Peter B. M. Vranas - Aristotle on the Best Good is Nicomachean Ethics 1094a18-22 Fallacious

7/28/2019 Peter B. M. Vranas - Aristotle on the Best Good is Nicomachean Ethics 1094a18-22 Fallacious

http://slidepdf.com/reader/full/peter-b-m-vranas-aristotle-on-the-best-good-is-nicomachean-ethics-1094a18-22 1/14

Aristotle on the Best Good: Is "Nicomachean Ethics" 1094a18-22 Fallacious?

Author(s): Peter B. M. VranasSource: Phronesis, Vol. 50, No. 2 (2005), pp. 116-128Published by: BRILLStable URL: http://www.jstor.org/stable/4182772 .

Accessed: 13/08/2011 15:01

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 [email protected].

BRILL is collaborating with JSTOR to digitize, preserve and extend access to Phronesis.

http://www.jstor.org

http://www.jstor.org/action/showPublisher?publisherCode=bap

http://www.jstor.org/stable/4182772?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp

http://www.jstor.org/page/info/about/policies/terms.jsp

http://www.jstor.org/stable/4182772?origin=JSTOR-pdf

http://www.jstor.org/action/showPublisher?publisherCode=bap



Aristotle on the Best Good:

Is NicomacheanEthics 1094a18-22 Fallacious?*

PETER B. M. VRANAS

ABSTRACT

The firstsentenceof NE 1.2 has roughly the form:"IfA [there is a universal end]

andB (because,if not-B, thenC), thenD [this endwill be the best good]".According

to some commentators,Aristotle uses B to infer A; but then the sentence is fal-

lacious. According to other commentators,Aristotle does not use B (until later

on); but then the sentence is bizarre.Contrary o both sets of commentators but

following Wedin 1981), I suggest that Aristotle uses B together with A to infer

validly that there is a non-instrumental and thus uniqueuniversal end (hence

D). On this interpretation he above two problems disappear,but a subtler prob-

lem emerges: not-B does not entail C.

1. Introduction: A multiplicity of interpretations

Aristotlecommentators ave rackedtheir brainstryingto make sense of

the sentence that opens Chapter2 in Book I of the NicomacheanEthics

(1094al8-22) - hereinafter"the Sentence".In Irwin's(1999) translation

(whichI use throughout),he Sentencereads:

Suppose, then, that [A] the things achievable by action have some end that we

wish for because of itself, and because of which we wish for the other things,

and that [B] we do not choose everythingbecause of somethingelse - for if we

do, [C] it will go on without limit, so that [C'] desire will prove to be empty and

futile[; c]learly, [D] this end will be the good, that is to say, the best good.

Ackrill (1974/1999: 68; cf. 1973:241) comments:

It is commonly supposedthat Aristotle is guilty of ... the fallacy of arguingthat

since every purposive activity aims at some end desired for itself there must be

some end desired for itself at which every purposive activity aims.... The out-

line structureof the sentence is "if [A] and [B], then [D]." Nobody will suggest

Accepted August 2004* I am very grateful to Travis Butler, Stephen Darwall, Stephen Everson, Heimir

Geirsson, Alan Hajek, Aviv Hoffmann, Joel Richeimer, Catherine Wilson, and an

anonymous reviewer for helpful comments. A version of this paper was presentedat

the 2003 annual meeting of the Central Division of the American Philosophical

Association.

C KoninklijkeBrill NV, Leiden, 2005 Ph-onesis L12

Also available online - www.brill.nl



ARISTOTLEON THE BEST GOOD 117

that [B] is here a condition additional to [Al. The one naturalway to read thesentence as a coherentwhole is to suppose that [not-B] is mentioned as the only

alternative to [A]. In that case a proof of [B] would be a proof of [A]. So when

Aristotle gives his admirable proof of [B] he is purporting o prove [A]; and the

sentence as a whole therefore amounts to the assertion that [D].

Commentatorswho accuse Aristotleof the above fallacy include (besides

Ackrill) Anscombe (1957: 34, 1967: 15-6), Darwall (1998: 192), and

Geach (1958/1972: 2); also, more tentatively, Bostock (2000: 9-10),

Broadie (1991: 12-4), andUrmson (1988: 10).' (Some of these commen-

tators also arguethataddingto B an extrapremise- which is implicitlyor explicitly endorsedby Aristotle resultsin a valid argument or A.)

Other commentators includingCooper (1975: 93), Hardie(1965: 277,

1968: 16-7), Irwin(1999: 173; contrast 1977: 52), Kenny (1966: 94-5),

Kraut 1989: 203-7, 217-20), Reeve (1992: 108-11), and Williams (1962:

292) - argue that Aristotlemay be innocentof the above fallacy because

he need not be understoodas purporting o prove A in the Sentence:he

may be understood nstead as stating A hypothetically.But then what

would be the role of B? As Ackrill notes: "If [B] were simply a correct

remark - irrelevant to, or a mere consequence of, [A] - it would beabsurdlyplaced and serve no purpose" 1974/1999: 68). Some commen-

tatorsargue thatB paves the way for what follows (e.g., for the function

argument n 1.7). Still, Ackrill's point stands that on such an interpreta-

tion the Sentence itself is bizarre.

In this paperI examine an interpretation n which Aristotle does not

committhe above fallacy and the Sentence s not bizarre ither.This inter-

pretationwas proposed by Wedin (1981), but to my knowledge it has

escaped notice so far.2 The interpretation uggests what Ackrill claims

nobody will suggest, namely that B is "a condition additionalto" A. I

I Anscombe 1957: 34 might be thought to refer to 1094al-3 (ratherthan to the

Sentence), but this is unlikely given Anscombe 1967: 15-6 (see also Kraut1989: 217-8 n. 14). Engberg-Pedersen 1983: 29-31), Hughes (2001: 28-31), and Joachim(1951:21) belong to a groupof commentatorswho (implicitly or explicitly) take Aristotleto

infer A from B in the Sentence but (for various reasons) do not accuse Aristotleof the

above fallacy. (See also Gauthier & Jolif 1970: 7; Sparshott 1994: 15-6.) Robinson

(1964: 17) accuses Aristotle of a slightly different fallacy (cf. Kirwan 1967: 110-1).2 A search throughthe Arts & HumanitiesSearch database failed to yield any ref-

erences to Wedin's paper. I found the paper mentioned only by Bostock (2000:9 n. 7). The presentpaper complementsWedin's in three respects. (1) Wedin did not

prove thatB guarantees he non-instrumentality f universalends; he provedonly that

B (given A) guarantees uniqueness(see ?2). (2) Wedin did not distinguish the three

versions of the literal interpretation distinguish in ?3; he defended only (what I call)



118 PETERB. M. VRANAS

prove a theoremto the effect thatB, in conjunctionwith A, entails thatthere is a non-instrumentali.e., not pursuedbecauseof any otherend)

and (thus) unique universal end. Such an end may well deserve to be

called "the best good", whereasan instrumental r a non-uniqueuniver-

sal end need not deserve this label. So I understandhe Sentenceneither

as "B;thus A; thus D" (as some of those who accuseAristotleof the fal-

lacy do) nor as "B; moreover, f A, thenD" (as some of those who try to

acquitAristotledo); I understandt literally instead,as (to a first approx-

imation)"if A and B, then D".

In ?2 I elaborateon the literalinterpretation.n ?3 1 distinguish hreeversions of the interpretation.n ?4 1 addressan objection.I conclude

in ?5.

2. A literal interpretation of the Sentence

Let Pxy standfor "x is pursued wishedfor, chosen)becauseof y",x and

y being ends (of "the thingsachievableby action").3The claim that (A)

"thethingsachievableby actionhave some end that we wish for because

of itself, and becauseof which we wish for the otherthings"can be for-malizedas:

(A) 3x(Pxx & Vy(y?x Pyx)).4

the hypotheticalversion. (3) Wedin did not address the objection to the literal inter-

pretationI address in ?4.1 Like Williams (1962: 289) and Kirwan(1967: 97), 1 take Aristotle to use inter-

changeably in (the vicinity of) the Sentence expressions like "we wish for" and "we

choose", and I use "we pursue" o cover such expressions. Like Williams (but unlike

Kirwan,to whom I refer the readerfor details), I do not index choices to people or

times; doing so would (for my purposes unnecessarily) clutter the notation. Unlike

Williams and Kirwan, I restrict the ranges of variables so that I need no one-place

predicatefor "is pursued" as opposed to "is pursuedbecause of").

Rackham(1926/1934: 5) translatesthe part of the Sentence that expresses A as

... we wish the others only for the sake of this" (emphasisadded); this would result

in a differentformalization,but I see in the Greek text no basis for the word "only".

Aquinas(c1271/1993: 7; cf. Broadie 1991: 12-3) understandsA as ". . . we wish other

things because of it" (emphasisadded);this would result in yet anotherformalization,

but it is clear that the Greek text says "theothers"(emphasisadded), and this is stan-

dardly understoodas "all the others"(see also Cooper 1975: 91 n. 2). This under-

standing of "the others"also rendersimplausibleanother readingof A (which I have

not seen defended in print but variants of which have been proposed in correspon-

dence by a couple of people), namely Vy3x(Pxx & Pyx). (Cf. Kirwan 1967: 97-8. This

is in fact a common reading of B ratherthan A: see footnote 5.)




This is equivalent to ]xVyPyx; i.e., there is an end because of which every

end is pursued. Call such an end universal. Now the claim that (B) "we

do not choose everything because of something else" can be formalized as:

(B) -_Vx3y(y?x & Pxy).5

This is equivalent to ]xVy(y?x -_ -iPxy); i.e., there is an end that is not

pursued because of any other end. Call such an end non-instrumental.

Note that A need not entail B: a universal end need not be non-instru-

mental. Indeed, if two or more universal ends exist, then each of them is

pursued because of each of the others, so none of them is non-instru-

mental. Conversely, B need not entail A: a non-instrumentalend need not

be universal. Indeed, if two or more non-instrumental ends exist, then none

of them is pursued because of any of the others, so none of them is uni-

versal. There is, however, a connection between A and B, and it is illu-

minated by the following theorem.

Theorem 1. If (A) there is a universal end and (B) there is a non-instrumental

end, then there is a unique non-instrumental nd, which is also the unique uni-

versal end.

Proof. Suppose that A and B are true, and let a be a universal end. Then every

end y different from a is pursued because of a, hence because of an end differ-

ent from y; so y is not non-instrumental, nd the only candidateleft for being a

non-instrumental nd is a. Given that there is a non-instrumental nd, a is the

unique non-instrumental nd. Now if some end b differentfrom a were also uni-

versal, then every end, hence also a itself, would be pursued because of b, so a

would not be non-instrumental. t follows that a is also the unique universal end.

Theorem 1 suggests that, rather than being "absurdly placed" or "irrele-

vant" to A, B is essential to Aristotle's reasoning in the Sentence.6 First,

B ensures that at most one universal end exists. (This point-

but not the

I Taken literally, B is the negation of a universal claim and is thus equivalent to

an existential claim. Commentators,however, typically write as if B were a universalclaim;e.g., the claim that"everypurposive activity aims at some end desiredfor itself"

(Ackrill 1974/1999: 68). This is arguably because they think that "the reason which

Aristotlegives for [B] . . . actually ustifiessomethingstronger han [an existentialclaim]"(Kirwan 1967: 107). I addressAristotle's argumentfor B in ?4 (cf. footnote 13). (Cf.also Wedin 1981: 247-8.)

6

Williams (1962: 294) proves a similar theorem, but does not formulate the literalinterpretation let alone relate his theorem to it); he says rather that B may be pro-

posed by Aristotleas a consequence of A, and that "[theSentence] is in any case con-

fusedly expressed, and it is perhapsimpossible to say exactly what it means" (1962:

292).



120 PETER B. M. VRANAS

secondone below - was also madeby Wedin(1981: 248-9).) This resultseems essentialbecause,if two or moreuniversalends existed, it wouldnot make much sense for Aristotleto say that "thisend will be... thebestgood"(1094a21-2;emphasisadded).7 econd,B ensures hatanyuni-versal end is non-instrumental, ot pursuedbecause of any other end.(Actually this entails that at most one universalend exists!) This resultseems essentialbecauseit mightbe inappropriateo call "thebest good"a universalend pursuedbecause of some otherend. Apparently ecog-nizing this possible inappropriateness, ome commentatorsargue that

Aristotleshouldhave included or shouldbe understood s havingimplic-itly included) he requirement f non-instrumentalityn A (Broadie1991:12; Cooper1975:92). On the literalinterpretationherewas no need forAristotle o do so: he can derivethe non-instrumentalityf universal ndsby usingB.

3. Three versions of the literal interpretation

As I pointedout in ?1, accordingto some commentatorsAristotlepur-

portsto proveA in the Sentence,whereasaccording o othercommenta-tors Aristotle may be understoodinstead as stating A hypothetically.Although,followingWedin(1981), I formulated he literalinterpretationas (to a first approximation) if A and B, thenD", TheoremI supportsequallywell the alternativenterpretationAandB; thusD". Call the twocorresponding ersions of the literalinterpretationypotheticaland non-hypothetical espectively.9 do not wish to take a standon whichversionof the literal interpretations preferable. n supportof the hypotheticalversionone mightnote - as Hardie(1968: 17), Irwin(1999: 180), Kraut

(1989: 205, 227-8), and Williams (1962: 292) do note - that in the

See Kirwan 1967: 108-9 for an objection,and Wedin 1981: 261 for a reply.8 As I said, if two or more universal ends exist, then each of them is pursued

because of each of the others, so none of them is non-instrumental.Conversely, it canbe shown that, if P is transitive (see ?4), then the claim that at most one universalend exists entails the claim that any universal end is non-instrumental.Note also that,since by definition a universal end is pursued because of itself, B ensures that auniversal end is pursued only because of itself and is thus - modulo temporalqualifications- what Aristotle calls "complete without qualification" 1097a33-4).

I One might claim that the non-hypothetical"version" s not literal, given that theSentence has the form of a conditional(cf. Wedin 1981: 250). But this considerationis not decisive: in an appropriate ontext (e.g., right after supportingp), asserting"ifp then q" may amount to puttingforward the argument"p; thus q".




beginning of 1.7 Aristotle writes as if he has not yet proven that a uniqueuniversal end exists: "if there is some end of everything achievable in

action, the good achievable in action will be this end; if there are more

ends than one, [the good achievable in action] will be these ends"

(1097a22-4; Irwin's brackets). This passage, however, is hardly conclu-

sive: although Aristotle does mention the possibility that more than one

universal end exists, he fails to mention the possibility that no such end

exists.'0 So he seems to take it for granted that at least one universal end

exists, which is precisely what A says. Now in support of the non-hypo-

thetical version one might note - as Ackrill (1974/1999: 68) does note -that right after the Sentence Aristotle writes: "Then surely knowledge of

this good also carries great weight for [determining the best] way of life;

if we know it, we are more likely, like archers who have a target to aim

at, to hit the right mark" (1094a22-4; Irwin's brackets). This passage,

however, is hardly conclusive either: Aristotle may be saying here that,

on the hypothesis that the best good exists, knowledge of this good is very

important (cf. Reeve 1992: 1 10).

Against the non-hypothetical version one might note - as von Wright

(1963: 89), Wedin (1981: 246), and Williams (1962: 292; cf. Kirwan1967: 102) do note - that, according to Aristotle, several ends are pur-

sued because of themselves: "Honor, pleasure, understanding, and every

virtue we certainly choose because of themselves" (1097b2-3). How then

could Aristotle accept premises entailing (as A and B on either of the

above versions of the literal interpretationdo) that exactly one non-instru-

mental end exists? I reply that an end pursued because of itself may also

be pursued because of some other end and thus fail to be non-instrumen-

tal. Reeve (1992: 108) gives the example of playing the cello both for the

sake of playing the cello and for the sake of contributing to a string quar-tet (see also Urmson 1988: 11). Aristotle himself, after saying that honor

etc. we choose because of themselves, says that "we also choose them for

the sake of happiness" (1097b4; emphasis added). So the claim that

exactly one non-instrumentalend exists is compatible with the claim that

0 In response one might claim that the phrase "if there are more ends than one"

does not refer to a case in which more than one universal end exists, but ratherrefersto a case in which more than one non-universal but no universal end exists. I grant

that this is a possible reading, but it is not the only possible reading, so my point

remains that the passage in question does not provideconclusive support or the hypo-

thetical version.




several ends pursuedbecause of themselvesexist: it is possible that allbut one of the latterends are instrumental.

One mightalso object to the non-hypotheticalersionby arguing hat,becauseA is prima acie implausible, t is unconvincingo claim- as thenon-hypothetical ersionaccording o the objectiondoes - thatAristotlein (the vicinityof) the Sentencejust assertsA withouttryingto supportit. I replythat rightbeforethe SentenceAristotledoes provide he begin-ning of an argument or A. Aristotlestartswith the claim thatthereare(what I call) subordination chains between pursuits (actions, crafts, sci-

ences); e.g., bridlemakingis subordinate o horsemanship,which is inturnsubordinate o generalship 1094alO-4).Aristotlecontinueswith theclaimthatthe ends of subordinate ursuits repursuedbecauseof the endsof superordinate ursuits 1094al4-6). But then it can be proventhat, ifall maximal subordination hains contain some "highestruling"pursuitwhich has a uniqueend pursuedbecause of itself, then this end is uni-versal."AndshortlyaftertheSentence,Aristotle ries to support heclaim

that there is indeed a highest ruling pursuit,namely political science(1094a27-bll). Now regardlessof what one thinks of the merits of the

above argument,ts existencecastsdoubt on the claim that(according othe non-hypothetical ersion)Aristotle does not try to supportA in the

vicinityof the Sentence.MaybeAristotleregarded he argumentas pre-liminaryand incomplete.If so, then maybe a third, ntermediate ersion

of the literalinterpretations preferable:n the SentenceAristotlestatesA neitherpurelyhypothetically or with totalconfidence,but ratheron the

basisof a tentativeargument. n any case, as I said, I do not wish to take

a stand;my primary oal in this section was to argue contraryo Wedin

(1981) - that the hypotheticalversion of the literal interpretations not

the only viable version.

" More precisely, it can be proven that: if (1) for any pursuits m and n, and for

any ends x and y, if x is an end of m, y is an end of n, and there is a subordination

chain from m to n, then x is pursued because of y, (2) there is a pursuit n such that

(a) for every pursuit m other than n there is a subordinationchain from m to n and

(b) n has a unique end, which is pursued because of itself, and (3) every end is anend of some pursuit, then (A) there is a universal end. (The proof is straightforward,

so the fact thatAristotle does not go over it raises no significant objection to my claim

that Aristotle tries to supportA in the vicinity of the Sentence.)




4. An objection to the literal interpretation

So far I have not examinedthe parentheticalnference n the Sentenceto

the effect that, if [not-B]we do choose everythingbecauseof something

else, then[C] "itwill go on without imit"."2 y contraposition,he claim

is thatnot-CentailsB, and this looks unproblematic:f it does not"go on

withoutlimit", in otherwords if there is no infinitepursuit-chain under-

stood as an orderedset of ends each memberof which - except for the

last, if a last one exists - is pursuedbecause of the next member), hen

every pursuit-chain as finitelymany members,and the last memberof

every maximalpursuit-chainunderstoodas a pursuit-chainhat is not a

subset of any otherpursuit-chain)s a non-instrumentalnd."3This rea-

soning is invalid,however:a maximalpursuit-chainwith finitely many

membersmay fail to have any non-instrumental ember.This is because

such a pursuit-chainmay be a pursuit-circle; or example, a is pursued

because of b, b is pursuedbecause of c, and c is pursuedbecause of a

(Pab & Pbc & Pca).'4So on the literal interpretation ristotledoes com-mit a fallacy in the Sentence afterall, although a fallacy different rom

the one of which he is commonlyaccused.

Is this a significantobjection o the literal nterpretation? ot if the fal-lacy which the interpretationttributes o Aristotle is sufficientlysubtle:

even the fatherof logic is not infallible, so an interpretationn which he

commitsa subtlefallacy need not be flawed.There s indeedevidence that

the fallacy is subtle:apparentlyt was missed by severalcommentators,

arguably ncluding Apostle (1975: 206 n. 10), Aquinas(c1271/1993: 7),Broadie (1991: 13), Burnet(1900/1973: xlvi-xlvii), Kenny (1966: 94),

Kirwan1967:107),andUrmson 1988: 10).Othercommentators, owever,

12 There is a furtherparentheticalnference in the Sentence, from C to [C'] "desirewill prove to be empty and futile"(cf. Aquinas c1271/1993: 8). Accordingto Wedin,in the Sentence Aristotledoes not provide an argument or B but states instead B hypo-thetically, because "the premise that [not-C'] our desire is not empty and vain [is] afrankly implausible contention save perhapsto those already versed in the ways ofvirtue"(1981: 244). I am not sure I agree, but I do not address this issue in the paper.

It is apparentlyon the basis of such a reasoningthat commentators ypicallywriteas if B were a universalratherthan an existential claim (see footnote 5).

14 In response one might argue that (1) Aristotle assumes there are infinitely many

ends (otherwise a worry about infinite pursuit-chainswould not arise - or so theresponse goes), and (2) there cannot be an infinite pursuit-circle.But even if (1) and(2) are granted, there can still be an infinitenumber of finite pursuit-circles, ike this:Pab & Pbc & Pca, Pde & Pef & Pfd,.... Therefore, even assuming there are

infinitely many ends, the inference from not-C to B is invalid.




apparently oticethe problembuttryto exonerateAristotle.Oneway outis proposedby Wedin,who claimsthat"inthe idiomof PhysicsIII, vi anoperation s infinitewhichreturnsuponitself in a circular ashion" 1981:249; see also Broadie & Rowe 2002: 264). If not-C is understoodasentailingthat no pursuit-chains "infinite"n Wedin'ssense, then not-Centails that no pursuit-circle xists: a pursuit-circlewith finitely manymembers) s an "infinite" ursuit-chain. here s a problem,however,withWedin'sway out. An end pursuedbecause of itself corresponds o "an

operation... which returnsuponitself in a circular ashion"and thus to

an "infinite" ursuit-chaina pursuit-circle ith a singlemember).So not-C on the above understandingntails that no end is pursuedbecauseofitself. But as we saw in ?3, according o Aristotlesome ends are pursuedbecause of themselves.So I find Wedin'sway out unattractive.'5

Another way out is based on Williams's (1962: 290) claim thatAristotleassumesP (the relation"beingpursuedbecauseof") to be (1)transitive and (2) what I call quasi-asymmetric:

(1) VxVyVz((Pxy & Pyz) -+ Pxz);

(2) VxVy((x?y& Pxy)_-Pyx).

Williams(1962: 291) derives from these two assumptions he result thatno pursuit-circle aving at least two membersexists, and Kraut(1989:204) uses a similarreasoning althoughhe does not mentionthat transi-tivity is needed) o vindicate he inference romnot-Cto B. Unfortunately,however,this way out for Aristotlespells trouble or the literalinterpre-tation. This is because (1) and (2), togetherwith A, sufficefor the con-clusion of Theorem1: B is notneeded.Thisis establishedby thefollowingtheorem.

'5 Wedin might respond by claiming that, according to Aristotle, only pursuit-cir-cles with at least two members count as infinite. I would reply along the lines of what

I say in footnote 18 and the correspondingtext. Moreover, it seems that the reasonAristotle adduces for not-C, namely that if C is true "desire will prove to be emptyand futile" (see footnote 12), may exclude pursuit-chainswith infinitelymany mem-bers but need not exclude pursuit-circles with one or more members). Broadie and

Rowe might object:"In the circularcase, I could actually gain all the things I desire,

but since I want none of them for its own sake, I gain nothing I really want" (2002:264). I reply that this reasoningdoes not exclude pursuit-circlessome (or all) mem-bers of which are also pursuedbecause of themselves (e.g., Pab & Pbc & Pca & Pcc)and thus does not guarantee the existence of a non-instrumental nd.




Theorem2. If (A) there is a universalend, (1) P is transitive,and (2) P is quasi-asymmetric,then there is a unique universalend, which is also the unique non-

instrumental nd.16

But if B is redundant,henthe literal nterpretationounders,becausethen

B is after all "absurdly placed and serve[s] no purpose" (Ackrill

1974/1999:68). Now one thingto be said in responsewas said by Wedin

(1981:243n. 1):Williamsntroduces substantivedditional remiseswhich

in effect give us a completelydifferentargument".'7his response,how-

ever, is not fully satisfactoryf it turnsout that Aristotledoes assume(1)

and (2). Kirwan(1967: 101) notes that Aristotledoes not actuallystate(1) or (2). Williams's statedgroundfor claimingthat Aristotleassumes

(I) and (2) is "[t]hediscussion of the architectonic elationsin cc. 1, 2

init." (1962: 290). I do not know whatexactlyWilliams has in mind,but

maybeit is somethingclose to what Kraut 1989: 201) proposes:

If A is desirable for the sake of B, then B cannot also be desirable for the sake

of A. For, Aristotleholds, when A is for the sake of B, B is more desirablethan

A (1094al4-16). And if B is more desirable than A, then A cannot also be more

desirable than B.

Kraut s attributingo Aristotlethe principle hat, if x is pursuedbecause

of y, theny is morechoiceworthyto use Irwin'stranslation f hairetotera)

thanx. Given,however,that no end is morechoiceworthy hanitself, this

principle has the (unacceptable)consequence that no end is pursued

becauseof itself. Krautmightrespondby restrictinghe principle o cases

in which x ? y, andby notingthat the restrictedprinciple ufficesfor the

quasi-asymmetryf P. But what would be the rationale or rejectingthe

unrestrictedwhile acceptingthe restrictedprinciple? f one findspursuit-

circles with a single memberunproblematic, hy findpursuit-circleswith

16 Proof. If there were two distinct universal ends a and b (so a?b), then every end,

hence also b, would be pursuedbecause of a (so Pba), and every end, hence also a,

would be pursuedbecause of b (so Pab); but then P would not be quasi-asymmetric.

It follows (given A) that there is a unique universalend, say a. This end is also (see

proof of Theorem 1) the only candidatefor being a non-instrumental nd. Now if a

were instrumental, hen it would be pursued because of some other end c. But since

every end is pursued because of a and P is transitive, if a were pursuedbecause of cthen every end would be pursuedbecause of c, so c would also be universal. It fol-

lows that a is the unique non-instrumental nd.

1' Let me clarify, however, that Williams did not formulate, and Wedin did not

address, Theorem2 or the currentproblem for the literal interpretation.



126 PETERB. M. VRANAS

two membersproblematic?'8 lthoughAristotledoes say "morechoice-worthy", t seems reasonable o understand im as meaning"at least as

choiceworthy" andon such an understandinghe principledoes not pre-clude any pursuit-circles. o I also find Kraut'sway out unattractive.prefer o saythatAristotledoescommita fallacyin inferringB fromnot-C,although he fallacy is subtleand thusexcusable.'9

5. Conclusion: Virtues and vices of the literal interpretation

The literal interpretation as several virtues.First, it is literal: it makessense of a passagewhichsomecommentatorsind"confusedly xpressed"(Williams 1962: 292) withoutdoing violence to Aristotle's text. Second,it is charitable: t acquitsAristotleof a fallacy of whichhe is commonlyaccused.Third, t is parsimonious:t vindicatesAristotle'sprimarynfer-ence in the Sentencewithoutrelyingon extraassumptionsike the tran-sitivityand quasi-asymmetry f P or the non-instrumentalityf universalends (the latterbeingderivedrather han- as Cooper 1975:92) claims

"take[n] or granted").Fourth, t is flexible: it is neutralbetweena hypo-

thetical,a non-hypothetical, nd an intermediate ersion,and also - asWedin(1981: 261-2) argues- betweendominantand inclusive accounts

of the best good.

The literalinterpretations not flawless,however. It is not completelycharitable: t does attribute o Aristotle a fallacy, althougha subtleonewhich occurs in a parentheticalnference.And it is not completelyiteral

either: it understandsAristotle's "morechoiceworthy"as "at least as

18 As an analogy, if one accepts the possibility that a thing creates itself, then why

reject the possibility that two (distinct) things create each other? It seems to me that

those who find pursuit-circles with two (or more) members problematichad better

replace talk of ends pursuedbecause of themselves with talk of ends pursuedbecauseof no end (analogous to uncreated as opposed to self-created- things); in this way

they would avoid pursuit-circlesaltogether." According to Bostock, Aristotle's argument from not-B to C "requires the

stronger [than not-B] premiss: if everything that we choose is chosen only for the sake

of something further" 2000: 9). This suggests a third way out for Aristotle: under-

stand not-B as Bostock's "strongerpremiss".I see, however, no basis for the word

"only"in the Greek text.

Moreover,if B is understood as the claim that we do

notchoose everythingonly because of somethingelse, i.e. -,Vx(-,Pxx & 3yPxy), then Bis equivalent to 3x(Pxx v Vy-lPxy) and is thus a trivial consequence of A (since A

entails 3xPxx).




choiceworthy".Nevertheless,despitethese flaws, the literalinterpretationseems to be on balancethe best availableinterpretationf the Sentence.

Iowa State University

References

Ackrill, J. L. (1973). Aristotle's Ethics. New York: HumanitiesPress.(1999). Aristotle on eudaimonia. In N. Sherman Ed.),Aristotle's Ethics: Critical

Essays (pp. 57-77). New York: Rowman & Littlefield.(Originally published 1974.)Anscombe, G. ElisabethM. (1957). Intention.Ithaca,NY: Cornell UniversityPress.- (1967). An Introduction to Wittgenstein's Tractatus (3rd ed.). London:Hutchinson.

Apostle, Hippocrates G. (Trans.). (1975). Aristotle: The Nicomachean Ethics.Dordrecht:Reidel.

Aquinas, St. Thomas (1993). Commentaryon Aristotle's Nicomachean Ethics (C. I.Litzinger,Trans.). Notre Dame, IN: Dumb Ox Books. (Originallywrittenc1271.)

Bostock, David (2000). Aristotle's Ethics. New York: Oxford UniversityPress.Broadie, Sarah(1991). Ethics with Aristotle. New York: Oxford University Press.Broadie, Sarah, & Rowe, Christopher(2002). Aristotle: Nicomachean Ethics. New

York: Oxford University Press.Burnet,John (Ed.). (1973). The Ethics of Aristotle. New York:Arno Press. (Originally

published 1900.)

Cooper,JohnM.(1975).ReasonandHumanGoodinAristotle.Cambridge,MA: HarvardUniversity Press.

Darwall, Stephen L. (1998). Philosophical Ethics. Boulder, CO: Westview Press.Engberg-Pedersen, roels(1983).Aristotle'sTheoryof MoralInsight. Oxford:Clarendon

Press.

Gauthier,Rene A., & Jolif, Jean Y. (1970). L'Ethique a Nicomaque: Introduction,Traductionet Commentaire(2nd ed.). Tome II: Commentaire. Premiere Partie:Livres I-V. Louvain: PublicationsUniversitaires & Paris:Beatrice-Nauwelaerts.

Geach, Peter T. (1972). Historyof a fallacy. In P. T. Geach, Logic Matters(pp. 1-13).Berkeley, CA: University of CaliforniaPress. (Originallypublished 1958.)

Hardie, William F. R. (1965). The final good in Aristotle's Ethics. Philosophy, 40,277-295.

(1968). Aristotle's Ethical Theory.Oxford:ClarendonPress.Hughes, GerardJ. (2001). Aristotle on Ethics. New York: Routledge.Irwin, Terence (1977). Plato's Moral Theory: The Early and Middle Dialogues.

Oxford: ClarendonPress.

(Trans.). 1999).Aristotle:NicomacheanEthics (2nded.). Indianapolis, N:Hackett.Joachim,HaroldH. (1951). Aristotle:TheNicomacheanEthics(D. A. Rees, Ed.).Oxford:

ClarendonPress.

Kenny, Anthony (1966). Happiness.Proceedings of the AristotelianSociety, 66, 93-102.

Kirwan,Christopher1967). Logic andthegood in Aristotle.ThePhilosophicalQuarterly,17, 97-114.




Kraut, Richard(1989). Aristotleon the HumanGood. Princeton:PrincetonUniversity

Press.

Rackham,Harris(Trans.). (1934). Aristotle: The Nicomachean Ethics (Loeb Classical

Library,Vol. XIX, 2nd ed.). Cambridge,MA: HarvardUniversity Press. (Originally

published 1926.)

Reeve, C. D. C. (1992). Practices of Reason:Aristotle's Nicomachean Ethics. Oxford:

ClarendonPress.

Robinson,Richard (1964). An Atheist's Values. Oxford: ClarendonPress.

Sparshott, Francis (1994). Taking Life Seriously: A Study of the Argument of the

Nicomachean Ethics. Toronto:Universityof Toronto Press.

Urmson,James 0. (1988). Aristotle's Ethics. Oxford:Blackwell.von Wright,Georg H. (1963). The Varietiesof Goodness. London:Routledge& Kegan

Paul.

Wedin, Michael V. (1981). Aristotleon the good for man. Mind, 90, 243-262.

Williams,BernardA. 0. (1962). Aristotleon thegood:A formalsketch.ThePhilosophical

Quarterly,12, 289-296.

Peter B. M. Vranas - Aristotle on the Best Good is Nicomachean Ethics 1094a18-22 Fallacious

Documents

Transcript of Peter B. M. Vranas - Aristotle on the Best Good is Nicomachean Ethics 1094a18-22 Fallacious