexicon,
liststhe followingsensesforaitia:
I. responsibility;
II. cause;III. occasion,
motive;IV. head,category
underwhicha
292
REASONS AND CAUSES
icance is farwiderthanthatoftheword"cause" as used nowa-

daysbothin ordinaryspeechand philosophicaldiscussions. I can
bestdo so by recallingsomeof the thingsthatcountas straight-
forward aitiai in Aristotle,
whosemetaphysical preconceptionsdid
not blunthis sensitiveness to thevaluesof the wordshe used:
i. Whydid thePersiansinvadeAttica?BecausetheAthenians
had raidedSardis.
2. Whyis thisstatueso heavy?Becauseit is made ofbronze.
3. Whyis he takingafter-dinner walks?Because of his health.
is
4. Why the angle at the semicircle
a rightangle? Becauseit
is equal to thehalfoftworightangles.7
I have deliberatelyavoided the word aitia in formulatingthese

examples,so as to bringout thefactthatto say thatX is theaitia
of r comesto preciselythesamethingas sayingthatrhappened,
ofX. In proofofthis,ifproofit
or happens,or is thecase, because
needs,I need onlyreferto the factthatAristotlespeaksofhisfour
thingcomes;V. caseindispute. In thispaper I shallbe concernedexclusivelywith

senseII, whichthedictionary rendersby"cause"-mistakenlyso,in myopinion,
since,I shall argue,thissensehas a muchwidersignification than thatof the
Englishword as commonlyused nowadays(the sensein whichwe speak ofa
blow as the"cause" oftheshattering ofa vase or ofair pollutionas the"cause"
of pulmonaryirritation),or as employedby philosophers(see, e.g., "Causa-
tion," by R. Taylor,in theEncyclopedia ofPhilosophy,ed. by P. Edwards [New
York, i967]; C. G. Hempelon causal explanationin Aspects ofScientific
Explana-
tion[New York, i965], pp. 347 if.; E. Nagel on causal laws in StructureofScience
[New York, i96i], pp. 73 if.)
7 Examples I, 3 and 4 are fromthediscussion ofaitiaiin thePosteriorAnalytics,
II, I I; examples2 and 3 fromthe chapteron the fouraitiaiin Phys.II, 3. I
have recastthephrasingforobviousreasons,and given a different twistto 2
(Aristotledoes not referto the weightofthe statue,and does notthinkin this
connectionofbronzespecifically as a naturalkind,butonlyas formablematter;
but he would not hesitateto recognizethekindofexplanationillustratedin 2
as a bona fideaitia). I disregardproblemsofAristotelian exegesisraisedby the
factthatnotall ofthefouraitiaiin thePost.Anal.are obviousduplicatesofthe
fourin thePhysics. For mypurposesit is sufficient to note theprominenceof4
in the Post.Anal.,whichleaves no doubt thatAristotlewould recognizeit as
a starexampleofa certainkindofaitia,eventhoughhe findsit hard to squeeze
it into the metaphysicalmould whichdictatesthe classification in the Physics
(and also in the corresponding passage in Metaphysics, V, 2).
293
GREGORY VLASTOS
aitiai as "all the ways of stating &

Sta -L (the because)."8
Aristotle'sso-called four "causes" are his four "becauses."
Now not everybecauserefersto a cause, thoughsome do, as does
the firstexample: the Athenian raid on Sardis would be a fair
example of a temporal antecedent which is the (supposed) suffi-
cient conditionof the occurrenceof an event,the Persian invasion
of Athens. Alternatively,a becausemay referus to an aitia which,
while not itselfa cause, has definitecausal implications. This is
broughtout well enough in the second example. We could hardly
speak of bronze as the "cause" of the weight of a bronze statue:
bronze scarcelycauses itselfto be heavy. What we have here is a
natural kind-that is to say, a clusterof propertiesregularlycon-
joined, among which is its characteristicspecificgravity.Though
the laws of the conjunctionof these propertiesare not themselves
causal, they have a networkof connectionswith causal laws by
means of which we are able to make relevant causal predictions,
such as that a bronze statue will outweighseveral wooden ones of
the same dimensions.
Now consider 3: here it would be not just awkward but posi-
tivelyabsurd to speak of the aitia as the "cause." The health for
which the ailing man submits to his peripatetic chores does not
now exist and may never come to exist,since his walks may not
avail to restoreit and he mighteven die on one of them of a heart
attack; how then could thisnonexistentand perhaps never-to-be-
existentthingcause his walks or anythingelse? To turnthe answer
to our why-question into a statementof a cause we would have to
take a detour in fullview of the intensionalcontextand make the
cause not the health the man expects from his walks but his
expectationof gettingit fromthem, backed up by a strongdesire
to improvehis health (strongerthan forany of the displaced alter-
natives). Some philosophersnowadays would deny us even this
maneuver,holding that it makes no sense to speak of the cause of
an action. Into this controversyI do not propose to enter.9I need
8 K 7&vrw) cbroaorfov T ri followed by a listing of the four aitiai, Phys.

o ta
I 98B5 if.This is one ofthecountlesspassagesin which So To'
tuI S1oT = to acLov
in Aristotle(forsome of themsee H. Bonitz, IndexAristotelicus, I77A50 ff.).
9 The best defenseof the causal account knownto me is by D. Davidson,
40 (i963), 685 if. Cf.
"Actions,Reasons, and Causes," Journalof Philosophy,
294
REASONS AND CAUSES
notin thisinquiry,whereall thatmatters is thatifwe weretalking

Greek,then,regardless ofourphilosophical persuasions, we would
nothave theslightest hesitationin sayingthattheman'shealthis
the aitiaof his walks,whilein Englishthe mostwe could do to
worktheman'shealthintoa causal accountofhisexertions would
be to citethe end-in-view and his desireto attainit as the cause
oftheseactions.
I have leftlastthemathematical example,themoststriking one
formypurposes, forherethegap betweenaitiaand causeisunbridge-
able by any ancillarydevice thatwill standup underexamina-
tion.We are givenP, "theangleat thesemicircle is thehalfoftwo
rightangles,"and Q, "theangleat thesemicircle is a rightangle,"
whereP is the penultimateformulain the stringof formulae
whichmakeup thecurrently acceptableproofofQin thegeometry
of the time.'0This leads Aristotleto take P as the aitia of Q,
construing theentailment of Q by P, alreadyproveda valid con-
sequenceof the axiom-setof the science,as an adequate ground
of the truthof Q. Since thisentailment is forAristotlea relation
whoserelata are abstractitems,he would not dream of saying
thatone ofthesepropositions causestheother.Yet thatis theway
he is made to talkby G. R. G. Murein theOxfordtranslation of
the Posterior by W. D. Ross in his Commentary,
Analytics, and by
countlesstextbooks, wherehe is represented as sayingthat the
premisesof demonstrative inferenceare the "causes" of its con-
clusions.1"Such statements have been defendedat timesby the
claim thatAristotle thinksthe premisescausaecognoscendi.'2 But if
causain thistime-hallowed phrasemeansno morethan "reason"
or "principle"the defenseis otiose,sincereasonsand principles
have no causal efficacy;whileif it does mean cause,the claim is
also M. White,Foundations of HistoricalKnowledge(New York, i965), ch. V,

"Reasons and Causes."
10 See T. Heath, Mathematics inAristotle(Oxford,1949), p. 72.
aLTcOV
CU TOV avIufrEpaqLCaLTTos (7iB22): "related to them as effect to cause,"
Mure in the Oxfordtranslation;"causative of the conclusion,"Tredennick
in the Loeb translation.I have protestedelsewherethisvenerablemistransla-
tion ("Anamnesisin the Meno,"Dialogue,4 [i965], 143 ff.,n. I5). The present
paper is an extendedversionoftheprotest.
12 Leibniz speaks of them as causes "de notre jugement"(NouveauxEssais,
IV 17, 3)
295
GREGORY VLASTOS
false,and may be collapsed by the simple reflectionthat it is not,

in general, true that knowing a given propositionis a sufficient
condition of knowing all of the propositionswhich it entails:13
thus one may know the axioms of a systemand be ignorant of a
whole raftof theoremswhich a cleverermathematicianwould be
able to deduce fromthe same axioms. Aristotle,so far as I know,
nevermade thisparticularblunder. But even ifhe had made it, we
would still have no good warrant for reading it into the many
passages in which he speaks of an entailingpremise as the aitia of
an entailed conclusion, since all such statementsmake perfect
sense if understood to express the simple fact that the premise is
a good and sufficientreason for the conclusion.
This briefdiscussionhas shown,I trust,how much more general
in significationis the Greek aitia than is the currentmeaning of
the word "cause" in English speech. Greek usage would entitleus
as a matter of course to speak of somethingas another's aitia,
regardlessof whetherwe are referring, as in i above, to a straight-
forwardcause, or, as in 2, to a law-like conjunctionof properties
or factorssuch that the instantiationof some of them is a sufficient
condition of the concurrentinstantiationof others,or, as in 3, to
the end-in-viewof a purposefulagent whose desire to attain it by
a certainaction we take to be thesufficient conditionofhisperform-
ing the action, or, finally,as in 4, to one propositionas entailing,
or implying,another, so that our assurance of the truth of the
formerwarrantsequal assurance of the truthof the latter. Here
then is the linguisticbase fromwhich the whole of the discussion
that is to follow will start: the mere fact that Plato speaks of the
Forms as aitiaiin our passage is not ofitselfthe slightestevidence-
not even prima-facieevidence-that he wants them to be causes.
There are plentyof otherthingshe may want to expressby saying,
and with the greatestemphasis, that theyare aitiai. What he does
mean can be decided only fromthecontext.Let us proceed to this.
13 Reason(New York, 1954), pp. 287 ff.

Cf. E. Nagel, Sovereign
296
REASONS AND CAUSES
II. THE "SAFE" BUT "IGNORANT"AITIA
Our passage fallsinto two divisions.DivisionOne (95E-99C)

recountsthe youthful infatuationof the PlatonicSocrates14
with
the physicalphilosophersand the disappointment in which it
endedwhenhe foundthatall theycould offerwas materialaitiai
and mechanicalcauses,whilehe had becomeconvincedthatonly
providesthe"true"(98E) or "real" (99B) aitiaiofnatural
teleology
phenomena.There is no talkof the Formsas aitiai-no mention
ofthemat all-throughoutthewholeofthisdivision,thoughthe
wayis preparedforthemby thelayingout ofa seriesofperplexi-
tieswhosesolutionwouldeludeSocratesuntilhe had hituponthe
TheoryofForms.ThispartofDivisionOne I shalldiscussat some
lengthin due course.The restof it I shall ignore.Though its
historical importance is incalculable-this is one of the great
turningpointsin European natural philosophy,the conscious
abandonmentofthelineofthoughtwhichhad led, in thesystems
of Leucippusand Democritus,to the firstrigorously mechanistic
of
conception the orderof nature-its messageis familiarand,
at least,quite clear. I shall,therefore,
superficially bypassit in
thispaperto concentrateon thefarmorehazardoustaskoffiguring
outwhatis goingon in DivisionTwo (99GC-Io5C), whereSocrates,
in his searchforteleology,fallsback on a second-best
frustrated
methodofinquiryofhisown.15
14 Let me say here once forall that in my opinion the "Socrates" of this
whole passage cannotbe identified in totowiththe historicalfigure,thoughit

may well be thatsome elementsof the accountin DivisionOne would fithis
intellectualbiography,as suggested,e.g., by R. Hackforth,Plato's Phaedo
(Cambridge,1955), pp. 127-131 (a book to whichI shall referhereafter solely
by the author'sname). I shall be using"Socrates" to mean the figurewhom
Plato makeshis mouthpiecein thisaccount,exceptin a fewcases wherethe
contextwillmake it clear thatI am referring to thehistoricalSocrates.
15 "Well, I formypart shouldbe delighted to learnfromanyoneabout this
sort of aitia [that of 'the good and the fitting,'99C5]. But since I have
been denied this aitia (eLfSJ SC' TavrrqgEaTEp?)O1)v) and have failed to either
findit myselfor learn it fromanother,would you like me to show you the
secondjourneywhichI have been pursuingin thesearchforaitia?" (99GC-D2).
Note that the referenceof Tav'rjsq in C8 (the subject of TaTEP-4?qvand also,
with the implied change of case, of the infinitiveselpEZv,paOeFv) is to TotavTqg
aoTras in C7; what Socrateshas failedto discoverby his own labors or from
297
GREGORT VLASTOS
This new method and its distinctiveaitiai are put forwardas the
logical pendants of a philosophical "hypothesis,"'6 that of the
Theory of Forms or Ideas, which is terselyformulatedas follows
in the more complete of the two statementsin our passage:"7
each oftheFormsexistsand it is in virtueofparticipating
in themthat
otherthingsare named afterthem(sc. forForms) [1o2AIo-B2].18
The formula has in view three sets of items and the relation of
"participation":19
(i) Forms-that is to say, those of the full-blowntheoryof

Plato's middle period-presented in this dialogue for the first
time. The very same termswhich he had used to designate the
definiendaof Socrates' moral inquiries in the earlier dialogues-
eidos,idea-he now applies to entitiesendowed with the follow-
thoseofothersand is preparedto do withoutforthepresentis the teleological

aitia itself.This leaves no room forunderstanding him to mean (as has been
done overand overagain in theliterature)thathis"second-best journey"is (a)
an alternativemethod of searchingfor teleological aitiai ratherthan (b) an
alternativemethodofsearchingforaitiai.The textoffers no directsupportfor
(a), since nothingis said of different methodsof lookingforteleologicalaitiai
(the naturalphilosopherswere condemnedforfailingto look forsuch aitiai,
not for lookingfor them by the wrongmethod). There would be indirect
supportfor (a) if we could assume that the earlierreferences to teleological
aitiai as the "true" or "real" aitiai (ras dS &ArO6) acrlasg,98EI; T6 a-rtov Kno
OYU, 99B3) mean thattheseare forPlato not only the preferred (mostfunda-
mental,mostilluminating)explanations ofnatural phenomena (which,of course,
theyare throughout thePlatoniccorpus),but theonlyadmissible aitiaiofanything
whatever. But thereis no case forsuch an assumption;thus Plato would not
hesitateto say withAristotlethatthepremisesofa deductiveargumentconsti-
tutethe aitiaiof its conclusion,withoutimplyingthe semanticabsurditythat
the premisesare the teleological aitia of the conclusion.For the interpretation
I have defendedheresee N. R. Murphy,pp. 145-146; Shorey,WhatPlatoSaid,
p. 534 and thereferences giventhere.
16 i ooB; I 02A I o-B2.
17 The first one occursin IooB5-7. It failsto mention"participation."
18 This formulais practicallyidenticalwiththe one in Parm.130E5-6. For
"named afterthem" (avth65vrovhrcov r,)v Sewvvwu'avaxELv) cf. n. 29 below.

19For thistrichotomy see Hackforth's notesand commentary on i02A-I05B;
R. G. Turnbull,"Aristotle'sDebt to the 'Natural Philosophy'ofthe Phaedo,"
Philosophical Quarterly, 8 (955), 131 fif.;D. Keyt, "The Fallacies in Phaedo
1o2A-107B," Phronesis, 8 (I963), i67 fi. I am particularlyindebtedto Turn-
bull's discussionand regretthatlimitationsofspace denyme the opportunity
to explainwhereand whyI disagreewithcertainfeaturesofhisinterpretation.
298
REASONS AND CAUSES
ing setofcategorialproperties:
theyare immutable,20
incorpo-
real,21 divine;22they cannot be known by means of sense-
but onlyby "recollection."24
experience,23
(2) The individualpersonsand objectsof ordinaryexperi-
ence,designatedby propernamesand definitedescriptions.
(3) The immanent characters
oftheseindividuals,
designated
by adjectives,abstractnouns,and commonnouns.The very
same wordsalso name Forms.This becomesstrikingly clearon
thoserare occasionson whichPlato explicitlyjuxtaposesthe
Form with the cognatecharacterto bringout the fact that,
thoughcloselyconnected,theyare ontologically distinct.He
does so twice in our passage, contrasting"Greatnessitself"
with "greatnessin us" (Io2D),25 and again "the Opposite
itself. . in the nature of things" (1r'v EDrj 9VCYEt) with "the
opposite itself. . . in us" (-r'vJ Wv),26 and both with "the
20 78D (citedin n. 44below); 79D; 8oB.

21 79B. Though theyare not called daco6laTahere,thisis clearlyimplied.Cf.
my commenton thisin "The Third Man Argumentin the Parmenides,"in
R. E. Allen (ed.), Studiesin Plato'sMetaphysics
(London, I965), pp. 231 if.,at
p. 247, n. 2. (This paper appearedoriginallyin thePhilosophical
Review,LXIII
[1954], 319 ff)
22 84A9 (cf.8oBi).
23 65A9-66A8.
24 72E3 if. Withbut a singleexception, not one ofthesepropertieshad ever
been ascribedby Socratesto theuniversalswhosedefinitions he soughtin the
earlydialogues.The exception(called to myattentionby Prof.JohnMalcolm)
occursin the HippiasMajor,whereSocratesspeaksofBeautyas being"always
beautiful"(&Ea KaAO'V,292E2 and 4; cf. 294D); but his point in saying this
in thisearlierdialogue is not metaphysical:he says thisonlyto disqualifyan
absurdlyparochialdefiniens ofbeauty.
25 JcapitalizeForm-naming wordsto distinguish themfromthesame words
used to name characters.Denied this,or any other,inscriptional device,Plato
had to use identicallinguistic tokenstoreferto twodistinctentities, distinguish-
ing theirreference by contextonly.
26 This shows that Plato does not reservethe emphaticuse of the relative
pronoun ("Justiceitself,"etc.) forreferences to the Forms,thoughthisis his
usual practicein his middleperiod.He may use it upon occasion,as he does
here,to referto a character;he had so used it in the earlierdialoguesto refer
to the Socratic universal,as W. D. Ross points out (Plato's Theory of Ideas
[Oxford,1951], p. I 7 and n. i). Commenting on Phaedo103B5,W.J. Verdenius
("Notes on thePhaedo,"Mnemosyne, S. IV, i i [I958], 193 if., at p. 232) says
that here "ro Ev 4'71,v 'vavriov is part of a,)or TO &vaov"; taking the latter
phraseto designatethe PlatonicForm,he infersthat Plato's Formsare both
299
GREGORY VLASTOS
oppositething" (To EVavTtov vrpaiyyta)-thatis, the individual that

has one of two opposite characters (Io3B).27
It will be convenientto use the followingsymbolsin schematic
referenceto these items:
the English capitals, F, G, as character-variables;
theirphoneticcognatesin the Greek alphabet, A, r, as Form-
variables;
a, b, c, as stand-insfornames of individuals, and x as a vari-
able whose values are names of individuals.28
What the theoryassertsthen is the following:forany character,
F, of any individual, x, there exists a homonymous29Form, P;
"immanent"and "transcendent."This would be unobjectionableif it were

meantto bringout thefactthatPlato's theoryin themiddledialoguesprovides
bothforimmanentcharactersand transcendent Forms.But Verdeniusappears
to mean more than this; he seems to deny Plato the ontologicaldistinction
between ro E'v lilv 6vavrov and ro 6vq-? q4'aa&. On this see the next note.
27 Cf. also thecontrastbetween"Similarityitself"and "the similarity which
we possess"in Parm.I 30B. This textbringsout explicitly, whatis implicitin the
Phaedotextsabove, that the word designatinga class 3 item has a referent
whichis ontologicallydistinctfromthe referent of the same word whenused
ofa class i item.It is, therefore, a mistaketo denytheontologicaldistinctness
ofclass3 and class i items,as did Shoreyin commenting on PhaedoI o3B: "there
are reallyonlytwothings:theidea, and theparticularaffected bythe'presence'
ofor 'participation'in the idea.... (The text)does notjustifytheduplication
of the idea, which [a] is a device employedhere only,and [b] withfull con-
sciousness,for the purpose of the argument" (The Unityof Plato's Thought
[Chicago, 1903], n. 283; I have interpolatedthe referencemarks). But [a]
is false,overlookingI02D and Parm.i30B. And [b] is ignoratio elenchi:
the fact
thatthe"duplication"heredoes servethepurposesoftheimmediateargument
in no way impliesthatit is void ofontologicalsignificance.
28 In thepaper citedin n. 2 I above I had notedthenecessity ofdistinguishing
systematically betweentheadjective,large,and theForm,Largeness,in analyz-
ing the regressargumentsin Plato's Parmenides. I there used "F-ness" as a
Form-variable.In thispaper I turnto the Greekalphabetonlyforreasonsof
typographicaleconomy.Were it not forthis,I would stillpreferto write"F-
ness" forthe Form-variablecorresponding to the character-variable"F": the
recurrence of"F" in bothsymbolsbringsout moreforcefully thelinguisticlink
between "Largenesss" and "large" and the ontologicalbond betweenthe
entitiestheydenote.
29 Wherex is characterized by F Plato speaksofx as "named" after.P, its
"namesake" (cf.Phaedo78E, Parm.133D, Tim.52A), interpreting thepredica-
tive statement,"x is F," as "naming" x "F"; cf. the formulain Rep. 596A,
"we are accustomedto posita singleFormforeach pluralityto whichwe apply
thesame name."
300
REASONS AND CAUSES
and x is F (that is, x has the character,F) if,and only if,x partici-
pates in 4P. "Participation" here designatesthat one-way relation
of ontological dependence between temporal things and eternal
Forms which is so fundamental a tenet of this philosophy. For
Plato nothingcould existin space and timewitha definitecharac-
ter,F, if theredid not exista correspondingA, while the converse
would not be true at all. The existenceof a specificForm, say, of
a chiliagon,would of itselfnot offerthe slightestassurance of its
physical instantiation;not only the Form of the Ideal City (Rep.
592AB), but infinitely many otherForms as well existwhich have
been uninstantiatedsince time began and may so remain forever
in Plato's universe.So much of his intentionis clear enough. But
if we probe further,pressing him to tell us just what it is that
happens when a particularF achieves the required "participation"
in a 0, Plato has no definiteanswer forus, and he is well aware of
this fact. He makes no effortto conceal fromthe reader that he
has yet to reach a clear-cut conception of what "participation"
involves,speaking of the relation of Beauty to beautifulthingsas
"presence (wcapovuca) or association (Kotvcvia) or whateverbe
the right word for it" (IooD5-6).30 Here is somethingPlato has
not yet cleared up to his satisfaction,though he doubtless expects
he will, remaining quite certain for the present that some such
relation exists and that, were it not for this, the fact that things
have characterswould be unintelligible.31
30The above translationproceeds on the assumptionthat Wyttenbach's
emendationof vrpoayevojuEVr7in the MS. to 7rpoaayopevojuEv7) (which appears
to have confirmation in a papyrus:cf. Hackforth'snote ad loc.) is correct.If
we stickto the MS. reading (cf. R. S. Bluck,Plato'sPhaedo[London, I955],
ad loc.) or accept Cornford'semendation (p. 77, n. I) to 7rpoayEvojuCvov, the
last clause in theabove citationwould read "in whatevermannerit may come
about" or "whatevertherelationship maybe." The difference willnotbe great
in eithercase, and will not affectat all the importantthingin thecitation,se.
theavowal of uncertaintyin ELTE &r on Kai orr. For Plato's use of ZiLe T7r?
(augmentedin thethirdexampleby sore Ogres)to avow uncertainty, see Parm.
i83D, Tra vap' q'jzrvotolucpairaETE,
7vre -Tcrravirad 1WEIrat, 7im.48C, EtTEacpx?)EbTe
apX&aelTE0'7m SOKEV Tep' TOcv~V,and Laws 899B, eoreCv a/aam evovaat . .. EL. 07er
E
TE Kal o0ws'.
Though theexpectationwas neveradequatelyfulfilled,

"1 Plato retainedthe
confidencethatsomehowor otherthings'must"participate"in theForms.In
at theend ofthesecondregressargument,Parmenidesdoes not
theParmenides,
concludethatthe notionofparticipationhas been invalidated,but onlythat
30I
GREGORY VLASTOS
Armedwiththis"hypothesis," Socratesfeelsempoweredto give

two complementary answersto the question,"Why is x F?"32
Let us beginwiththefirst, whichhe calls the "safe" but "simple-
minded"and "ignorant"(iooDE; 1o5C) aitia.Thisisjust thatx is
F because it participatesin 0.33 What could he mean by that?
Two interpretations have been advanced, which I take to be
mistaken.
On one ofthem,theFormwouldbe a teleologicalaitia.34I fail
to see how thiscould be squaredwiththefollowing featureofour
passage:Socratesmakesit abundantlyclear thathe is still,at the
timeofspeaking,"deprived"oftheteleologicalaitiahe had been
lookingfor.35But it is no less clear that the alternativeline of
investigationhe is aboutto explain-the "secondjourney"-is not
something thathas poppedintohishead atjust thatmoment;it is
a methodof inquiryon whichhe has beenalready engaged.36This
methodtakesitsstartingpointfromthe hypothesis ofthe Forms.
Hence,ifSocrateshad thoughtofthe Formsas teleologicalaitiai,
"we mustlookforsomeotherway [i.e., a way otherthansimilitude]by which

theyparticipate"(133A5-6).
32 Also, "Why does x come to be F?" which is the implied questionin
IoIC2-7: "you would loudlyprotestthatyou do notknowhow else each thing
in the peculiar
comes to be [F] exceptby coming to participate(ptEcaaxo'v)
essence (-ns gaasaovialag) of that [Form] in which it comes to participate
(EKaacov ov av peiraax7) and so here you have no aitia of their coming to be
two except their coming to participate (eracaXeawv) in the Dyad-in this all
thingsmustcome to participate(pe-aaXeZv),if theyare going to be two-and
in the Unit if they are going to be one" (IOIC2-7). For the translation
of ,IE-raaxo'v /IEzaal2) ,ueaxfa, /ETaUXElv, cf. Cornford ad Parm. I 29A3
(p. 69, n. i): "As in the Phaedo,/IeraAa/4laivetv(peta'aXeamg,Phaedo ioiC,
to partakewhen the thingbecomeslike
peLEaA)lbssParm.I 3 iA) meansbeginning
(yLyvEaOa&),whereas Iuexev is used of having a share and corresponds to
beinglike (elvaL)"; "beginning," however, should be corrected to "coming"
forobviousreasons(Cornforddoes not even use it in his own translation!).
33 Whichmay be abbreviatedto "x is F because of 4P,"as, e.g., at ioiD7-8,
TW Ta KaAa KaAa.
KaA(Pvrav'Ta
34 See, e.g., Taylor, p. 203 and n. 2; Bluck,op. cit.,p. i99; Crombie,pp.
159 ff.The "confusions"whichCrombiefindsin thepassagearisein partfrom
his assumptionthat "it is apparentlytaken forgranted that whereverthis
[i.e., a formalaitia] is achievedsomethinglike a teleologicalexplanationwill
be forthcoming" (p. I59; "apparentlytakenforgranted"concedesthatnothing
ofthekindis said, or distinctly implied,in the text).
35 Cf. n. I5 above.
38 Note thepreterites: ixrepay,aEV/IaLin 99Di and I ooB4, Jp/nraa in iooA3.
302
REASONS AND CAUSES
he would not have said thathe is still"deprived"ofteleological

aitiai.He would have said thathe does have them,thoughonly
on the basis of a hypothesis. On thisgroundthe suggestedinter-
pretationmustbe rejectedas not onlyunsupported by thetext-
thereis no mentionof teleologyafterthispointin our passage-
but as contrary to theunambiguousimplications ofour text.It is,
therefore, unnecessary to inquirehow Plato could have assigned,
withoutgraveconfusion, to his Forms-entitieswhosemostcon-
spicuousfeatureis theirabsoluteimmutability-theteleological
function which,bothin thisdialogueand in the Timaeus, pertains
exclusively to mindor soul.37
A secondinterpretation deservesa little-if onlylittle-more
consideration. This is thattheFormis meantto taketheplace of
Aristotle's"efficientcause." So Aristotlehimselfexpoundedour
passage, complaining that "Socratesin the Phaedo"thoughtthe
Forms"sufficient aitiaofgeneration,"38 thoughit is hardto know
just how seriouslyhe tookthisreading,forit is clear fromother
remarks ofhis39thathe knowsquitewellthatthiscannotbe Plato's
doctrine.40 Echoes of thesame readinghave recurredin modern
37The sole patternofteleologicalexplanationenvisagedin thePhaedois that

exemplified in thepurposefulagencyofa mind (thecosmicnousofAnaxagoras,
97CI ff.;the mind of Socrates,98C2 if.). In the Timaeusthosefeaturesof the
cosmos which admit of teleologicalexplanationare exclusivelythosewhich
are imputeddirectlyto the activityof divinesouls: the Demiurge (46C-E et
passim)and "his offspring" (69C ff.).And cf.n. 45 below.
38 De Gen.etCorr. 335B9-i6: "But one partyhas thoughttheFormssufficient
aitia ofgeneration,as did Socratesin thePhaedo:forhe too, afterreproaching
othersfor having explained nothing,hypothesizesthat some existentsare
Forms,othersparticipantsin the Forms,and thateach thingis said to be in
virtueoftheForm,and tobecomein virtueofparticipating [intheForm]and to
perishin virtueofshedding[theForm]. Hence, ifthisis true,he must believe
the Formsaitiai both of becomingand of perishing."Cf. also Met.99iB3-4:
"In thePhaedothematteris put thus:The Formsare aitiaibothofbeingand of
becoming."
39 "It is evidentfrom whathas beensaid thathe has used onlytwoaitiai:that
of the essence,and the materialaitia" (Met. 988A8-Io).
40 In spiteofwhathe saysin thepassagesquotedin n. 38 above,whathe most
likelymeansin the light of the statementquoted in n. ;9 is not that Plato
himselftaughtthattheFormsare efficient causes,but thatsince (in Aristotle's
opinion)Plato had made no provisionforefficient causesas such (Met.992A25-
26, "For we [Platonists]say nothingabout the aitia which is the source of
change"-i.e., the "efficient" aitia),Plato wouldhaveto invoketheFormsto do
303
GREGORYVLASTOS
scholarship in spiteofShorey'svigorousprotests.Thus Hackforth

(p. 144) claimedthattheIdeas are meantto be causesof"qualities
ofconcretethings,"ifnot ofconcretethingsthemselves: "Beauty
itselfis not the cause of a beautifulthing,but of a thing'sbeing
beautiful."Whatsensecould be made ofthisHackforth does not
seemto have considered.If he had done so, I doubtifhe would
have ever committedhis interpretation to print.For since all
Formsare absolutelyfreeof spatiotemporal limitations,then if
one of themwere supposedto be actingon a particularspatio-
temporalobject,a, witha determinate property, P, we wouldhave
to suppose(i) thatit is also actingon all otherobjectsin theuni-
verse,includingthosewhichdo not have the property, P, and,
further (ii) thatall otherForms,includingFormscorresponding to
propertiescontraryto P, are simultaneously actingon a. How
then(i) could thegivenFormhave thatspecificcausal effect on a
whichwould accountforits beingP ratherthannot-P,without
havingthesame effect on all otherobjects,includingthosewhich
are not-P?And how (ii) could it have any determinate effecton
a at all, if all thoseotherFormsare simultaneously actingon a
withcontraryeffect?The only way to avoid the absurd conse-
quences of the suppositionwould be to creditFormswith the
power to act selectivelyon different objects in the universe,
directingtheircausal influenceto some of them,withholding it
fromothers.And howcouldPlatohaveso particularized hisForms
as causal agentsin the worldof space and timewithoutfouling
up the mostfundamental of his metaphysicalprinciples
?4- Only
the mostdirectand explicitevidencecould persuadeus thathe
thisjob in additionto thatof the "formal"aitia. On the chargeagainstPlato

in Met. 992A25-26,see H. Cherniss,Aristotle's Criticism
of Plato andtheEarly
Academy (Baltimore,I944), pp. 383 if.
41 See Cherniss, op.cit.,p. 452 and n. 397, on thepropositionthat "the ideas
themselvesare nevermade productiveagents" in Plato's philosophy.In view
of this interpretation, which Chernissmaintainsforcefully and consistently
throughout thiswork,he mighthave done betterto avoid speakingoftheIdeas
as "causes" of their"approximations"in the sensibleworld (p. 2I8), "of the
particular'sbeing" (p. 373), "of thatwhichin particularsis similarwhenever
and whereverit occurs" (p. 375). He makesit quite clear, of course,thatby
sayingthat the Ideas are "causes" he does notmean that theyhave causal
efficacy;what thenremainsunclearisjust what is meant,whenthisis not,to
justifythe use of the term"cause" in thisconnection.
304
REASONS AND CAUSES
blunderedso grossly.And thereis no suchevidence.All we have

to go on is thefactthathe usesthesamelanguageoftherelation
oftheFormto a thingwhichhe would also have used ifhe were
speakingoftherelationofa cause to itseffect:P is that"because
of which" (& ' or, iooDi) x is F;42 it is that which"makes"
(ITQoLE, iooD5) x to be F; itis theaitiaofx's beingF (forexample,
IooC6-7; IOIC4-5). Is there no other way of construingthese
statements thatwill make bettersense?
Considerthefollowing exchange:"Whyis thisfigurea square?"
"Becauseit has fourequal sidesandfourequal angles.If it had just
thefourequal sidesthatwouldnotmakeit a square; it couldhave
been,forall that,a rhombus."43 Here itis clearthatthe"because"
whichanswersour "Why?" is not meantto explainthe occur-
rence of a squareshapedchalk mark on our blackboard.The
occurrence is presupposed is takenin itscause.Our
and no interest
questionis not "What made that chalkmark?" nor yet "What
made that chalk mark square?" but rather "What makesit
square?" which could only mean in this context:why do we
classifyit as a square,ratherthanas a figureofsomeothershape?
Our questionis answeredwhenwe are shownthatthechalkmark
happens tohave-nothow or whyit happened toget-the shape that
meetsthe logical conditionsforbeing square. The aitia we are
givenis a logicalone.
At firstsightthismay seem suspiciously deflationaryof what
is said in our text.When Socratesmaintainswithsuch dogged
emphasisthata beautifulthing"is beautifulforno otherreason
42 The "because" can also be expressedby the "instrumental"dative: r47

KaAi iooD7, E2; eyeEt i ooE5, etc.
43 Compare: "Those characteristicsthatare indispensableto an act's being
just are the characteristicswhichmake the act a just act," Susan Stebbing,
A Modern toLogic,2nd ed. (London, I933), p. 429. "To teachwhat
Introduction
makes a member of any class a good member of the class . .. ," R. M. Hare,
TheLanguageofMorals(Oxford,1952), p. i02. We findsimilarusesof"makes"
with transparently logical importin Plato's earlierdialogues: Charmidesis
asked to consider"what sortof man temperancemakesyou by beingpresent
in you" (67rolo'vnva ae Grotve - awppoavv-q 7rapoviaa) and answer accordingly
the question,"What is temperance?"(Charm.i6oD6-7). Hippias agreesthat
"that which is fittingfor each thing is what makes provedd)each thing beauti-
of "beautiful"
ful," where"fitting"is being consideredas a possibledefiniens
(Hipp.Maj. 29oD).
305
GREGORr VLASTOS
thanbecauseit participates in Beauty"(IooC5-6), he is certainly

puttingforwarda thesiswhichcould not be reducedwithany
plausibilityto thelogicaltruththata particularthinginstantiates
a conceptif, and only if, it satisfiesthe definition.But such a
reductionis the last thingI would wish to suggest.To do so I
wouldhave to arguethatforPlato logicis a metaphysically non-
committalbusiness;and who would want to say such a thing
on his behalf?What is his TheoryofFormsifnot the claim that
logical statements presupposemetaphysicalones and would be
mumbo-jumbo withoutthem?For Platothedefinition ofa concept
is "theaccountoftheessence"ofitsForm, P.44The reasonwe can
speak significantlyand trulyof thingsbeingsquare or beautiful,
he would insist,is that thereexistsan incorporeal,immutable,
intelligibleobject,named "Squareness" or "Beauty," in which
corporeal,mutable,sensibleobjects occasionally"participate"
and, when theydo, are rightlycalled "square" or "beautiful."
So what I have called the "logical" aitia is at the same timea
metaphysicalone forPlato; the logical functionof Squareness,
Beauty,and so forth,he is convinced,could not be discharged
aside fromtheirmetaphysical status.But once thatis granted,it
is the logical functionof the metaphysicalentitythat does the
explanatoryworkof the "safe" aitia. When I want to knowwhat
makesthisfigurea squareratherthana pentagon,whatanswers
my questionis not the existenceas such of the Form, Square-
44 "That realityitself(avtrn n ov'ata), of whose essencewe give the account
(f; Ao'yov&tolmv To6 etval) when we ask and answerour questions,is it ever
invariablythesame or does it vary?Equalityitself,Beautyitself,each 'whatis'
in itself,the reality(avror &aaaTov S carnv, TrOov), does it ever admit the least
alteration?"(Phaedo78DI-5). Burnetrightlyremarksad. loc.in his Commen-
tarythat"we musttakeAo'yov Toi dvat together as equivalentto Aoyovng ovrala
or 'definition'(comparingRep. 534B3, S&aAElKOV KaAcis TOV AOyov EKaTaTov
rqs ovalas) and as governing
Aacqgaivovrra the genitive4w." Cf. also Rep.533A8-B2,
and also Laws895DE, wherethedistinctionis drawn betweenthe name, the
ovaia it names,and the AMyog of this ovtata. It is a mistake to suppose that
thereis any Formofwhichthereis no Ao'yog.The view that in Symp.2 I iA it is
said or impliedthatoftheFormof Beauty "there is no logosnor knowledge"
(R. C. Cross,"Logosand Formsin Plato," Mind [63], 433 ff.,at p. 443) is
not warrantedby the text,which only says that Beauty "will not appear"
(ae'Sc pavraaOXc7asta)as logosor knowledge(sinceit will appear as thatofwhich
we have logosand knowledge).
306
REASONS AND CAUSES
countlessotherFormsalso existwhichdo not help to answermy

question-but the logical contentof its definition:thisis what
marksoffthe Form, Square, fromall theseotherForms and,
isomorphically, marksoffeverysquarein ourworldfrominstances
of all the otherfigures.And the factthatthislogicalfunctionis
performed by a celestialForm,ratherthanby a nominalisticftatus
vocisin no way altersthestrictly noncausalimportoftheformula
"F in virtueof satisfying the definition."Plato's Squarenesshas
no morecausal efficacy thanhas thenominalist's; it has no power
to spawnearthlysquares;ifit did,so wouldtheForm,Myriagon,
and each of the countlessothersthat have had no mundane
progenyand neverwill.45
Thisinterpretation ofthe"F in virtueofI" formulafreesPlato
fromso muchembarrassment and is so consonantwitheverything
elsewe knowofhismetaphysical viewsthatit wouldhavea strong
claimon us evenwithoutfurther confirmation.6 In pointoffact
we do getconfirmation foritfromtwodistinct data in ourpassage.
45Note thatwhen Plato says in the Timaeusthat the Ideal Model may be
"likened"to the "father"ofgeneration,theReceptacleto the "mother,"and
the thingsthatcomposeour world to the "offspring"(5oD2-4), he makes it
veryclear thathe assignsno causal functionto the Ideas in respectofeitherof
the two kindsof causality(teleologicaland mechanical)whichhe recognizes
in that dialogue (46C7 if.). The metaphoricalremarksin Rep. 5o6E, 507A,
and 5o8B, about the sun as the "offspring" of the Idea of the Good mustbe
interpreted in the lightofwhat Plato meanswhenhe saysin the Timaeusthat
the whole of the naturaluniverse,notjust the sun, is the "offspring" of the
Ideas generally,notjust oftheIdea ofthe Good. In the Timaeusthemetaphor
is employedin thecontextofa cosmologicalschemewhichenablesus to control
the intendedmeaningin a way whichis altogetherdenied us by the allusive
unexplicateduse ofthemetaphorin BookVI oftheRepublic. Whenthe"father"
metaphoris used in the Timaeusin a contextwhich makes it clear that the
metaphordoes express(teleological)causal agency,it is applied not to the
Forms,but to the Demiurge(28C3-4) in contradistinction to the Forms.
46 A numberof commentators entertainand, to all appearances,endorse
a substantially onlyto shyaway fromit a page or two
identicalinterpretation,
later,or even a line or two later. To recountthesevagariesin detail would
requirea special monograph(a rathertediousone). A singleillustration must
suffice.Cornford(in the workmentionedin n. 5, above) beginson p. 77 with
an impeccableglosson IooC4-6: "thefactthatthisroseis beautifulis thesame
thingas thefactthatthisrosepartakesofBeauty.We learnnothingabout any
causewhichwould bringthatfactintoexistence."But see what happenswhen
he proceeds(immediately)to iooDi-8, wherePlato, havingspokenthreelines
earlierofBeautyas (a) thereasonwhy (Sd&o)x is beautiful,goes on to speak
307
GREGORT VLASTOS
In thefirstplace,itmakesgoodsenseofthefactthatthisformula
is proposedas the "safe" but "ignorant"or "simple-minded"47
aitia. This is what it would obviouslybe for anyone who has
alreadyaccepted the metaphysical"hypothesis"on which this
aitia is so explicitly forall x, x is F
pegged.48On thishypothesis,
of it also as (b) what makes (proses)x beautifuland to say also (c) that x is
beautifulbecauseof (ace KaAli: instrumental dative; cf. n. 42, above) Beaut,
Cornfordbecomesgreatlyexercisedover the use of "makes" and wonders:
"Does it [the 'makes'] mean that the thing'sbeauty simplyconsists in the
presenceeitherof the Form itselfor of the characterlike thatof the Form,as
we saythatthepresenceofa gay colour'makes'thethinggay? Or does itmean
thatthe Form,existingindependently, causesthe thingto be (or to become)
beautifulbysomehowimparting itsowncharacterto thething?This isprecisely
thedilemmaon whichSocratesrefusesto pronounce.The languagemightbe
expresslydesignedto leave it unresolved"(italics are Cornford's).Now (b),
"makes," could not have causal importunless(c), "because of," did, sincethe
latteris usedtosaythesame thingin IooD7 as was said bytheuse of"makes" at
D4-5; and thisin turnis thesame thingas was expressedby the use of (a), 8V'
f&fin Di, theveryexpressionused at C5, where Cornfordwas certainthatit
had no causal import;how thencould he be leftuncertainas to theimportof
"makes" at IooD5? Moreover,any uncertainty on thisscorecould have been
resolvedby notingthat the "makes" is used in Socraticdialogues (cf. n. 43)
whereCornford wouldnotthinkofreadingcausal importintoit.That Cornford
himselfcannot be takingvery seriouslythe "dilemma" on which Socrates
supposedly"refusesto pronounce"in iooDi-8, appears on p. 79, wherehe
talksas thoughthe supposed"dilemma" has been firmlyresolvedin favorof
itsnoncausalhorn.He remarkson "Simmias comesto partakeofTallness" as
the Platonicanalysisof "Simmias becomestall": "This is a descriptionof the
same eventin otherwords.Nothingis said as to any 'cause' in our sense,which
would makesuch an eventtakeplace as itseffect."But he backslidesagain on
p. 8o in glossingIo3D (where,e.g., snow perishesat the approach of heat):
"Socratesseemstobe unawarethatonlytheefficient cause ofchangehe actually
describesis a physicalcause of preciselythe kindwhich,in the account ofhis
youthfulexperiences, If "comingto partakeof
he had rejectedas unsatisfying."
Tallness" has no causal import,then Socrateswould not have the slightest
reasonforbeing"unaware" oftheexistenceofphysicalcausesofbecomingtall;
whythenshouldhe be "unaware" ofsuchcausesin thecase ofa thingceasingto
be snow (i.e., melting)and becominghot whenit comesto partake ofHeat?
47 It is safe: iooD8, Ei; IoIDI-3; I051B7-CI. It is "simple, artless,and
perhaps simple-minded (or stupid, ev7cos)" IooD3-4; "ignorant,"Io5C.
48 Plato,ofcourse,would notapplythesame description to the "hypothesis"
-i.e., to the TheoryofIdeas. Obviouslyhe would notthinkofthisas "simple-
minded,"noryetas havingthekindof"safety"he is now talkingabout,since
he presentsit as a "hypothesis"and refrains fromclaimingthathe has proved
it conclusivelyin thisdialogue (which,of course,does not implythathe does
not thinkit susceptibleof conclusiveproof:in the exchangewithSimmiasat
308
REASONS AND CAUSES
if,and onlyif,x participatesin (D. From thisit followswiththe

"'safety"ofanalyticinference thata, or b,or c, or anyotherx, is F
in virtueofparticipating in A. Forjust thesame reasonthisaitia
is "ignorant"- that is, uninformative.49 Not only does it not
professto giveus theslightest helpin finding out thecause ofany
specifichappeningin the world; it does not even aid us in dis-
coveringits correctdescription:only if we alreadyknow that
something is F (thatis,ifwe have alreadyso describedit) can we
proceedon thestrength ofthisaitia to say thatit is F in virtueof
1, showingthatourdescription matchesthedefinition and thereby
clinchingthe proprietyof calling it "F" ratherthan "G" or
something else.
Secondly,theinterpretation I have offered makesgood senseof
anotherthingin ourpassagetowhichI haveyettomakereference:
the fact that the formula"F in virtueof participating in P" is
expectedto resolvethe puzzlesabout aitia whichweredisplayed
in DivisionOne ofourpassage.50The first fourofthese(96D8-E4)
Io7A8-B9SocratesconcedesthattheTheorycallsforfurther
investigation,but
makesno avowal of uncertainty,
as was suggestedbyJowett'sinfluentialbut
gratuitousrendering of KaC' 0aov SvvaTov d'Ata-r' &vOpdiry by"witha sortof
confidence
hesitating in humanreason"nowhappilycorrected to "as faras
humanly possible"in the4threv.ed.,Oxford, I 953). The"safety"ofthepres-
ent aitia is due to its beingso immediatea consequenceof the "hypothesis"
thatonceyouacceptedthelatter youwould risknothing
furtherin maintaining
thisaitia.
49 ThisisnodoubtwhatShorey hadinmindwhenhespokeofthe"tautolog-
icallogic"ofthetheory thattheIdeasareaitiai(cf.thecitation in theopening
paragraph ofthispaperandthereferences inn. 3). Unhappilyhefailedtonote
thatthiscouldonlybe saidofthe"safe"aitia,andnotat all ofthe"cleverer"
one (to be discussedin thenextsectionofthispaper)whichis conspicuously
nontautological inform, thisbeingtheveryreasonwhyitiscalledthe"clever"
aitia.Evenin thecase ofthe"safe"aitiaa certainqualification is necessary
(cf.n. 78 below)ofwhichShoreytookno notice.
50 Theyarestated in 96D8-97B3, and resolved in IooE5-B7.I do notlump
96C2-D6withthepuzzles:thebelief thata mangrows bytheintakeoffoodand
drinkinvolves no absurdity whichneedsto be clearedup by themachinery
ofthe"safe"aitia:itwillbenoticed thatwhentheclearing up is donein iooE5
ff.thereisnomention orallusionto thebelief thatwe growbytakingin food
anddrink.The pointof96C2-D6issurely toillustrate wayinwhichan
another
infatuationwiththemethodology ofthenatural coulddoa tyro
philosophers far
moreharmthangood:itcouldlead himto"unlearn"(96C6)familiar truths,
persuading himthathe knewonlythosecauseswhichhe couldformulate in
thecategories ofan elaborate physicalsystem andhencetospurnexplanations
309
GREGORY VLASTOS
are peculiarlymystifying to the modernreader.5'He is likelyto

findthem not so much puzzles as meta-puzzles:what perplexes
himis whatthereis in anyofthemthatPlato couldhave thought
perplexing.The firstpair (98D8-Ei) look almost like spoofs.
Socratesis supposedto havebelievedin hisbenighted youththatif
one man overtopsanotherby a head, he does so because ofa head,
if
and that the same is true of horses: one horseis tallerthan
anotherby a head, theaitia ofitssuperiorheightis-the head!52
The nextpair (EI-4), on the contrary, look like commonplaces:
Socratesthoughtat thattimethattenthingsare morethaneight
becausethereare twomoreofthemin theten (thanin theeight);
and that,giventwoobjects,twoand one yardslong,respectively,
thefirstwillbe thelongerbecause it exceedsthesecondby halfits
ownlength.One wondershowanyone,no matterhowyoungand
callow,could have been expectedto swallowthe absurditiesin
the firstpair, or blamedforcountenancing the platitudesin the
second.
Lightdawnsin thefollowing paragraph,wherewe comeupon
and a sixthpuzzle: Whydoes i, added to i, make2 (96E6-
a fifth
97A5)? Why does i, dividedby 2, make 2 [halves](97A5-B3)?
Here themode ofpresentation changes:53Instead of recounting,
oftheusual kindwhichare castin homely,everydaylanguage,yetare farmore
worthyof credencethan the windy theorizingswhich the physiologoi would
have put in theirplace. (Thereis no foundation forBurnet'ssuggestionad locum
[apparentlyswallowedby Hackforth,p. I31] that "Socrates means that his
former beliefswereupsetby thequestionofAnaxagoras[fr.iO], 'forhow could
haircomefromthatwhichis nothairand fleshfromthatwhichis notflesh?'"
Anaxagoras' highlyspeculativeanswer to this question,if true, would not
"upset" but onlyaccount forthe beliefin 96C8-D5 thata man can increasehis
bulk by eatingand drinking,"when fleshwas added to flesh,bone to bone,"
etc.)
51For manyyearsI could not make head or tail of thesepuzzles,and used
to takethemas symptomatic ofsomekindofmuddlein Plato'sthinking, though
withoutsharingCrombie'sconfidencethat it was the kind of muddle that a
tyroin philosophycould diagnose.It was onlywhenI reachedtheinterpreta-
tion of the "safe" aitia which I presentin thispaper and saw what kind of
solutionit wouldprovideto thesepuzzles thatI began to see whatgave riseto
theirperplexitiesin the firstplace.
52 On theface ofit, thisis a low-gradepun: the dative, Vn KE9paAJ-,
could be
used to mean both"by a head" and "because ofa head."
58 The changein toneand perspective is markedby Cebes' question,imme-
diatelyafterthepreviouspuzzleshad been laid out: "And how do you thinkof
310
REASONS AND CAUSES
poker-faced,mistakesincurredin his distantyouthwhenhe had

been "utterlyblinded" (96E5) by his obsessiveattachmentto
mechanisticaetiology,Socratesnow refersto similarerrorsfrom
hispresent,entirely different
philosophicalperspective,
and gives
us broad hintsas to whyhe can no longerstomach54 theanswers
his previous"method"of thinking had made compelling.Thus,
theold answerto thequestionin thefifthpuzzlewouldhave been:
theadditionofone unitto theotheris whatmakestwoofthem.55
This, he now says,he no longeraccepts,and forthe following
reason:
For I wouldbe astounded
if,wheneachofthemwas apartfromthe
other (o7re IEv EKaTepovawrcovXcpts-aA74Awv
iv), each was one
and theywerenot thentwo,but aftertheyapproachedeach other
(E&rAr)crIacav this became the aitia of theirbecomingtwo:
dAAX4Aots-)
the conjunction in theirbeingputcloseto each
involved
(7quv'vobos-)
other(rov31rTArgaov -rEOWvat)
aAA4Acwv [97A2-5].
Here at lastwe see thegaffeSocrateshad beenperpetratingin that

periodwhichantedatedhis discoveryof the "safe" aitia. He had
been confusingthe arithmeticaloperationof addition with a
physicalprocess-thatof takingthingswhichwere "apart" to
beginwithand puttingthem"closeto each other."56 And he had
been supposingthatthefeasibility ofthismaterialprocesswas the
thesethingsnow?" (96E5). ThereafterSocrates' verbschange fromthe past

tensein D8-E4 to thepresent.
54 "I do not accept," 96E7; "I would be [literally,'am'] astoundedif,"
97A2; etc.,windingup at B7 with"I will have no truckwiththis[methodof

thinkingwhichgivesriseto thepuzzles]" (TOiTOV SC' ov'Saatf wpoaLcaL).
55 6E8-97Az. To simplify the expositionI abbreviatethispuzzle whichin
the text (Burnet's,with his expansion)presentsthe resultof adding A to B
as a disjunction:eitherB becomestwo,or A becomestwo,or A and B taken
togetherbecome two. The last of the threedisjunctssuffices formy purposes
above.
56 The literal meaning of the verb used for addition, wpoaTIO77,4t A i-4 B,
is literally"put A nextto B." This dead metaphorcomesalive in thepassage,
Socratesshiftingback and forthfromthisverb to variantexpressionswhich
referunmistakablyto puttingobjectsclose to one another,whilehe denotes
theconverseoperationnot by the usual termsfor arithmeticalsubtraction
(aQpatpeats)or division (taipeapsa), but by words which have strong physical
evocations: "splitting" (a&.ats, A7; cf. &taaXlaet, A6), "leading apart"
(dwayeTat, B3), and "separating" (Xwpicpeat, B3).
3II
GREGORY VLASTOS
aitiaofthe logico-mathematical truththatthe same itemswhich

count as units,if taken disjointly,will count as a pair, if taken
conjointly.Lookingback at thisbonerfromhispresentphilosoph-
ical perspective,Socratessayshe would be "astounded"if such
a thingweretrue:he wouldsuffer thatpeculiarsenseofintellectual
outragewe all feelwhenasked to believea proposition whichis
notjust materiallyfalsebut logically absurd. For obviously, the
thingsbeingtalkedabout are twoby hypothesis, and theywould
stillbe tworegardless of whethertheywerejammed up together
in a cupboardor situatedin different galaxiesa millionlight-years
apart. How absurdthento offertheirpropinquity as the reason
'
whytheyare two! So the puzzle can now be solvedor, more
exactly,dissolved,Plato's solventbeing the "F in virtueof P"
formula. 58If thingsare one in virtueofparticipatingin Unity,two
57This is reinforced by a furtherobjection.If we wereto taketwo-nessas the

effectofconjunctionwe would be facedwiththe (supposed)paradox thatthe
oppositeprocessofdisjunctioncauses thesame effect.Socratesis goingon the
assumptionthatifa givenprocesscauses a certaineffect, the oppositeprocess
could notalso cause thesame effect. There is a fallacyhere,but apparentlynot
an obviousone,forI have seenno noticeofit in theliterature.Even ifwe were
to concede the truthof the assumption,the conclusionwould followonlyifit
were truethatthe disjunctionand conjunctionofthesameitemsproducedthe
same effect. But thelatterwould notbe truein thetwocasesSocratesis discuss-
ing: he getstwo-nessin thefirstcase by conjoiningunitsA and B; he getsit in
the second by disjoining,not the same units,but parts inside each of them.
This fallacydoes not invalidate the fundamentalinsightI expoundabove,
however;thiscan dispenseentirelywiththisadditionalsupport,which,as it
happens,is unsound.
58 This becomes clear in Division Two of our passage where (:ooE5-B7)
all ofthepuzzles are resolvedseriatimby applyingto each the "F in virtueof
V" formula.This fundamentalpoint,whichshould be made the pivot ofthe
interpretation of the puzzles, is not even mentionedby Hackforth(p. I31):
he givesno indicationthatthepuzzle laid out in 96E6 ff.is stuffed withhints
of the correctsolutionand that in Plato's opinion the "F in virtueof 0"
formulaprovidesthecorrectsolution.So it is hardlysurprising thatHackforth
should labor under the misapprehensionthat Plato is himselftaken in by
"unreal problems" (cf. the citationfromHackforthand my commentin the
concludingparagraphof thissection).Crombietoo thinksthatPlato wallows
in theveryconfusions whichthe "safe" aitia is meantto clear up. Thus he says
that "phrases like 'the puttingof one alongsideone is not the cause of the
occurrenceof2' (ioIB9-CI) are used withoutany clearindicationwhetherthe
questionis: (a) 'Why are theretwothingshere?' (to whichan answerin terms
ofputtingone thingalongsideanotherwould be appropriate);or whetherthe
questionis: (b) 'How does thenumber2 arise?'" (p. i69; I have interpolated
312
REASONS AND CAUSES
in virtueofparticipatingin theDyad, thenit willbe clearthatthe

"Why?" in "Why do and i make 2?" cannotbe a physical
i
"Why?" and thatitsanswering"because" mustbe extractednot
fromaccountsof what happensto objectswhentheyare moved
about,but from"accountsoftheessence"59 ofthenumbers,One
and Two.60
All fourofthepuzzlesin 96D8-E4 willyieldto thesame treat-
menton thehypothesis6' thatall ofthemcropup becausein this
thereference marks).Crombie,mostsurprisingly, failsto takeintoaccountthe
factthat forPlato the number,Two, a Form, could no more "arise" than
perishand hence could not "arise" fromtheputtingtogetherof two physical
objects.Conversely,the factthat,ifwe did put togethertwo physicalobjects
"here," we would get "two thingshere" would be as obviousto Plato as to
anyone else. Hence Plato could have denounced "because of puttingone
(object)alongsideanother"onlyifhe thoughtthisan answerto (b); forhe would
have seen it as a triviallytrueanswerto (a). How thencould he have failedto
distinguish(a) from(b) ?
5 Cf. the citationin nn. 44 and 32 above, noting the forceof s 181asg
ovaasa in the latter.In the RepublicPlato says thatit is the philosopher's job
to ask "what is theessenceof the One itself"(Tr .7OrefETflV arcro Ev,524E6;
and by the same tokenof Two and othernumbers).When one had done this
one would see thatthe "accountsoftheessence"ofOne and ofTwo allow for
the participationof the same objects,takensingly,in One, takenjointly,in
Two (cf. the precedingnote subfin.). Cf. Hip. Maj. 30iD8-9 wheremuchis
made of the factthat Two can onlybe instantiatedby a pair of individuals,
each ofwhomis one.
60 A substantially similarinterpretation will be foundin Moreau (in the
workcitedin n. 5 above), myonlyobjectionsto it being(a) thathe has to make
Plato a neo-Kantianidealistto bringit off("La causedelaproduction du 2, c'esta
dired'unobjetderepresentationdouble,cen'estdoncpas le rapprochementoula separation
danslespace,maisdansl'esprit.... Toutesles difficultis de cettesortesontdonclevdes
par l'idialismemathematique,quifaitdel'unitiunactintellectual etdunombre
indivisible
unepurerelation," p. 382) and (b) that he does not realize how inappropriate
"cause" (cf.thestartofthecitation)becomeswhenit is clear that (i) thisaitia
is not a physicalone and (ii) Plato is notproposingthata psychologicalcause
be substituted forthephysicalone. In spiteoftheseobjections,I mustrecordmy
heavydebt to Moreau. I have derivedgreaterhelp fromhis discussionofthe
puzzles in thePhaedothanfromany othersinglesource.
61 I say "hypothesis,"forcertainlythereis nothingwhateverin thewording
ofthesesixlineswhichstatesor impliesthattheirpuzzlesarisebecausephysical
factorsare beingconfusedwithlogicalones.For thishypothesis I claimno more
than indirectverification fromthe context.We startat 96A8 withthe tale of
young Socrates' addiction to a methodologywhich restrictsthe quest for
aitiai to physicalcauses. We are thengiven a sequence of six puzzles, all of
themillustrativeof the same perverseline of thought(cf. Cebes' questionin
96E6, whichconcernsthe firstfour,but is answeredby the presentationand
313
GREGORY VLASTOS
benighted phase of his philosophical evolution Socrates62 was

confusingphysical aitiai with logical ones: he was assuming that
a materialfactor,like a head, or the materialpresenceoftwo units,
or the material projection of a part of one thing beyond another
could account forthe respectivestatements,all of which are true
a priori, and could be accounted for only by referringto the
meaning of the termstheyuse. Thus, take the most interestingof
the puzzles: Why are ten thingsmore than eightthings?Reflecting
on this in Part Two, now that he is well out of that particular
fly-bottle,Socrates declares:
So you would be afraidto say [thatis, you would notsay] thatthe io
thingsare morethanthe 8 in virtueof2 things,and thatit is because
of thisthattheyexceed,insteadofsayingthattheyexceedin virtueof
numerousness and because of numerousness [ioiB4-6].
What Socrates is tellingus, put into modern language, is that the

reason why the group of ten is more numerous than the group of
eightis simplythat it satisfiesthe logico-metaphysical63
conditions
discussionofthelasttwo; and notethatthesamesolutionis offered forall sixin

iooE8- ioi C9). The last two puzzles are discussedin extenso(twelvelinesfor
thesetwo as againstsix linesforthe firstfour); and these,as I argued above,
turnout to be cases ofconfusionofphysicalwithlogical aitiai.It is, therefore,
reasonableto assumethatthefirstfourare also casesofthesame confusion and
thatthiswouldhave come out intotheopen iftheyhad been discussed,instead
of merelymentioned,in the text.
62 It shouldbe noticedthatPlato does notsay thatthenatural had
philosophers
made thisconfusion, but onlythatSocratesdid so whenhecame underthespell
oftheirteaching.Plato does not hesitateto attackhis predecessors sharplyfor
theirobnoxiousdoctrines(see,e.g.,Phil.28D-29A; Tim.48B; Laws 888E-89oA
and 967A-D). So it is unlikelythathe would have hesitatedto lambastethem
in thispassage if such absurditiesas those in 96D8 if. had figuredin their
writings.And the factis that no survivingfragmentof theiroriginalworks
indulgesin thiskindofthing.In the lightof theseconsiderations we had best
refrainfromascribingsuch logical solecismsto the physicalphilosophersand
mathematiciansof the time (as is sometimesdone in the literature:e.g.,
Crombie,pp. i6o-i6 I), and understandPlato to meanin ourpassageno more
than he actuallysays and directlyimplies-i.e., thatsince theyhad failedto
clarifythe conceptofaitiaand to sortout its categoriallydifferentimportfor
categoriallydifferent subject matter,theyhad lefttheirreaders defenseless
againstsuch confusions as thoserecountedin our passage.
63 I say "logico-metaphysical"
ratherthanjust "logical," in deferenceto the
pointI made earlier,thatforPlato thelogicalrelationofa termto theconcept
underwhichit fallsis at thesame timethemetaphysicalrelationofa sensible
314
REASONS AND CAUSES
of (greater)numerousness. If thiswereto strikeus as uninforma-

tive,Plato wouldagree(thisis an "ignorant"aitia) but insistthat
itis notuselesson thataccount,forit wouldsaveus frommisdirect-
ing our searchforaitiaito irrelevant factors,suchas-in hisown
formulation ofthispuzzle-the presencein one groupoftwounits
whichare not in the other.This would be the reddestof red
herrings,unlessit werelogicallyrelatedto therelativenumerous-
nessofthetwogroups-forexample,byshowingthatthefirst has
as many unitsas does the second and more unitsbesides,not
necessarilythesetwo units,nor necessarily twounits:anynumber
ofunitsin thefirst groupoverand above thosein it whichmatch,
unitforunit,theonesin thesecondgroupwouldfulfill thelogical
requirements ofgreaternumerousness, and thusenable us to say
preciselywhythereare morein the firstthan thereare in the
second.64If the "safe" and "ignorant"aitiadid thiskindofwork
to an eternalForm.To say thatsomethingis the case "in virtueofnumerous-

ness and because ofnumerousness"(the same thingsaid twiceover again for
emphasis,firstby the instrumental and thenby an accusative
dative, vTAXOEt,
witha preposition,&arTovAdos) is expansibleinto"in virtueofparticipating in
the Form,Numerousness."
64 In thisdiscussion I have deliberately gone beyondwhatwe getin thetext,
in orderto bringout thefurther implicationsofPlato's basic insight.If he had
had at his disposaltechniquesof analysissuch as are available nowadaysto
beginners,he could have offereda generalformulato cover all fourof the
puzzlesin 96D8-E4, layingdownthecontextualdefinition, "whereA, B, C are
(positive)magnitudesor cardinals,A is greaterthan B if,and only if,there
existsa C such that A = B + C," and then showingthat this definitionis
satisfiedin all fourcases: In puzzles i and 2, A = theheightofthefirst(man or
horse); B = the heightof the second; C = the lengthofa head. In puzzle 3
(the one discussedin thetextabove), A = io units;B = 8 units;C = 2 units.
In puzzle 4, A = 2 yards; B = i yard; C = A/2 yards (= i yard). Had Plato
been able to clean up theproblemin thisway,he wouldhave sparedhisreaders
two blemishesin hispresentaccountwhichhelp explainwhyhis soundinsight
may be so easilymissed.
(a) He says "numerousness"(1XTOEL,TArOos)in ioiB6, instead of "greater
numerousness," and "magnitude" (pEye'OEt) in IoIB7, instead of "greater
magnitude,"therebyfailingto bringout that both are special cases of the
"greaterthan" relation,and thatthe absolutenumerousness or bignessof the
thingshe is talkingabout is irrelevantto the reasoning.
(b) He givesspuriousreasonsforrejecting"a head" as the reasonwhyA is
biggerthanB in the firsttwopuzzles,saying(ioiA5-B2) thatthiswould lead
to thefollowing"contradictions":itwouldimply(i) thatthesame cause would
producecontraryeffects(makeA bigger,B smaller)and (ii) thata cause ofa
315
GREGORY VLASTOS
for Plato one can see why he could find it so enlighteningwhile

ascribingto it no causal agency whatever.
It is sad then to see him charged by seriousscholarswithhaving
made the very errorwhich, if I am right,he was the firstto spot.
Thus Hackforthtakes him to task forposing a pseudo-problemin
asking forthe aitia of ten being more than eight:
The questionwhethertheadditionof2 is thecause of io beinggreater

than 8 is meaningless,because thereis no more a cause of i o being
greaterthan 8 than thereis of Thursdaycoming afterWednesday
[p. I3I].
Certainlythere is no causehere, and who should know this better

than does Plato, who gives us, as one showpiece of cockeyed
thinkingabout aitia, the puzzle generated by assuming that there
is a (physical)cause forthe truththat I + I = 2 ? But though
thereare no causes forsuch truths,thereare mostcertainlyreasons
forthem, and it was a mark of genius to see that where one type
of aitia, with its peculiar methodological commitments(those of
physical inquiry), is inapplicable, another type of aitia, with its
entirely different(logico-mathematical) methodology, is appli-
cable, and to make his metaphysical theory the vehicle of this
insight.
certaincharacterwill producean effectwitha contrarycharacter(the head,

a smallthing,will cause A to be big). The reasoningis fallacious.In (i) there
is no contradictionin the same cause producingcontraryeffectson different
things;and thereis none in (ii), ifonlybecause it is not beingclaimedthatthe
head makesA big,but thatit makeshim bigger thanB. and thereis no reason
whya man or a horsehas to be a big man or horsein orderto be biggerthan
anotherman or horse (for other examples of this fallacyin Plato see my
"Degrees of Realityin Plato," in New Essaysin Plato andAristotle, ed. by R.
Bambrough[London, i965], p. I4). I submitthathere,as in the case of the
fallacyI pointedout in n. 57 above, Plato's residualconfusionsand fallacies
(whichwouldbe entirely understandable in a thinker wholackedtherudiments
of the logic ofrelations)do not cancel thevalidityofthefundamentalinsight
expressedin the "F in virtueof participatingin A" formula.To see thetraps
intowhichPlato fallsis to admireall themorethetenacitywithwhich,in spite
of thesemishaps,he pursuedthe truthhe saw.
3i6
REASONS AND CAUSES
III. THE "CLEVER" AITIA
We can now considerSocrates'secondanswerto the "Whyis

x F?" question.Insteadof mentioning just the one Form, P, he
now refersus also to another,r, so relatedto 1 thatwhateveris
"named" (thatis, characterized)65 afterr, willalso be "named"
after 1 (io3E2-Io4B4). The firstexample given of the r- -
relationis the pair, Three-Odd:whateveris a triowill also be

odd-numbered. Fromthisand otherexamplesit is clear thathe
has in view a transitive, nonsymmetrical66 relation.He has no
technicalname forit and is contentto use a metaphor:he speaks
of I "bringingon" 0.67 I shall speak of it as "entailment,"
extendingthis termbeyondits normal use as a propositional
connective and allowingittoconnectconcepts, as we sometimesdo
in informal contexts.68 The formulathenforthisaitiacouldbe put
as follows:"x is F because in F and 1 entails0"; or,
itparticipates
moreelaborately:"x is F because, beingG, it must in r;
participate
and sinceI entails(P, x mustalso in
participate i, andhencex mustbe
65Cf. n. 29 above.
66
Plato's relationhas to coverboth cases such as thoseof the Three-Odd,
Snow-Cold couplings,where the relationis clearlyantisymmetric, and also
othersin which,forall we knowto the contrary, Plato perhapsthoughtofthe
relationas symmetrical,as in thecase oftheFire-Hotcoupling.Whatis certain
(fromexampleslikeThree-Odd,Two-Even,etc.) is thatPlato thinksofpartici-
patingin I as a sufficient,but not also a necessary,conditionofparticipating
in (.
67 Having startedoffat Io3E2 ff.usingstillmorecumbersome language to
expressthe I-0 relation(which I have abbreviatedabove to "whateveris
'named' by r will also be 'named' by 0"), he shiftscasually to lE'7rnpEpEtVat
I04E io and usesitfrequently (I surmise:simplybecauseitis shorter),
thereafter
varyingit withthe expressiondiscussedin theprecedingnote,wherethe P-k
relationis indicatedvia theisomorphicG-F relation:ifG comesto be present
in x, thenx will be F. As Shoreypointsout (p. i i and n. 3 ofthesecondpaper
mentionedin n. 3 above; formoreexamplessee Bonitz,IndexAristotelicus) the
terms 7rnpe'pEv and avve7mlpleV as well as some of the othertermsused here
by Plato to expressrelationsbetweenForms are also used by Aristotleto
express relations
entailment between generalterms.
68 Ifwewere toconvert Plato'stalkaboutFormsintoset-theoretical
language
takingA, r, etc. to namesets,the"bringing-on" woulddenotethe
relation
inclusion ofthe"bringer-on" in the"brought-on," notthemembership ofthe
former inthelatter:Platoclearlyhasnointerest thattheForm,Fever,
insaying
is sick,or thattheForm,Fire,is hot.
317
GREGORr VLASTOS
F." Plato does not spell out any such formula as this. But an
examination of his text will show, I believe, that this is what the
sketchierphrasingsthereimply.For what he understandsby them
we must rely on his examples. There is a flockof them. First,he
gives additional arithmeticalcases of 1- P linkages: Five-Odd,
Two-Even, Four-Even, Ten-Even.69 Then, withoutany apparent
shiftingof gears, still talking about precisely the same relation,
explainingpreciselythe same aitia,he bringsin physical,biological,
and other examples of Form, F, entailing Form, P: Fire-Heat,
Snow-Cold, Fever-Sickness,Soul-Life.70Does this answer to the
"Why is x F?" question give us more reason thandid thepreceding
to thinkthat his Forms are meant to be causes ?
Let me press one of his examples, where Fever is the aitia of a
sickness.We may assume the followingbackground: A man dis-
plays the cluster of symptomswhich would have justified us in
classifyinghim as sick beforediagnosing his particular ailment:7'
he suffersfromweakness,loss of appetite, pain, and otherpsycho-
logical registersof physical distress.We examine him and see he
is veryhot.72 We inferthat he is sick becauseof this.7 Socrates steps
in at thispoint to tell us we are entitledto make thisinferenceonly
because the man participatesin the two Forms, Fever, Sickness,
69 104A4-B2; Io5A6-7; 105G9 ff.
70 He had alreadyintroducedtheFormsHot, Cold, Snow,Fire,to illustrate
therelationof incompatibility betweenForms(Io3CIo ff.)-a relationwhich
I leave out of my discussionin thispaper to avoid burdeningstillfurtherits
alreadyoverburdenedexposition.
71 These symptoms mustbe sufficient to warrantthe classification"sick,"
but not sufficientto warrantthe narrowerone "feverish,"else we would lose
the"cleverness"ofthepresentaitia: we would be back in theformally tautolo-
gous "safe" and "ignorant"aitia.And cf. the nextnote.
72 This is how we mustunderstand "fever"here (takingpyretosin its literal
sense,"burningheat,fieryheat" [Liddell and Scott,Greek-English Lexicon, s.v.,
I]-i.e., as that "excess of heat" in the body whichPlato takesto be a cause
of a varietyof ailmentsin all of which the patientis feverish:Tim.86A2-3,
'rO R-EV oVV (K 7vpog v7EpfoArp... voacrav . . ), if Fever-Sickness is to parallel the
Fire-Heat,Snow-Cold couplings,as it is surelymeantto do: if"fever"were
understoodhere to meana species of sicknessin the firstplace, the coupling
would not constitutean example of the "clever," i.e. informative, aitia.
73 A verysubstantialinference: ifvalid, it empowersto inferthathe is sick
just fromknowingthathe suffers from"excessofheat," as ofcoursewe could
notifwe werenotjustifiedin movingfrom"he is sickandsuffers fromexcessof
he suffers
heat" to "he is sickbecause fromexcessofheat" in thefirstplace.
318
REASOJNSAND CAUSES
thefirstofwhichentailsthesecond.Whenthisis abbreviatedto
"the Form,Fever,is what makeshim sick,"it has an alarming
ring.It soundsas thoughthe Form were a ghostlystand-infor
bacteria.But we need onlyrecalltheforegoing argumentwhich,
I trust,clearedtheFormofimputation ofcausalagencyin thecase
ofthe "safe"aitia,to assureourselvesthatthesame clearancecan
be givenitin thecase ofthepresentaitiaas well.Ifl isnotexpected
to be a cause whenitis said to "make"x to be F, thenbythesame
tokenP cannotbe expectedtobe a causewhenit is said to "make"
x to be F.74
May we then conclude that here, too, the "makes" in the
Platonicformulahas a strictly logico-metaphysical force-thatno
greatercausal significance is to be read into "the Form,Fever,
makesthemansick"thaninto"theForm,Sickness, is whatmakes
him sick"? Such had been Shorey'sclaim when he maintained
againstZellerthatin thewholeof thispassagePlatois concerned
withlogic,notphysics, adumbratinga theoryof syllogistic infer-
ence,notofcausal explanation. 75This is an attractiveinterpreta-
tion:one could wishit weretrue.And nothingwouldhave stood
in theway ofour takingit as trueifPlato had givenonlylogical
and mathematical examplesofthe F- 0 relation.If we had only
Three,Odd, Two, Even,and thelikeamongthe examples,then
certainlythe F- 0 couplingcould be strictly noncausal. That
Jones'sfamilymustbe odd-numbered because it happensto be a
threesome is indeedausterely irrelevantto thecausal orderofthe
world.Not so whenwe are toldthatJonesis sickbecausehe has
a fever,thata burninglog is hot because it is on fire,thatthe
whitestuffon the groundis cold becauseit is snow.To be sure,
none of the entailments holdingbetweenthe relevantFormsare
beingcreditedwithcausalagency.Buttheyare certainly expected
That theoccurrenceoffeveris thecause
tohavecausalimplications.
of the occurrenceof sicknesswould be a textbookexampleof a
74 Cf. mycritiquein n. 46 above ofthelastofthecitationsfromCornford in

that note.
75 On pp. 7-8 ofthe secondpaper mentionedin n. 3 above. And thisis the
less extremeof Shorey's claims: cf. the strongerone (which I cited in my
openingparagraphand towhichI alludedin n. 49 above) thatPlato is offering
"only a tautologicallogic." And cf.n. 78 below.
319
GREGORr VLASTOS
cause in Greek medicine.76And since the F- P entailmentis

beingoffered as thejustification
ofthecausalinference, howcould
it be emptyofcausal significance? The samewouldbe trueofthe
Fire-Heatand Snow-Coldcouplings."Fireand snow,likebronze
in thesecondexamplein SectionI above,are naturalkinds,and
theinvarianceoftheconcomitance ofthecharacteristicproperties
in each of themsignifies a multitudeof causal interconnections
withotherkindsof matterin the universe.Thus when Socrates
maintainsthattheForm,Snow,is theaitiaofcold,he is asserting
neitherthe metaphysicalabsurditythat the Form,Snow, chills
selectedregionsof the universe,nor the semanticabsurditythat
snowcausesitselfto be chilly;butwhathe does assertis neverthe-
lesstiedfirmly to thecausal structure
oftheworld-forexample,
to thefactthatifwe raisethetemperature beyonda certainpoint
snow mustchange to water.This "must" is a causal one. And
since in Plato's theoryit is groundedin relationsof entailment
betweenForms,it wouldhave to be a fantastically strong"must":
it would have to expressa physical law thathas logicalnecessity.
Since Plato claimsthatthesnowofour experienceis cold because
the Form,Snow, entailsthe Form,Cold, and sinceall Forms-
thoseof physicalstuffs and processes,no less than thoseof logic
and mathematics-areeternaland sustainonlyimmutablerela-
tionsto each other,he is implyingthatthelaws ofnature,could
we butknowthem,wouldhave thesamenecessity as do thetruths
ofarithmetic and logic.78
76
Cf. n. 72 above.
77 I ignorethe Life-Soulcoupling,whichraisesotherproblemsthatcannot
be discussedin thispaper.
78 We could have reachedthe same conclusionin Sec. I above, ifPlato had
used exampleslike Snow and Fire in illustrating the "safe" aitia, insteadof
stickingto logico-mathematical ones (Numerousness,Greatness,One, Two)
and to thattantalizingabstraction, Beauty.To simplify theexpositionI played
hisgame,using"square" as myownexample.Had I shifted to,say,"x is bronze
because it participatesin the Form,Bronze,"it would have becomeapparent
that even the "safe" aitia, thoughexpressedin a tautologicalformula,has
far-reaching substantiveimplicationsforthe causal orderof the universe:for
as I remarkedabove, the regularconcomitanceof the propertieswhichmake
up that naturalkindhas causal implications;to say that the relevantcausal
laws are instantiatedin x because x participatesin a PlatonicForm,Bronze,
is to creditthoselaws withabsoluteimmutability and to implythattheymay
be knowna priori.
320
REASONS AND CAUSES
A theorysuch as thisis not likelyto get a sympathetichearing

fromphilosophersnowadays. Most of us have been brought up
to think that the laws of nature are in the last analysis radical
contingencies-deficto uniformitieswhich we must eitherexhibit
as special cases of still more general defacto uniformitiesor else
accept as thingsforwhich no furtherreason can be given. Coming
upon Plato's reductionofphysicalto logical necessityin the Phaedo,
we may then be temptedto thinkof it as not only false but unrea-
sonable, wrong-headed,indeed light-headed, a kind of whimsy.
We would do well then to reflectthat in the modern period, too,
a substantiallysimilarview has been propounded by philosophers
-by Leibniz, for example, who held that all syntheticand con-
tingent truthsmust representnecessary, analytic truths,imper-
fectly comprehended by finite minds; that the neo-Hegelians,
from F. H. Bradley to Brand Blanshard, find in Hume's alter-
native to Leibniz a dissolution of causality into casualty and
insistthat "being causally connected involve(s) being connected
by a relationoflogical necessity";79 and finallythat, to speak from
the other,far more populous side of the fence,while it is generally
admitted that causal laws must support counterfactuals, the
problems of explaining counterfactualson a regularitytheoryof
the laws of nature is a troublesomeone and its solution is still
under debate.
There had been no Hume in Plato's past. The physicalphilos-
ophers had proceeded on the faith that, as Leucippus had ex-
pressed it, "nothinghappens at random, but everythingby reason
and by necessity."80But when theylooked criticallyat thisaxiom,
79The Natureof Thought, II (London, 1939), 515; theitalicsare Blanshard's.

Cf. A. C. Ewing, Idealism(London, 1934), p. I71: there is an "intrinsic"or
"inherent"bond betweencause and effect ofwhich,he says,he can onlythink
of"cause and effect as connectedbya relationoflogicalentailment";and Proc.
ofAristot. Soc.,supp. vol. 14 (I935), 66: "The cause logicallyentailsthe effect
in such a way thatit would be in principlepossible,withsufficient insight,to
see whatkindofeffect mustfollowfromexaminationofthecause alone without
havinglearntby previousexperiencewhat were the effects ofsimilarcauses"
(quoted fromEwing by E. Nagel, The Structure of Science[New York, 196i],
P. 53).
'I H. Diels and W. Kranz, Die Fragmente (Berlin,19394-193 7)
derVorsokratiker'1,
Frag. B I.
321
GREGORY VLASTOS
as Democritus, the last in the succession,was the firstto do, all

theycould findin nature on which to base theirfaithin rational
necessitywas defictoregularity.He taught,we are informed,that
natural explanation reduces to the principle that "things always
are or happen thus" and that "there is no sense in looking for a
reason for that which always happens."81 We cannot tell from
survivingfragmentsor reportswhat conclusionshe drew fromthis
remarkable reflection.But regardless of what Democritus may
have made of it, we can see with what force Plato could have
retorted: "If you must have rational necessityin nature, you
cannot get it fromregularitieswhich are mattersof brutefact.The
only kind of rational necessityknown to me is that which I findin
mathematics82and dialectic. Do you know of any others? If so,
explain yourself.Until you do, I will continue to believe that
nature could exhibit rational necessityonly if its laws mirrored
the interrelationsof the Forms we explore in logico-mathematical
reasoning."
To say this is not, of course, to suggest that Plato's view is
unobjectionable. Its mostglaring faultis its methodologicalsteril-
ity for natural science. What knowledge of the laws of nature
could one hope to secure a priori by followingout lines of entail-
ment from termslike fire,snow, and fever? The entailmentsin
our passage are depressingcommonplaces. But even so, it is not
clear that they would warrant the certitude with which they
would be credited on this view. How could we know that the
Form, Fire, really entails the Form, Heat? It would be no use
tellingus that we would know thisifwe had "recollected" the two
concepts correctly.For how could we be sure of that? What guar-
antee would we have that what we learned about firefromour
sense-experience,sadly limited by the parochial contingenciesof
our time and place, would not have led us astray? If there were
81Phys.252A32-BI (= frag.A65 in Diels and Kranz).
82 Plato's criticismof contemporary mathematiciansin Rep. 51 OC-5 iD
does notimply(or evensuggest)lack ofconfidencein theabsolutecertitudeof
mathematics, whosesubjectmatteris eternal(527B7-8)and,therefore,"draws"
or "leads" thesoul away fromthefluxto eternalbeing(524E-525B). (It should
be notedthatPlato does notimputeto themathematicians theabsurdassump-
tion that theirsubject matterconsistsof visible figures;he says explicitly
[5ioD6-71 thattheyare notreasoningabout these.)
322
REASONS AND CAUSES
stuffswhichburnwitha cool flameelsewhere in theuniverse,and

we had known of them,our notion of fire would have been
different, and thenwe would not have thoughtof claimingthat
theeternalFireis eternally linkedto Heat. It is impossibleto tell
fromourpassageto whatextentPlato was assailedat thistimeby
such doubts. Here, as elsewhere,he has a way of keepingthe
spotlightof his discourseon just thoseareas wherehe is most
confident oftheanswers,contentto leave muchelse in obscurity.
This artfulchiaroscuro makeslifedifficult foranyonewho triesto
expound his thoughtsystematically. Time and again we come
acrossgaps in his thought,not knowinghow he would expectus
to fillthem.This way ofwritingphilosophyis not to be excused,
and I have no desireto excuseit. But thismuchcan at least be
said forPlato: his silencesare themselves suggestivenot ofconfu-
sionbutofa canny,self-critical awarenessofthe limitations of his
theory.The problemshe persistently declinesto discussin the
middledialoguesare thosewhosesolutioneludeshim.
Thisis conspicuously truein thepresentcase. If Platohad really
thoughtwe could syllogizeour wayintothesecretsofthenatural
universe, hisconfidence in sucha fantasy wouldhavebeenpathetic.
But thefactis thathe offers us no suchpseudo-science ofnaturein
thisdialogue.83His ideason geography and astronomy he presents
onlyin theframework ofa myth. 84 If notalreadyin thisdialogue,
thensoonafterhe musthave drawntheonlyconclusionopen to a
sensibleand honestmanwhohad to livewithPlato'smetaphysical
theory:that therecan be, strictlyspeaking,no such thingas
knowledge ofnature-onlyeducatedguesses, verisimilitudes,
plausi-
bilities.Such a conclusionis clearlyimpliedin theRelpublic. When
he drops empiricalscienceslike physics,biology,and medicine
fromthecurriculum ofhigherstudies,thereis no suggestion that
theirsubjectmatterwillbe reclaimedat a higherlevelbydialectic.
FormslikeFire,Snow,and FeverneverdarkenthepagesofBook
83 He appears to be
disclaimingit by emphasizingSocrates' ignoranceof
naturalcauses (99C6-9; IooD3; ioiC9-DI).
84 Includingsome importantscientific doctrines,such as the sphericityof
the earth,itsstability"at the centerofthe heavens" (Io8E4-IoqA6), and the
impliedrepudiationof the ancientnotionof an absolute "up" and "down"
(I 12C1-2).
323
GREGORY VLASTOS
VII of the Republic.His point of view remains the same in this

respectwhen he comes to deal directlyand at lengthwith empiri-
cal topics, as he does in the Timaeus.There the Forms of Fire,
Water, and the like are accorded a curious and revealing treat-
ment. Their existence is formallyproved,85and they are placed
ceremoniouslyon their metaphysical pedestals, only to be left
there and quietly ignored in the rest of the treatise where the
workingsof nature are explored. The plethora of ingeniousexpla-
nations of natural phenomena displayed in the sequel is spun out
of the theoryof the geometrical configurationof matter. We are
not told that firecauses water to evaporate, meltsmetals, cuts up
foodstuffsinto digestible and assimilable particles because the
Forms of Fire, Water, and so forthentail the Forms of the corre-
sponding processes.The aitiai of physical,chemical, and biological
phenomena are not deduced from"accounts of the essence" of the
Forms, but are derived syntheticallyfrom the structureof the
atom. And what is claimed forthemis not certainty,but verisimil-
itude,86the atomic theoryitselfbeing presentedas no more than
a plausible hypothesis,87 having no more than aestheticelegance88
and the saving of the phenomena to recommend it.
I implied at the startofthispaper thatour passage in the Phaedo,
rightly understood, is not unworthy of Plato's philosophical
stature.The reader can now see why I made this claim and may
assess its meritsfor himself.If my interpretationis correct,Plato
has not only distinguishedmechanical fromteleological causes-
thispart of his contributionI have not attemptedto discuss-but
has also come within sight of the still more radical distinction
between both of these and the logical aitia of classificationand
entailment.Had he availed himself,as Aristotlewas to do, of the
expositorydevice of philosophical lexicography,this achievement
would have been more perspicuous and also, no doubt, more
complete,forin making his thoughtmore explicit he would have
attained greaterlucidityhimself.We should not, however,be put
85 5iB7-52A4.
86 Cf. my "DisorderlyMotion in the Timaeus,"in R. E. Allen,op. cit.,pp.
382-383and notes.
87 53D4-6.
88 53D7-E8.
324
REASONS AND CAUSES
offby thefactthatat no pointdoes he sayin thestyleofhisgreat

pupiland critic,"aitia has manydifferent senses."Thereare other
waysofexhibiting distinctions,and one way ofdoingso is to use
them.This,I have argued,is whatPlato doesin ourpassage,most
ofall in thepart whichhas been leastunderstood
successfully in
the scholarlyliterature,wherehe uses the "safe" aitia to explode
pseudo-problemswhich arise when the categorial difference
betweenlogical and physicalaitiai is ignored.89If it were then
suggestedthatPlato cannotbe afterall so clearon thispoint,else
hewouldnothaveusedindiscriminately arithmetical,physical,and
medicalconceptswhen illustratingthe "clever"aitia, I trustthe
answeris now apparentfromwhat I have said in theconcluding
partofthepaper.Thereis no confusionhere,but the expression
ofhisfirmconviction thatall intelligible physicalno less
necessity,
thanmathematical, mustbe groundedon logicalnecessity, sinceit
representstheinterrelations ofeternalForms,be these articulated
in discourseorimagedin thephysicalworld.Thisconviction could
easily have set him startedin pursuitof a will-o'-the-wisp, a
physicalsciencewhichdeducesthe laws of naturea priori.It is
a mark of good sense, no less than of clear thinking,that
in his subsequentwritingshe claimed the certitudeof logical
necessityonlyforpropositions of mathematics and dialectic,and
was contentwitha physicaltheorywhich,he conceded,was no
morethana beautifulguess.
GREGORY VLASTOS
PrincetonUniversity
thelast twoofthesixpuzzles; and,

89 I.e., in 96C6-97B3,wherehe discusses
ifmyhypothesis concerningthepointofthefirstfouris accepted,in 96D8-E4

as well.
325

Philosophical Review Volume 78 Issue 3 1969 (Doi 10.2307/2183829) Gregory Vlastos - Reasons and Causes in The Phaedo

Загружено:

Сведения о документе

Оригинальное название

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Philosophical Review Volume 78 Issue 3 1969 (Doi 10.2307/2183829) Gregory Vlastos - Reasons and Causes in The Phaedo

Загружено:

Авторское право:

Доступные форматы

Philosophical Review

Reasons and Causes in the Phaedo

THERE is a passagein thisdialoguewhichhas led many

1 An earlierversionofthispaper was deliveredat MichiganState University

More recently,commentatorshave not hesitatedto take theirown

I. ON THE MEANING OF AITIA

Since so much will turn on the meaning of the word aitia,6I

dissertation:see his citationfromthe latterin "The Originofthe Syllogism,"

icance is farwiderthanthatoftheword"cause" as used nowa-

I have deliberatelyavoided the word aitia in formulatingthese

thingcomes;V. caseindispute. In thispaper I shallbe concernedexclusivelywith

aitiai as "all the ways of stating &

8 K 7&vrw) cbroaorfov T ri followed by a listing of the four aitiai, Phys.

notin thisinquiry,whereall thatmatters is thatifwe weretalking

also M. White,Foundations of HistoricalKnowledge(New York, i965), ch. V,

false,and may be collapsed by the simple reflectionthat it is not,

13 Reason(New York, 1954), pp. 287 ff.

II. THE "SAFE" BUT "IGNORANT"AITIA

Our passage fallsinto two divisions.DivisionOne (95E-99C)

whole passage cannotbe identified in totowiththe historicalfigure,thoughit

(i) Forms-that is to say, those of the full-blowntheoryof

thoseofothersand is preparedto do withoutforthepresentis the teleological

"named afterthem" (avth65vrovhrcov r,)v Sewvvwu'avaxELv) cf. n. 29 below.

20 78D (citedin n. 44below); 79D; 8oB.

oppositething" (To EVavTtov vrpaiyyta)-thatis, the individual that

"immanent"and "transcendent."This would be unobjectionableif it were

Though theexpectationwas neveradequatelyfulfilled,

Armedwiththis"hypothesis," Socratesfeelsempoweredto give

"we mustlookforsomeotherway [i.e., a way otherthansimilitude]by which

he would not have said thathe is still"deprived"ofteleological

37The sole patternofteleologicalexplanationenvisagedin thePhaedois that

scholarship in spiteofShorey'svigorousprotests.Thus Hackforth

thisjob in additionto thatof the "formal"aitia. On the chargeagainstPlato

blunderedso grossly.And thereis no suchevidence.All we have

42 The "because" can also be expressedby the "instrumental"dative: r47

thanbecauseit participates in Beauty"(IooC5-6), he is certainly

44 "That realityitself(avtrn n ov'ata), of whose essencewe give the account

countlessotherFormsalso existwhichdo not help to answermy

if,and onlyif,x participatesin (D. From thisit followswiththe

are peculiarlymystifying to the modernreader.5'He is likelyto

poker-faced,mistakesincurredin his distantyouthwhenhe had

Here at lastwe see thegaffeSocrateshad beenperpetratingin that

thesethingsnow?" (96E5). ThereafterSocrates' verbschange fromthe past

97A2; etc.,windingup at B7 with"I will have no truckwiththis[methodof

aitiaofthe logico-mathematical truththatthe same itemswhich

57This is reinforced by a furtherobjection.If we wereto taketwo-nessas the

in virtueofparticipatingin theDyad, thenit willbe clearthatthe

benighted phase of his philosophical evolution Socrates62 was

What Socrates is tellingus, put into modern language, is that the

discussionofthelasttwo; and notethatthesamesolutionis offered forall sixin

of (greater)numerousness. If thiswereto strikeus as uninforma-

to an eternalForm.To say thatsomethingis the case "in virtueofnumerous-

for Plato one can see why he could find it so enlighteningwhile

The questionwhethertheadditionof2 is thecause of io beinggreater

Certainlythere is no causehere, and who should know this better

certaincharacterwill producean effectwitha contrarycharacter(the head,

III. THE "CLEVER" AITIA

We can now considerSocrates'secondanswerto the "Whyis

relationis the pair, Three-Odd:whateveris a triowill also be

74 Cf. mycritiquein n. 46 above ofthelastofthecitationsfromCornford in

cause in Greek medicine.76And since the F- P entailmentis

A theorysuch as thisis not likelyto get a sympathetichearing

79The Natureof Thought, II (London, 1939), 515; theitalicsare Blanshard's.