plato's division of goods in the republic

7/30/2019 Plato's Division of Goods in the Republic

http://slidepdf.com/reader/full/platos-division-of-goods-in-the-republic 1/28

Plato's Division of Goods in the "Republic"

Author(s): Robert HeinamanSource: Phronesis, Vol. 47, No. 4 (2002), pp. 309-335Published by: BRILLStable URL: http://www.jstor.org/stable/4182706 .

Accessed: 13/08/2011 15:45

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/4182706?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/4182706?origin=JSTOR-pdf

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



Plato's Division of Goods in the Republic

ROBERT HEINAMAN

ABSTRACT

In the Republic Plato draws a distinction among goods between (1) those that

are good in themselves but not good for their consequences, (2) those that are

good both in themselves and for their consequences, and (3) those that are not

good in themselves but are good for their consequences.This paper presents an

interpretation f this classification,in particular ts applicationto the case of jus-tice. It is argued that certain causal consequences of justice as well as factors

that are not causal consequences of justice are relevant in explaining why justice

is good in itself; and that it is only the reputation or justice and the causal con-

sequences that follow from thatreputation hat are relevant in explaining why it

is good for its consequences.

In the Republic, Glauconand Adeimantusask Socratesto prove that a

person s betteroff by beingjust rather hanby being unjust.'Whenthey

explainhow they want Socratesto provethis, they say the following:

1. Socrates should show what effect justice and injustice have on the soul irre-spective of reputation 358b, 366e).

2. Socrates should praisejustice itself (358d, 366e).

3. Socrates should show how justice in and by itself affects a man for good, and

how injustice in and by itself affects a man for evil (367b, d).

Therefore, o prove thatjustice is superior o injusticePlato must provethatjustice is an intrinsicgood and injustice s an intrinsicevil. There isnothing else (prior to Rep. 612b) that Plato attempts o argue over andabove this in tryingto establish thatthejust life is betterthan the unjust

life.Hence,to understand lato'sargumentt is crucialto understandwhat,

in Plato'sview, makes it the case thatjustice is an intrinsicgood. And tounderstand his we must understand he division of goods set out at thestart of Book II. There Plato divides goods into those that are goodin themselvesbut not for theirconsequences, hose thatare good both inthemselves and for their consequences,and those that are not good inthemselvesbut are good for theirconsequences.

AcceptedMarch 2002I Republic 348b, 358c-d, 366e-367a, 367b, d, e; cf. 347e, 352d.

? KoninklijkeBrill NV, Leiden, 2002 Phronesis XLV1114

Also available online - www.brill.nl



310 ROBERT HEINAMAN

There is disagreementonhow this classification s to be understood.

Two interpretationsave been proposed:

1. Plato divides goods into

i. things which are good independently of any consequences they may

have,

ii. things which are good both independently of any consequences they

may have and because of certain consequencesthey have, and

iii. things which are good only because of certain consequencesthey have.2


i. things which are good because they have one kind of consequence,

ii. things which are good both because they have a consequence of the

sort specified in (i) and because they have a second kind of consequence,

and

iii. things which are good solely because they have the second kind of con-

sequence.3

I believe thatboth interpretationsre wrong, and will arguefor the fol-

lowing view:


i. things which are good (a) independentlyof any consequences they mayhave, and/orare good (b) because they have one kind of consequence,

ii. thingswhich are good both because of (i) and because they have a second

kind of consequenceother than that referred o in (i)(b), and

iii. things which are good solely because they have the second kind of

consequence - that specified in (i)(b).

Or,at least,this is what is requiredn order o understand hat Platosays

about the cases of justice and injusticeand the Republic'sargumenthat

justice pays. It may well be that a coherentgeneral interpretation f

Plato's divisionof goods is not possible.

2 C. Kirwan,"Glaucon'sChallenge",Phronesis 10 (1965), 162-73; T. Irwin,Plato's

Moral Theory(Oxford: ClarendonPress, 1977), 184 (hereafterPMT);J.D. Mabbott,

"Is Plato's RepublicUtilitarian?"n G. Vlastos (ed.), Plato ll (Notre Dame:University

of NotreDame Press,1978), 57-65;C.D.Reeve, Philosopher-Kings,Princeton:Princeton

UniversityPress, 1988), 282 n. 19.

3 M.B. Foster, "A Mistake of Plato's in the Republic,"Mind 46 (1937), 386-93;

David Sachs, "A Fallacy in Plato's Republic," in N. Smith (ed.), Plato: Critical

Assessments(London:Routledge, 1998), 208-10; N. White, A Companionto Plato's

Republic(Oxford:Basil Blackwell, 1979), 78; "TheClassificationof Goods in Plato's

Republic,"Journal of the History of Philosophy 22 (1984), 393-421; J. Annas, An

Introductionto Plato's Republic (Oxford:Clarendon Press, 1981), 348-9; N. Pappas,

Plato and the Republic (London:Routledge, 1995), 54, 156.



PLATO'S DIVISION OF GOODS IN THE REPUBLIC 311

To avoid confusionit is necessaryto distinguishbetweenPlato's useof the word"consequence" nd the contemporary otion of a causalcon-

sequence.For, as I will argue,while Plato calls (e.g.) justice an intrinsic

good,4and hence something hat is good independently f what he calls

"consequences," ts intrinsic goodness is not independentof what we

would call its causalconsequences.So I will distinguishbetween"conse-

quences" consequences rrelevant o a thing's intrinsicgoodness- and

causal consequences."Consequence"n its own will always refer to the

typeof consequencewhich Plato has in mind. On my interpretation,ome

causal consequencesof a thing maybe excludedfrom the "consequences"of that thing and,hence,may determinet to be somethingwhich,for that

reason, is intrinsically good or evil.5

I

I begin by arguing or two claims. First,the descriptionsof class (iii) in

the above interpretations ail to make clear a crucial point: anything

belongingto this class must be an intrinsicevil. Secondly, the fact that

somethinghas good causal consequencesmay be partof what makes it"good in itself."

Membersof the ThirdClass of Goodsare IntrinsicEvils

Here is what Plato has to say aboutthe differentkinds of good when he

introduces hem at 357b-358a:

First class of goods:

1. We do not welcome them for their consequences.

2. We do welcome them for the sake of themselves.

3. Examples: 1. enjoyment(@oXaipetv)2. pleasures (ai Bovai) which have two features:

a. they are harmless

b. nothing comes to be from them except enjoyment (ro

xaipElv).

I I use "intrinsic good" and "instrumental good" to mean what Plato means

by "somethingthat we welcome for itself' and "something that we welcome for itsconsequences."

I I assume that Plato divides evils (KaKc) in a way that matches his division of

goods, and in particular hat the manner n which Plato applies it to justice also applies

to the case of injustice.



312 ROBERT HEINAMAN

Second class of goods:

1. We welcome them for the sake of themselves.

2. We welcome them for the sake of their consequences.

3. Examples: thought,sight, health.

Third class of goods:

1. They are painful or laborious(Firi1Lova).

2. They benefit us because of their consequences.3. We would not choose them for themselves.

4. We would choose them for the sake of their consequences.

5. The consequences for the sake of which we would choose them includerewardsand other benefits that they produce.

6. Examples:exercise, receiving medical treatmentwhen ill, the practiceof med-icine and other ways of making money.

Plato draws up similar classifications in other dialogues. The Gorgias(467c-468a)presentsa tripartite ivision betweengoods, evils and inter-mediates,where intermediatesan be good, evil or indifferent, ependingon the particular ircumstances.TheRepublic's hirdclass of goods doesnot includeanything hat could be regardedas something hatis in itselfindifferent r intermediate. hat class is notadequately pecifiedby notingthat its membershave consequences hat areintrinsicgoods.6It is also es-sential o thisclassthat tsmembers re ntrinsic vils. AsCrossandWoozleysay, they are thingswe choose "in spite of whatthey arein themselves."7

Several things make this point clear. First,membersof this class arecharacterized s painfulorlaboriousEniiovov: 57c,358a, 364a;cf. 365b).Secondly,they are difficultor irksome(XaXe7o6v:58a, 364a; cf. 364d).8

Thirdly, n themselves hey are to be avoided(cxuro1' aito (pmKEtCov)

6 This applies to Terry Irwin's characterizationof the thirdclass in his books on

Plato's ethics, Plato's Moral Theory, and Plato's Ethics (Oxford: Oxford UniversityPress, 1995) (hereafterPE). PMT (184) says the thirdclass of goods contains those"chosen only for their consequences." PE (181) says the third class includes thosegoods "we value for their consequences but not for their own sake." Both characteri-zations fail to exclude from the third class such things as walking and runningthatare in themselves indifferent.This misunderstandings important ince, as we will seelater, it underminesIrwin's main objection to the view that the fact that x has causal

consequences which are goods may be relevant to determining hat x is good in itself.PE (185) may no longer maintainthe same view as PMT. But even there, doing whatis just, in the view set out by Glaucon, is described as something we are averse tobecause we are "giving up something attractive."That is, Irwin sees it as involving

the absence of a good, not the presence of an evil.R.C. Cross and A.D. Woozley, Plato's Republic (London:Macmillan, 1971), 66.

8 The references to 364, which is partof Adeimantus' speech, shows that no dis-tinction can be drawn here between what Glaucon and Adeimantus say.



PLATO'S DIVISIONOF GOODS IN THE REPUBLIC 313

(358a). "To be avoided" (pEVuK'dov)s the oppositeof "worthyof choice"(aipexov),and as it is what is good in itself that is worthchoosing for

itself, so it is what is evil in itself that is to be avoidedfor its own sake.9

Thereis no otherexplanationof why this thirdclass shouldbe described

as "to be avoided."Since membersof this class are good for theircon-

sequences, t could not be becauseof theirconsequences hatthey are"to

be avoided."In any case, the claim is that they are to be avoided"for

themselves," o theirconsequencesare irrelevant o the questionof why

they are to be avoided.The only remainingalternativewouldbe thatthis

thirdclass is - in itself - neutral,but thenagaintherewouldbe no basisfor describing ts membersas in themselvesto be avoided:why should

walkingand sitting- theGorgias' (468a) intermediates be describedas

in themselves o be avoided?

Nor would the first two descriptionsof membersof this class make

sense if they were thingswhich are in themselves ndifferent.For exam-

ple, walkingis not in itself painfulor laborious,or difficultor irksome.

Fourthly,when Glauconsays that mostpeopleput justice intothe third

class of goods (358a), he says thattheyregard t as something hatis not

good but ratheras somethingthat is necessary(avayccatov) nd they doit unwillingly(a`ov?E;: 358c, 359b; cf. 360c, 366d). Walkingor sitting,

which belong to the Gorgias' intermediates, re not things that could be

said to be, in themselves,things that we do unwillingly."Unwillingly"

means not merelythat we have no positivedesire to do the thingin ques-

tion but thatwe have a positivedesire not to do it. It is what we regard

as evil that we do unwillingly.Peopleconsider temsin the thirdcategory

as "notgood"because,although hey belong to the third class of goods,

this means only thattheyare instrumentaloods.Theyareintrinsic vils.'0

Take the exampleof treatmentwhen ill, which Glaucongives (357c)

9 Cf. Aristotle,Nicomachean Ethics 1104b30-32, 1153bl-2, 1172bl8-20, 1175b24-

25, Topics HI, De Anima431al5-16.

10 The fact that the many can say thatjustice is "fine"(icaX6v:364a) does not mean

that they do not regardit as an evil (KWaCo).Polus in the Gorgias is committedto the

same sort of position, which allows the same thing to be both fine and evil, or both

disgracefuland good (cf. Rep. 348e). He holds (474c-d; cf. 482d-e) that doing what

is unjust is a good (ayao6v) but disgraceful, and hence not fine (KXAk6), since dis-graceful is the opposite of fine. Polus thinks that the fine must be distinguishedfrom

the good and the evil must be distinguishedfrom the disgraceful,as he holds that to

do injustice is good but disgraceful and to suffer injustice is evil but not disgraceful.

Cf. Gorgias 482d-e, 488e-489a.



314 ROBERT HEINAMAN

to illustrate he kindof thinghe wants to put into the thirdclass. A sim-ilar example is given in the Gorgias(467c), where Socratesmentions ak-

ing a drug and suffering or the sake of health.This is not somethingwe

regardwith indifference,as we regard walking or sitting (Gorg. 468a).

Taking a painful drug is given as an exampleof a positiveevil and we

do it because its good consequencesoutweighits intrinsicevil."

When Glauconsays that the many put justicebetweenthe best andthe

worst (359a-b), he does notmeanthatit belongsto the class of thingsthat

are indifferent.Whathe says is slightly misleadingsince he is using "jus-

tice"to refernot simply to acting justly but to the combinationof doingwhat is just and not being treatedunjustly 359a2). The situationsbeing

comparedare

1. acting unjustlyand not paying any penalty

2. acting justly and not being treatedunjustly

3. being treatedunjustlywithoutthe powerof revenge.

Glaucon's statement hat the many put justice betweenthe best and the

worst means that 2 is intermediaten orderof preferencebetween he best-

1-

and the worst-

3 (cf. 344c). There is no warrant for taking theview attributed o the many in this passage to contradict their view,

reported n the previouspage (358a), that ustice belongsto the thirdclass

of goods.Plato's division betweenthreeclassesof goods picksout threeof a pos-

sible six classes of goods and evils:

1. Thingsthatare good in themselvesandgood for theirconsequences.2. Things thatare good in themselvesandhave evil consequences.

3. Things thatare good in themselves and have no good or evil conse-

quences.4. Thingsthatare evil in themselvesand evil for theirconsequences.5. Thingsthatareevil in themselvesand have good consequences.

6. Things that are evil in themselvesand have no good or evil conse-

quences.

(1) is Plato's second class, (3) his first class, and (5) his thirdclass of

""goods."Glaucon and Adeimantuswill present the view that justice

belongsto (5) andinjusticebelongsto (2),12while Socrateswill arguethat

I Cf. Protagoras 354a-c.12 And in the best circumstanceswhere perfect injustice is attainable the evil con-

sequences can be avoided.




justicebelongs to (1) andinjusticebelongs to (4). Thingsthat are neutralbut have good or evil consequencesplay no role in the argumentof the

Republic,and they are passedover in silence.'3

IntrinsicGoodnessand CausalConsequences

Does Plato believe that all causal consequencesof an item x are conse-

quencesof x, and hence irrelevant o the issue of x's intrinsicgoodness

or evil? Or do some causalconsequences all outside of the class of con-

sequences(as that term is used by Plato)'4so thatthey may berelevant

to determiningwhethersomething s good in itself? I think t is clearthat

Plato allows thatcertainkinds of causal consequencescan establish that

something s good in itself.

First argumentthat the casual consequences of x may be relevant to its intrin-

sic goodness

My first argumenthat the causal consequencesof somethingmay be rel-

evant to establishing ts intrinsicgoodnessor evil is based on two claims:

(1) The argumentsof Glaucon and Adeimantus n defense of injustice are meantto prove that it is good in itself, and hence they show what sort of consid-

erationmay be relevant in establishingthat something is good in itself.

(2) The argumentsoffered by Glaucon and Adeimantusto show that injustice is

good in itself appeal to the causal consequences of injustice.

I offer two arguments in support of (1). To begin with, all partiesto the

discussion in the Republic agree that the consequences of justice are goods

and the consequences of injustice are evils.'5 Several points show this.

(i) Socrates obviously believes that the normal consequences of justice

are good (612a-614a), and the many are said (358a) to put justice into thethird class of goods, viz. the class of things which are evil in themselves

13 There is a passing referenceto things that are neither good nor evil in Book X's

argumentfor the immortalityof the soul (609b).

'4 Plato uses "'a uanoIakivovta" nd "taz 'ytyvo6jeva &o x" as synonyms (357b6,

8, cl-2, dl-2, 358a2).

'5 Cf. Reeve, Philosopher-Kings,25. Foster("A Mistake of Plato's in the Republic,"

387) claims that Glaucon and Thrasymachusregard justice as good neither in itselfnor for its consequences. But his view is based on the claim that Plato regards the

reputation or justice as a third class good whose consequences are distinct from the

consequencesof justice. I'll argue below thatPlato identifies the consequencesof jus-

tice with the consequences of the reputation or justice.



316 ROBERT HEINAMAN

but which have goodconsequences.

So both Socratesand themany agreethat the consequencesof justice are goods.

(ii) Agreementabout the consequencesof justice and injusticecomesout in Glaucon'ssetting up of the questionwhich the Republicis sup-posed to answer.For in orderto focus on the worth of justice and injus-tice themselveshe considersa case where the consequencesof justicearegiven to the unjustman andtheconsequencesof injusticearegiven to thejust man(361a-d). And the consequencesof justice areall good (362b-c;cf. 363a-d)and the consequencesof injusticeare all evil (361e-362a;cf.

363d-e).Glaucon s settingout the view of the many,so the manyas wellas Plato accept the point that the consequencesof justice are good andthe consequencesof injusticeare evil. Thus,when in Book X we finallycome to theend of theargumenthat ustice is an intrinsicgood andinjus-tice an intrinsicevil (612c), Socrates refersto Book II's detachmentofjustice and injusticefromtheir usualconsequences 360e-362c)and says(612c-d):

I grantedyour requestthat the just man should be reputedunjust, and the unjustman just, even though it would be impossible that the falseness of their reputa-

tions should escape both gods and men. I yielded in this for the sake of argu-ment, so thatjustice itself could bejudged withrespectto injusticeitself... sincethat judgement has now been made, I ask on behalf of justice the returnof thereputation t has in fact among gods and men...

(iii) Agreement between Socrates and the many on the point that the

consequences of justice are good and the consequences of injustice are

evil is also shown by Adeimantus' speech. Reversing Glaucon's argument

which explained how the many praise injustice and condemn justice, he

considers how the many praise justice and condemn injustice (362e).

When people praise justice they do not praise justice for itself but onlybecause of its good consequences (362e-363e, 366d-e). So the many agree

that justice has these beneficial consequences. With respect to the con-

demnation of injustice, Adeimantus points out that it is only condemned

for its consequences: people attribute to the unjust "all the punishments

which Glaucon enumerated in the case of the just with a reputation for

injustice, but they have nothing else to say" (363d; cf. 366e, 367b-c), i.e.

in blaming injustice they say nothing about its being evil in itself.

(iv) This agreement on the value of the consequences of justice and

injustice s also shownby thefact that the beneficialconsequencesof jus-tice which Glaucon and Adeimantusmention n Book 11are the same asthose mentionedby Socrates n Book X (612a-614a)when he pointsout




the consequences of justice and injustice. With regard to the rewardswhich men grant to the just, Socrates says:

Will you allow me to say about them [thejust] what you yourself said about the

unjust?'6 shall say that, when they get older, just men fill the public offices of

their own city if they wish [cf. 362b2, 363a3], marny nto any family they want

to [cf. 362b3, 363a3], give their children in marriage to anyone they wish [cf.

362b3], and all you said about the unjust I now say about the just.

Socrates explicitly states that he agrees with the view of the many, as

describedby GlauconandAdeimantus,aboutthe beneficialconsequences

of justice.'7 t is truethatthe consequencesof injustice istedby Glaucon

and Adeimantus 36le-362a) are not identicalwith the consequencesof

injusticementionedby Socrates n Book X (612e, 613b-e). But the pun-

ishmentsmentionedare identical(361e4-5, 613el-2), and, in any case, it

is clear that all regard he consequencesof injusticeas evils.

Sincebothsidesto theargument greeon the valueof theconsequences

16 Because, for the sake of argument, the perfectly unjust man was given the rep-

utation for justice and this reputationand its consequences are now being returned othe just man (612c-e).

17 There are differences between Books II and X with regardto the response of the

gods to justice and injustice. In the earlier passage Glaucon is setting out the view of

the many that the gods will treat well those who are really unjust because the gods

are immoral enough to be bribed by the unjust (362c; cf. 364b-365a, 365e-366b). This

is part of Glaucon's argument on behalf of the many that injustice is good in itself.

In Book X (612d-613b), with the virtue of the gods restored, their response to justice

and injustice is based on their knowledge of the charactersof the just and the unjust,

and is part of the reason why justice is good for its consequences while injustice is

bad for its consequences. The response of the gods is also used by Adeimantus in

Book II to argue that justice is good not for itself but for its consequences (363a-e).

He reverses Glaucon's procedure o examine in what way the many praise justice, and

argues that they praise it only for its consequences.

The quotedpassage's list of beneficialconsequencesof justice proves thatthe Republic

is not arguing merely that the just man is better off in the ideal city (cf. R.W. Hall,

"Justiceand the Individual n theRepublic,"Phronesis4 (1959), 151, 158).The beneficial

consequences regarding marriage and going into politics "if they wish" are not avail-

able to Plato's prime examples of just men in the ideal city, the guardians. Likewise,

591d-e's remarkson how the just man will acquire wealth could not apply to the

guardians.And Plato goes on to say that the just man he is talking about will only

go into politics if by divine luck he lives in the ideal city (592a). The argumentof

the Republic aims to show that justice pays "anyone"(589d), including the man with

the ring of Gyges (612b), so it is also wrong to say that the argument claims merely

that justice pays the philosopher.



318 ROBERT HEINAMAN

ofjustice andinjustice,'8

heirdisagreement

boutthevalueofjustice

and

injustice can only concern the intrinsicvalue of justice and injustice.In

which case it is clear that the argumentswhichGlauconandAdeimantus

bring forward o demonstratehat injustice s a good thingare meantto

show how it is good in itself.9

A second argumentor (1): Thatthe praiseof injusticecontained n the

speeches of Glaucon and Adeimantus s meant to indicateits intrinsic

goodnessis also provedby Glaucon'sstatement 358d4-6)that he wantsto hearSocratesdefendjustice in the sameway as he himself will praise

injustice.The "way" n question,the manner n which Glauconwants tohearjustice praised, is clearly specified by his immediatelypreceding

statement hat"Iwant to hearit Uustice]praised or itself' (358dl-2); and

by his previousassertion 358b) that he wants to hear the natureof jus-

tice and injustice, and their effects on the soul, not an account theirrewardsand consequences.20

So: Glauconsays thathe wants to hear Socrates praisejustice in the

same way he is aboutto praise njustice.He wants Socrates o praise us-tice by showing that it is good in itself. Therefore, he reasonsGlaucon

18 This is part of the explanationof why, when Glaucon's initial speech compares

the lives of the just and the unjust n order to assess their relativehappiness and unhap-

piness, he contrasts the extreme case of injusticewith the just man who is thought to

be unjust(360el-6). For Glaucon is arguingfor the claim that the unjust is better off

than the just, and it is in the extreme case of injustice where the unjust man can keep

his injusticehidden 361a4-5; cf. 365c-d) thatthe evil consequencesof injustice described

in 361e4-362a3; cf. 363d-e) are avoided (360e6-361b5; cf. 366b3-4). So it is only in

the extreme case of injustice where these evil consequences of injustice are avoided

that there is any plausibility in saying that the unjust life is better than the just life.

The claim that, ordinarily, he unjust life is overall worse thanthejust life is a pre-supposition of Glaucon's explanation of the origin and nature of justice at the start of

his speech. It is only because ordinary people are unable to avoid the consequencesof injustice that they prefer to make an agreementto be just and thus avoid injustice

themselves (359a6-b4), and Glaucon's argumenttakes this to be a rationaldecision.

In theory they rank higher than this compromisethe situationwhere they are unjust

and avoid punishmentfor their unjust behavior (359a6). But as things actually are,given the world and their own powers as they exist, ordinarypeople are unable to

avoid the consequences of injustice (359bl, b6). Though they regard injustice as an

intrinsicgood andjustice as an intrinsicevil, they rank (injustice + the consequences

of injustice) as worse than (justice + the consequencesof justice), and these are their

only options in the real world. But that means that they rank the consequences ofinjustice as so bad that theirevil outweighs any benefits from injustice.

'9 As noted by Cross and Woozley, Plato's Republic,69.20 A point repeated by Adeimantus(366d-367e).




goes on to give for preferring injustice to justice are reasons why injus-tice is supposed to be good 'in itself,' in the sense of the phrase that Plato

has in mind.

So we can affirm:

(1) Theargumentsf Glaucon ndAdeimantusn defenseof injustice re meantto prove hat t is good in itself, and hence they show what sort of consid-erationmaybe relevantn establishinghatsomethings goodin itself.

But we can also affirm

(2) Thearguments fferedby GlauconandAdeimantuso show that njusticesgood in itself appeal o the causalconsequencesf injustice.

To begin with, at 359c-d, while explaining why "injustice is much more

profitable to [the unjust agent] thanjustice," i.e. why injustice is supposed

to be an intrinsic good, the desire for undue gain (n4Xove4ia) is said to

be the motive for unjust action (cf. 366al-3). But clearly enough, in the

standard cases which Plato mentions such as the desire for more wealth

than is one's due, the gain one gets from unjust action is a causal conse-

quence of the action, not the action itself. It is because people want the

money that will result from unjust action that people do what is unjust.2'

Hence, Glaucon is saying that it is because unjust action produces undue

gain that unjust action is seen by people as intrinsically good.

Undue gain, i.e. the possession of more than one's due, is not a part of

the action of acting unjustly. A can steal from B and hand over the goods

to C. Then C, not A, will (let us suppose) have more than his due.22

It might be said that in this case unjust action is not an intrinsic good

to A, but I use this example merely to show that the undue possession of

a good is not a part of an unjust action but a causal consequence of unjust

action. And when, as in the normal case envisioned by Glaucon, A

unjustly takes B's goods and as a causal consequence has more than his

due, Glaucon takes this to show that injustice is an intrinsic good, i.e. the

unjust action is taken to be an intrinsic good because of a causal conse-

quence of the action.

Other passages in the speeches of Glaucon and Adeimantus reinforce

the point that the causal consequences of unjust action play a role in argu-

ing for its intrinsic goodness. At 360b-c, in the case of the man with

21 I, like Plato, will ignore cases of people who steal simply for the joy of stealing.22 This is evidently Plato's view of the matter too. For example, at 591a the pos-

session of goods is a result of (Eic) an unjust action. Cf. Laws 862c.



320 ROBERTHEINAMAN

Gyges' ring,we get some examplesof unjustactions, ncluding aking he

propertyof another.Taking the propertyof anotherwould clearly not be

"welcomed" f it did not result in one's being free to possess and use the

stolen property.Again,if I stole thegoodsand handed hemoverto some-

body else, I wouldhave committedan unjustact butto me the act would

(at least in this regard)have no value. Glaucon s explainingwhy people

regardunjustaction as an intrinsicgood,23 nd he construes he example

given with Gyges' ring as providing he reason why people regard t as

an intrinsicgood (360c-d). So unjustaction is regardedas an intrinsic

good becauseof a causal consequenceof the action.

23 Irwin (PE, 184) describes the case of Gyges' ring as one where Glaucon "points

out that if we value justice for its consequences, we must admit it is no longer valu-

able if the good consequences are removed." This may be misleading because there

is no question of removing the good consequences of justice in the case of Gyges'

ring. It is a case where the consequences of injustice are removed, i.e. the evil con-

sequences that usually ensue on unjust action, in order to show that people do not

value justice for itself. For, Glaucon says, whenever the evil consequences of injus-

tice are eliminated people will choose injustice over justice. To have a case where the

good consequences of justice are removed we would need an example in which some-one does what is just but does not receive the usual rewardsof justice. No such case

is consideredin the passage with Gyges' ring.

LaterIrwin says the following when discussing the case of Gyges' ring (PE, 187):

"Glaucon's counterfactual uppositions ... make clearer .., a consideration hat is rel-

evant to our decision in actual circumstances. If the only things that matterare the

consequences of justice and injustice, then we must prefer injusticewhen it has bet-

ter consequences."But the idea that "the only things that matter are the consequencesof justice and injustice" s irrelevant o the case of Gyges' ring. The aim of this argu-

ment is to establish that injustice is good in itself so it can hardlyassume thatthe only

things that matterare the consequences of justice and injustice.The only counterfac-

tual feature of the situation is that the usual consequencesof injustice are removed.But it is precisely because injustice is consideredan intrinsic good andjustice is con-

sidered an intrinsic evil that all, according to the argument,would choose injustice if

the usual bad consequences of injustice could be avoided. It is not a situation where

we are supposed to imagine that "the only things that matter are the consequencesof

justice and injustice."

I also disagree with Irwin's description of the overall purpose of the speeches of

Glauconand Adeimantus.He says (PE, 188) that"thefour-stageargument s carefully

constructed o show thatI cannot both advocate being just purely for its consequences

and give a good reason for being just; if I am concerned only for the consequences

of justice, I must admit that I can secure these by appearing o be just rather hanby

being just." But if the normal situationis that I cannothide my acts, then the normalsituationis that I can appear to do what is just only if I do what is just. So, as in the

social contract theory, I can both advocate doing what is just purely for its conse-

quences and give a good reason for doing what is just.




At 362b, after isting the benefitsarisingfrom the reputation or justice(cf. 613d), i.e. after listing the goods that are consequencesof justice,24

Glauconrefersto the intrinsicgood of doing whatis unjustby sayingthat

"besidesall these advantages i.e. the advantageshatfollow from the rep-

utationfor justice, i.e. the advantageswhich are normallyconsequences

of justice], he [the unjust man] benefits in the pursuit of gain (KcEp&a-

ivovta) becausehe does not scrupleto practice njustice."The gain that

results from practicing njustice s not the same as the practice tself.

Even more clearly, the benefits of injustice listed after this are both

causal consequencesof injusticeand do not dependon the reputationorjustice, and hence25 renot regardedby Plato as consequencesof justice:

Besides all these advantages [thatfollow from the reputation or justice - 362b2],

he benefits in the pursuitof gain because he does not scruple to practice injus-

tice. In any contest, public or private,he is the winner, getting the betterof his

enemies and accumulating wealth; he benefits his friends and does harm to his

enemies. To the gods he offers grandsacrificesand gifts which will satisfy them,

he can serve the gods much better than the just man, and also such men as he

wants to, with the result that he is likely to be dearer to the gods.26

All of these benefitsare unrelated o the reputationorjustice.Forexam-ple, there is no reasonwhy a wealthy man's aiding his friendsshould

dependon otherpeople thinkinghim just. The benefits derive from the

accumulation f wealth that follows on unjust action by the unjust man

and are partof why unjust action is supposedto be good in itself. But

24 To repeat, Socrates has reversed the connections between justice and injustice on

the one hand and their normal consequences on the other, so that the consequences of

justice are given to the unjustman (and the consequences of injustice to the just man).

There is a clear division between (i) the benefits for the perfectly unjustman that are

consequences of justice, i.e. those that follow from his misleading reputationfor jus-

tice (362b2-4), and (ii) the benefits of injustice itself (362b4-c6). In the case of the

just man with the reputation or injustice, only the evils thatareconsequences of injus-

tice are mentioned (361c-362a).

25 As we'll see later, the consequences of justice which Plato is concerned to rule

out in the assessment of the intrinsic goodness of justice are those that follow from

the reputation or justice.

26 Note that whereas a few lines later (363a-b) Adeimantus says that the reputation

for justice leads to popularity among the gods, here in 362c the basis for the unjust

man's being dear to the gods is not his reputationfor justice but the fact that, being

rich as a result of his injustice, he is able to offer the gods "grandsacrifices and gifts

which will satisfy them."In 363a the just man is underdiscussion and justice is praised

for its consequences, while in 362c the unjust man is at issue and it is argued that

injustice is good in itself. Cf. n. 17.



322 ROBERT HEINAMAN

these are obviouslycausal consequencesof unjustaction and not identi-

cal with it.27

Again, at 364a Adeimantus ontinuesGlaucon'sargumentby pointingout that people regard unjust action as more profitable han just action

since the unjustwill have wealth. As we have seen (he is arguing that

injustice s good in itself and assumes that its consequencesareevils), in

saying that people regard njustice as profitable,Adeimantusmeansthat

they regard t as good in itself. And it is the fact thatunjustactionleads

to the possession of wealth which is said to explain why injustice is

regardedas profitable.The possessionof wealth is a causalconsequence

of unjustaction,and the fact that unjustaction has this as a causal con-

sequence is taken to show thatunjustactionis good in itself.

Glaucon and Adeimantus are defending the view put forward byThrasymachusn BookI (357a, 358b-c, 367a). Since they defendthe view

that justice is evil in itself and injustice s good in itself, Thrasymachus'position s that ustice is evil in itself and injustice s good in itself. Thus,

Thrasymachuss said to hold the view thatjustice belongs to the thirdclass of goods (358a; cf. 358b7-c7),28o he regards t as evil in itself.

27 These examples of causal consequencesof injusticewhich are supposedto make

it good in itself rule out White's suggestion that when x is an intrinsic good because

it is the aitia of some effect, the connection betweencause and effect is a kind of non-

empiricalsynthetic necessity ("The Classification of Goods in Plato's Republic,"408).

The connection between injustice and aiding one's friends is not the same as that

between three and odd.

Also, White's view ("The Classification of Goods in Plato's Republic,"395) that

"x is good for itself" means that x "by itself" leads to happiness cannotbe right.Plato

asserts that health is good for itself but he obviously does not believe that health by

itself leads to happiness. Nor do I see any reasonto believe that Plato thinks that thetortured ust man is happy, contrary to White's position that, since justice is good in

itself, justice necessitates happiness. Likewise, although the many think that injustice

is good in itself, they think that the normalconsequences of injustice are evils - being

whipped, stretched on the rack, imprisoned, having one's eyes burned out, and being

impaled (361e; cf. 613d-e). It is unlikely that they regard such a man as happy.

28 Irwin is surprisedby the fact that when justice is put into the third class of goods,

Socratessays thatjustice "has previouslybeen condemned by Thrasymachus or hav-

ing this character" 358a). For, Irwin says, "Thrasymachus id not say thatjustice was

any sort of good at all" (PE, 181). But what Thrasymachus ays aboutordinary njus-

tice suggests that he could agree that, normally, ustice has valuableconsequences.For

he did point out the evil consequences of ordinaryinjustice (344b), i.e. injusticethatis not able to avoid punishmentand the opprobriumof others. Only injustice done in

secret (as in the extreme case of Gyges' ring) or on a large scale (as in the case of a

tyrant)can avoid these consequences (344a-c, 345a; cf. 360e-361b, 365c-d, 367b-c).




And, clearlyenough,he holds that injustice s good in itself (cf. 358e3).Hence,the reasonswhichThrasymachusrings orward or praising njus-

tice and for condemning ustice can also be taken to indicatewhat, for

Plato,makes somethinggood or evil in itself.

At 343d-344aThrasymachusointsout evil causal consequencesof jus-

tice. They are all independent f the reputationor justice, and therefore

they providereasonsfor saying thatjustice is evil in itself. The causal

consequencesof justice are listed as follows:

1. In contractswith other people, "when the partnership nds"the just

have less thanthe unjust.

2. In paymentof tax, the just man acts justly and as a resultpays more

thanothers.A resultof the action is that thejust manhas less money.

3. In the distribution f benefits thejust man gets less.

4. When holding public office, the affairs of the just man deteriorate

throughneglect and he receives nothingfrom holdinghis office since

he is honest.

5. The just man is dislikedby his householdand acquaintances ecause

he refuses to do them unjustfavors.

Clearly,at least (1), (2), (4) and (5) describecausal consequencesof jus-

tice. Injustice s said to be good becausethen the oppositeof (l)-(5) holds

(343e), so the corresponding ointsdescribegood causalconsequencesof

injusticewhich are independent f the reputationor injustice.Hence(see

below)29 hey are arguments hat injusticeis good in itself. This fits the

So although Thrasymachussays that injusticeis profitableandjustice is not profitable

(348c), his position is that it is only large scale or secret injustice that is profitable

(344c, 348b, d; cf. 345a). Likewise, it is reasonable to suppose, Thrasymachusmay

think that, ordinarily,justice has good consequences, while it is a worse option for

one who is in a position to be unjustin secret or to practicelarge scale injusticewhich

involves the ability to avoid punishment.

It is possible that the statementat 358a that "Justice has now for some time been

objected to by Thrasymachuson this score" only picks out the point that justice "is

to be avoided in itself for being difficult,"rather than that fact plus the claim that it

has good consequences.But the point remainsthatThrasymachusagrees that injustice

normallyhas evil consequences. The fact that he thinks there may be exceptions to

this rule does not affect the point since, as we'll see (and as should be clear already),

Plato does not think thatwhat he calls the consequences ofx must always or neces-

sarily follow from x.

29 Where I argue that, for Plato, it is only the good or bad causal consequences of

justice and injustice that follow from the reputation or one or the other that provide

reasons for saying that they are good or evil for their consequences.



324 ROBERT HEINAMAN

previously noted point that all parties to the discussion, includingThrasymachus, gree thatinjusticehas evil consequencesandjusticehasgood consequences.So Thrasymachus' rguments gainst ustice must bedesignedto show that it is evil in itself.

Second argumentthat the casual consequences of x may be relevant to its intrin-

sic goodness

Strongevidence that the causal consequences of x may be relevant inestablishingits intrinsic value is providedby Plato's comment on the

course of the argumentof the Republicat 612a-d.3?There,he points outthat he has hithertobeen arguing for the superiorvalue of justice overinjusticeby considering ustice and injusticethemselvesapart rom their

consequences,and that he will now, for the firsttime, turnto theircon-sequences.This provesthatall thathas gone before s understood y Platoto have presented easonswhyjusticeis good in itselfandinjustice s evilin itself.

If so, it follows that the causal consequencesof justice and injusticematter or establishing hat the firstis good in itself andthe latter s evilin itself. For

earlier 588b-590a)Platoargued hat ustaction s good becauseit promotespsychic ustice,while unjustaction is evil because t promotespsychic injustice.It is clear thatpsychicjustice is a causal consequenceof just actionand thatpsychicinjustice s a causalconsequenceof unjustaction. And according o 612a-d, the fact that ust andunjustactionshavethese effectsestablishesthat one is good in itself and the other is evil initself. Hence, the causal consequencesof somethingmay determine t tobe good or evil in itself.3'

`0 As others have recognized. See Kirwan,"Glaucon'sChallenge,"170-71; White,"The Classification of Goods in Plato's Republic,"401; Cross and Woozley, Plato'sRepublic, 62. Cf. Reeve, Philosopher-Kings, 25. White correctly points out that 612suffices to refute the view that Book IX's argument hat the just life is more pleasantthan the unjust life is not an argumentthat justice is good in itself.

31 Note too 613e-614a's contrastbetween the good consequences of justice with the"goods"providedby justice itself which make it good in itself. If a pluralityof goodsprovidedby justice itself makes it good in itself, then the presenceof the single goodjustice itself cannot be the sole thing that makes it good in itself.

Irwin (PMT, 210-11) describes the argument of 588b-590a as though it were an

argumentthat just action is good because of its consequences. He does not mention612's clear implication that 588-90, togetherwith everything else up to 612 said indefense of justice, is meant to establish thatjustice is good in itself. PE too does notdiscuss the clear implication of the passage on this point.




Further,or the samereason hat he speechesof GlauconandAdeimantusindicatewhat, for Plato, makes somethinggood in itself, the discussionof

the degenerate onstitutions nd men in Book VIII indicatewhat can make

something evil in itself. For Glaucon and Adeimantuswant Socrates to

show not only that ustice is good in itself but that injustice s evil in itself

(366e, 367b, d). As 612 shows, all that has preceded s relevant to this

proof. Hence, the discussion of unjustmen in the discussion of different

types of men and constitutions s partof the argument hat injustice s evil

in itself.

Take,for example,the extremecase of the tyrannicalman.The follow-ing evils will apply to him as causal consequencesof his injustice:the

satisfactionof his old desires no longeraffordsany pleasure 574a, 577e-

578a, 579e), he is full of fear, convulsions and pain (578a, 579b, e), he

is friendless(576a, 580a). For Plato, these causal consequencesof injus-

tice show how it is evil in itself.32

32 The fact that x may be considered an intrinsic good because of its causal con-

sequences may even be built into Plato's initial description of the first class of goods,

which are welcomed for themselves but not for their consequences. At 357b5-8 Plato

says that they are things which we do not welcome for their consequences

but welcome for their own sake, such as enjoyment(Xcipetv) and the harmless

pleasures (i'ovai) from which nothing comes to be at a later time other than

enjoyment (Xacpetv). (&XX'a a6 cxi'aoio 'Eveccxacaxa~O'gvot, olov 'O xaipslviai. al i Oval O"aat &Xap?i; alt i18EVl' TOc Vrv EtTi a xpOvov &c TaxuTa;

yiyvevatta'Uo i] XavipetvEXovra.)It is at least plausible to suppose thatxaipetv and iiSovi are distinct since the first is

said to come to be from the second. If they are distinct, then it is very plausible to

say that 'Sovii is a kind of activity andXaipEtv

is the sensation of pleasure that it

produces (cf. Sachs, "A Fallacy in Plato's Republic," 209). In which case Plato will

be saying that a kind of activity (ilovi') can be countedas an intrinsicgood because

of the pleasure (xaipetv) it causes.

Irwin (PMT, 325, n. 8) claims that "from which nothingcomes to be at a later time

other than pleasure(Xaipetv)" in the quoted sentence applies to the previousXaip&lv(in rOxaipetv cal ai 'Iovai in b7) as well as a'til8ovai. If so, Irwinargues, it can-

not be that Plato is saying that the Xaipetv which "comes to be at a later time" is a

causal consequenceof ii5ovai since then he would also, absurdly,be saying that that

second xaipetv is a causal consequence of Xcctptv, i.e. of itself. Hence, the quoted

sentence cannot be saying that pleasures that are sensations are caused by pleasures

that are activities.Irwin's objection fails because it rests on the unjustifiable assumption that "from

which nothing comes to be at a later time other than enjoyment (Xaipeiv)" cannot

modify 'jovai alone. But it can, and if it does Plato is giving two examples of thingswhich we welcome for their own sake but not for their consequences:



326 ROBERT HEINAMAN

Irwinarguesthatthe causalconsequencesof somethingcannotbe rel-evant to its intrinsicgoodness.He says (PMT,325, n. 8) that,on thisview,Plato will haveproblemsdistinguishinghe firstandsecond class of goodsfromthe thirdclass of goods. The objection s repeated n Plato's Ethics(190-91) where he says that the supposition hat the causalconsequencesof an item mighthelp to determinewhether t is an intrinsicgood "doesnot explainwhy Plato divides the second fromthe thirdclass of goods inthe way he does." For, he says, even if exercise on its own33 one ofPlato'sexamplesof the thirdclass of good) normallyproduceshealth, tis clear that Plato would put it into the third

class, not the second.Whereas,on the proposal hat the causalconsequencesof x may be rele-vant to making it an intrinsicgood, Irwinthinks,exercise should be putinto the second class since it causes a good consequenceand hence, onthe proposal,should count as good in itself.

But Irwin's objection is a resultof his own misunderstandingf thethirdclass of goods. He evidently thinksthatthey includethings thatareneutral n themselvesbut have good causalconsequences.For all he saysabout them in PMT (184) is that they are "goodschosen only for theirconsequences,"not takinginto accountthe point,arguedabove, thatanymember of the third class of goods is an intrinsicevil. The misunder-standingbecomes clear when he says (PMT,185):"WhenSocratesagreesthatjusticeis a b-good [i.e., belongsto the second class of goods], ... herejects he positionof Book I, andequallyof theGorgias andthe Socraticdialogues,whichall treat usticeas a c-good [i.e., as belonging o thethirdclass]."Irwincan say this only if he thinks that the thirdclass of goodsincludesthings thatare, in themselves,neutralor indifferent.For it wouldbe absurd o claim thatSocratesever suggeststhatjustice is an intrinsicevil, andI assume that Irwin does not ascribe such a view to Socrates.

In fact, unlike the Gorgias, the Republic's classification does notincludeany class of thingsthat are in themselves neutralor indifferent.Once we acknowledge he pointthat Plato's thirdclass of goods includes

(1) enjoyment(Qaipetv),and

(2) the harmless pleasures(fjovai) from which nothing comes to be at a later timeother than enjoyment(Qaipetv).

If this is what Plato means, then (2) may be saying that some i8ovai are intrinsic

goods because of their causal consequences.33 Irwin is addressing the proposal that the fact that a good is a consequence of x

counts for x being an intrinsic good only if x has that consequence in normal cir-cumstances,not merely if it has that consequence in certainspecial circumstances.




only intrinsicevils, the interpretationhat sees the causal consequencesofx as possibly relevantto determining ts intrinsicgoodness has no prob-

lem in distinguishing he second fromthe thirdclass of goods. All things

that belong to the first two classes of goods are not intrinsicevils, while

all things belonging to the thirdclass of goods are intrinsicevils. This

suffices to distinguish he first and second class of goods from the third

class of goods. Exercise is excluded from the second class of goodsbecause it is an intrinsicevil.

II

If Plato allows that causal consequencesof x may be relevant to deter-

miningthat x is good in itself, what are the "consequences" f x which

he wishes to rule out as irrelevant o determininghe intrinsicvalue of x?

With regardto Plato's generaldivision of goods into threeclasses, I do

not know the answer to this question.I doubt that there is a consistent

interpretation hich will handle all of Plato'sexamples.But it is the case

of justicethat interestsme here, and in this case the consequenceswhich

Plato wishes to exclude in assessingits intrinsicvalue are the reputationforjustice and the benefits that follow from the reputationorjustice. In

otherwords, the consequencesthat follow for the just man because of

otherpeople's awarenessof his justice are those that are irrelevant o its

intrinsicworth.34

The evidence for this claim is overwhelming.It can hardlybe acci-

dental that in every passagewhere Plato talks aboutexcludingthe conse-

quencesof justicefrom consideration f its intrinsicworth,35t is the rep-utation for justice and the consequences following on that reputation

which are at issue. Here are some of them:

1. 358a: After Socratessays that he puts justice into the class of goodsthatare to be welcomedboth for themselves and for theirconsequences,Glauconsays: "That s not the opinionof the many...; they would putit in the wearisomeclass, to be pursued or the rewardsand popularitywhich come froma good reputation i.e., for its consequences],but to beavoidedin itself as being difficult."

I Here I agree with Jerome Schiller, "Just Men and Just Acts in Plato's Republic,"

Journal of the History of Philosophy 6 (1968), 5.

35 Rep. 358a, 361b-362c, 362e-363a, 365b, 366d-e, 367b-e, 392b, 392c, 444e-445a,

580b-c, 612b-c.



328 ROBERT HEINAMAN

This is the first passage in which there is a referenceto the conse-quencesof justice,andtheyare identifiedwith the rewards ollowingfrom

thereputationorjustice.The rewards ollowingfrom a reputationorjus-tice are distinguishedromany intrinsicvalue of justice, the many beingsaid to regard ustice as evil in itself, and to think that it is rational o

want it only because of its consequences,viz. the rewards ollowingfrom

the reputation orjustice.

2. 361b-362c:In comparing he goodness of the lives of thejust and the

unjust, n order o ensurethat the just man is one who choosesjustice foritself rather hanits consequences, he just man is deprivedof the repu-tation for justice, and, hence, is deprivedof the rewards ollowingfrom

the reputation orjustice.If therewere any otherconsequencesof justicebesidesthose followingfrom thereputationorjustice whichPlato wantedto rule out in assessing its intrinsicworth,thensimply excluding he con-sequences hat follow from thereputation orjustice would not insure hat

the consequencesof justice were not being considered.With those otherconsequencesstill in the field, it wouldremainpossible that thejust manchooses

justice because of those consequences of justice that do notdependon the reputation orjustice insteadof choosingjustice for itself.But as Plato presentsthe case, excludingthe rewards or the reputationforjustice by itself suffices to exclude all of the consequencesof justicewhich he wants to put out of consideration.Hence, there are no conse-quencesof justice which Plato wishes to exclude from considerationtherthan those that follow from the reputationorjustice.

3. 362e-363a:WhenAdeimantus akes over Glaucon'sargument,n order

to provethat the manydo not valuejustice itself he pointsout:

When fathers speak to their sons, they say one must be just... but they do not

praise justice itself, only the high reputations t leads to, in order that the son,thoughtto be just, shall enjoy those public offices, marriages,and the rest which

Glaucon mentioned,as they belong to the just man because of his high repute.

Here, the only consequencesof justice said to be valued which are dis-

tinguished romjustice itself are good reputation nd the goods resultingfrom thatreputation.

4. 367b: Adeimantus xplainswhat he wantsSocratesto do:

So do not merely give us a theoreticalproof thatjustice is better than injustice,but tell us how each, in and by itself, affects a man, the one for good, the otherfor evil. Follow Glaucon's advice and do not take reputations nto account, for




if you do not deprivethem of truereputationand attach false reputations o them,we shall say that you are not praisingjustice but the reputation or it, or blam-

ing injusticebut the appearanceof it...

Adeimantuswantsjusticeitself to be praisedandinjustice tself to be con-

demned.But suppose that Plato thoughtthat the good consequencesof

justiceincludedotherthingsbesidestherewards hatfollow from the rep-

utation or justice.Then, in praising ustice, ignoring he rewards hatfol-

low from the reputationfor justice would not suffice to insure that

Socrateswas notpraisingusticefor its consequences ather hanfor itself.

For the possibilitywould remainthat after removingthe beneficialcon-sequencesof justicethat follow fromthe reputationorjustice,othercon-

sequenceswouldremainand thatSocrateswas praising usticebecauseof

those consequencesrather hanfor itself.

I could go on, but the above passagesprovethe point.In the case of

justice and injustice,the consequenceswhich Plato is concernedto rule

out of considerationn decidingwhether ustice and injusticeare good or

evil in themselvesare those consequenceswhich follow fromthe reputa-

tion for each. Hence,causal consequencesof justiceor injusticewhichdo

not involvethereputationor justiceor injusticemaybe relevant n decid-ing whether hey are good or evil in themselves.

This fits the evidence set out above on the causal consequencesmen-

tionedby Glauconand Adeimantuswhenthey present heirargumentsor

injusticebeing good in itself: the causal consequenceswhich they men-

tion are independent f the reputationor injustice.Likewise,the causal

consequencesof injusticewhich Socratesmentions n Book VIII in argu-

ing that injustice is evil in itself are independentof the reputation or

injustice.

This interpretationmakes it clearwhy, when Plato ends the argumentfor justicebeing good in itself and pointsout why it is good for its con-

sequences,he speaks in terms of the contrastbetweenthe goods which

follow from being just and the goods which follow from appearing

(6oxeiv) to be just (612d). One has the reputation60'a) for justice pre-

cisely when one appears o be just to otherpeople.The benefits that fol-

low from the realityof justice do not dependon such an appearance nd

makeit good for itself.

Thereis one seriousobjection o the proposal hat the consequencesof

justice which Plato wishes to rule out of consideration n assessing itsintrinsicworthare hosethat ollow from hereputationor ustice.36 onsider

36 Cf. Kirwan, "Glaucon'sChallenge," 165; Irwin, PE, 182.



330 ROBERTHEINAMAN

Plato's exampleswhen he firstsetsout the division of goodsat 357b-358a.Examplesof the second class of goods which are good for theirconse-

quencesaswell asfor themselvesareknowledge, ightandhealth.Examples

of the thirdclass of goods which are good for theirconsequences nclude

being treatedwhen ill. Now, whatever he good consequencesof knowl-

edge, sight, healthand being treatedwhen ill which Plato has in mind,

surely manyof them mustbe consequences hat do not dependon having

the reputationor these things.

So, the objection goes, when Plato sets out his division of goods, he

gives as examples of things that are good for their consequences hings

thatare instrumental oods for consequenceswhich do not dependon the

reputationor those things. If, in his own examples,the consequencesof

x thatmatter or determiningts instrumental alue do not dependon the

reputationorx, how can Platowish to say that the consequencesof jus-

tice thatmatter or determiningts instrumental alue dependon the rep-

utationforjustice?

To this there are two replies. First, the evidence cited above still

remains.In particular, mmediatelyafter setting out the general division

of goods (358a), Plato specifiesthe consequencesof justice thatneed to

be excludedfrom the assessmentof its intrinsicvalue as those rewards

thatfollow fromthe reputationorjustice.

Secondly,considerwhat Plato says when he returns o his division of

goods in a laterpassage (367c-d):

Now since you agreed thatjustice is among the greatest goods - those that are

[all worth having for what comes from them but much more [b1] [worthhav-

ing] for themselves, such as seeing, hearing, thinking, and, of course, being

healthy and all the other goods that are [b2] fruitful by their own nature(yo'vtta

mlairWv {aEt) but [a2] not [fruitful]because of reputation (AkX'oi5 0'p) -

praise this aspect of justice, [b3] the way in which it in itself benefits the manwho has it, and the way in which injustice harms the man who has it; leave [a3]

the rewardsand reputations 864ag) for others to praise.

This passage may be open to different readings. It might, perhaps, be

understood so that [b2] but not [a2] is explanatory of [bl] in this sense:

specification of the intrinsic worth of seeing, hearing, etc. specifies the

goods they produce by their own nature - not the goods they produce

through the reputation [50ta] for possession of the goods, which would

make them good for their consequences. Then, throughout, [b] specifies

intrinsic value while [a] concerns the value of a thing for its consequences.

Suppose this is right. The examples of health and sight mentioned here

werealso mentionedn the originaldivisionof goods at 357c as instances




of the second class of goods- things that are good for theirconsequencesas well as for themselves.And [all, [a2] and [a3] connect the beneficial

consequencesof healthand sightwith the reputation or possessingthem.So understood, his passageanswersthe claim that Plato'soriginalexam-

ples at 357c show thatthe good consequencesof justice in the Republic

are not those which follow from the reputation or justice. For the argu-mentwas that in the examples mentioned e.g. health and sight - their

consequencesevidently do not dependon the reputationor the thingsin

question.But here Plato connectsthe consequencesof even these things

to reputation.Suppose nsteadthatwith [a2] Platomeans thathealth, sight,etc. have

no consequences hatfollow from theirreputation.Thatis, on this inter-

pretation,both [b2] and [a2] elucidate[bl]. Nevertheless,when the case

of justice is returned o in [b3] and [a3], the consequencesof justice are

once againtied to the reputationor justice. So, on this interpretationf

[a2], in the same sentence n whichPlatomentionssome of the examplesfrom357c andaffirms hat their nstrumental aluehas nothing o dowith

consequencesresulting romthe reputationor having them,37 e repeats

the idea of the dependenceof the consequencesof justice on possessingthe reputationor justice.Therefore,on this interpretationf 367c-d, the

fact that 357c mentionsexamplesof goods whose instrumentalalue has

nothingto do with reputation annotshow that the instrumental f valueof justice does not dependon the reputation or justice.38

I conclude that Platoregards he causal consequencesof x as possibly

relevantto assessing the intrinsicvalue of x. But it would be wrong tothinkthatPlato believes thatin all cases it is onlythecausalconsequencesof x that matter n determiningts intrinsicvalue. Considerthe view of

the manywho hold thatjustice is an intrinsicevil because it is difficult(xaXE7,O')ndpainfulorlaboriousEsuc6vo;)358a,364a;cf. 364c-d,365b),

37 Although I said that the present interpretation onstrues both

[b2] fruitful (y6vtga) by their own nature(r~ aiov (p'aEt), and

[a2] not [fruitful]because of reputation o' 8061)as elucidatingthe idea that these thingsare fall worth having for themselves, the inter-

pretationconstrues [a2l as meaning that seeing, hearing, etc. have no consequences

because of their reputation.If so, it follows that the good consequences that make

them good for their consequences include nothingthat follows from the reputation or

having them.38 361b5-362e7 is another passage that contrasts the reality and appearanceof jus-

tice (and injustice) (361b7-8, dl, 362e6-7), and connects the appearanceof justice with

the consequences of justice.



332 ROBERTHEINAMAN

and so put it into the third class of goods. It is possible that F1ct1o6vo;should be translated"painful," nd it is possible that Plato regardspain

andpleasureas a causalconsequencesof just action.Still, there s no rea-

son to thinkthat Plato regardspain as an intrinsicevil only because of

its causal consequences.And, in any case, the fact that a kind of behav-

ior such as doing what is just is "difficult" as nothing o do with its causal

consequencesbut is rathera featureof the action.However,the fact that

justice is difficult s offeredas a reasonwhy it is evil in itself. Hence,the

fact that x is evil (or good) in itself need not be determined olely by its

causal consequences.39 Being difficult is not even a "formal" consequenceof just behavior.

If the fact that justice is difficult supports he point that it is evil in

itself, then the fact that injustice s "easy"(364a; cf. 364c) must support

the claim thatit is good in itself. Being easy is not a causalconsequence

of unjustaction.

III

I will concludeby pointingout some other featuresof Plato's division ofgoods that are worth noting. First, the consequencesof x do not neces-

sarily follow fromx.40 n orderto set up the examinationof the intrinsic

value of justice and injustice,Glaucon presents he cases of the just manwho does notenjoy the consequencesof justice butinsteadsuffers hecon-

sequencesof injustice;and of theunjustmanwho does not suffer he con-

Which is not to deny that even in the case of pain and pleasure their conse-

quences may be relevant to determiningtheir intrinsic worth. At 357b Plato says that"harmless pleasures (i'ova')" are good in themselves. If these pleasures are sensa-

tions of pleasure, Plato may be implying that when they have harmful consequences

they are not good in themselves because of those harmfulconsequences. It is also pos-

sible, however, that Platowould classify a harmfulpleasureas something good in itself

which has evil consequences. This may be the view found at Protagoras 353c-354c.

At 351c-e, pleasure qua pleasure is good. But immediately after 354c we get a dif-

ferent use of "good"and "evil." Now (354c-e), the overall goodness or evil of x is

assessed by comparing ts intrinsic pleasantness or painfulness (goodness or evil) with

its pleasant or painful (good or evil) consequences. Socrates' subsequent attempt

(354e-355e) to refute the many's explanationof akrasiaappearsto presenta difficulty

only because the differencebetween these uses of "good"and "evil" is blurred.40 Noted by Cross and Woozley, Plato's Republic,66. Reeve claims (Philosopher-

Kings, 30) that if x is an instrumentalgood because it brings about y, x is sufficient

for y. See also n. 27.




sequencesof injusticebut enjoys the consequencesof justice.While Platolater points out that this separation f justice and injustice romtheir con-

sequenceswas made solely for the sake of argument 612c-d),and that it

is impossible hatboth gods and men shouldbe ignorant f a man's justice

or injustice, here is plainly no reason why someone's justice or injustice

could not escape the knowledge of other people. Then the consequences

of justice or injusticethat follow from the knowledge of other people

would not be enjoyed by the just or unjustman. And obviously,as Plato

clearly knew,even if A is just and has the reputationorjustice, a tyrant

might preventall those goods which Plato says are the consequencesofjustice.A tyrantmay escape at any rate some of the consequencesof his

injustice,or if A is unjustbut not a tyrant,A may be clever enough to

conceal his injustice,or find himself in a positionwhere he can hide his

injustice. Similarlywith Plato's example of treatment or illness - the

hoped for consequenceof restoration o health does not always follow.

Consideralso the historicalSocrateswho did not enjoywhat Platowould

regardas all the consequencesof justiceeven though, n Plato's view, he

was the mostjust man of his time (Phaedo 118a,SeventhLetter324e).

So: at least not all of the consequencesof x that make it an instru-mentalgood necessarily ollow from x. It need not be a sufficientcondi-

tion for thoseconsequences,and thisevidentlyappliesto theconsequences

of justice and injustice.Plato is drawinga generaldistinctionbetween

types of good, and the claim that y is a consequenceof x entails at best

thaty usuallyfollows x.

The same applies to those causal consequencesthat make x good or

evil in itself. For example,while the fact thatunjustaction leads to the

accumulation f wealth is given as a reasonwhy injustice s good in itself

in the speechesof Glaucon and Adeimantus, here is no reason to thinkthat Plato believes that this necessarily follows, even when the unjustaction is undertakeno obtainmoney.Adeimantus ays (364a)that"unjustdeeds arefor the mostpart moreprofitablehanjust ones."

Nor is x on its own a necessary conditionfor the consequencesof x.

The consequencesof x may follow from somethingotherthanx: in Plato's

case of the perfectlyunjustman, the consequencesof justicefollow from

the appearanceof justice even when justice is not present n the unjustman. So x is not a necessarycondition or theconsequencesof x that make

x good for the sake of its consequences.Another feature of Plato's division which has attracted ittle if any

notice is this: The notions of intrinsicgood and evil are relativeto some-

one affectedby the item in question.While any good or evil is good or



334 ROBERT HEINAMAN

evil for some person, the point here is that the same thing may be anintrinsicgood for A and an intrinsicevil for B.

GlauconandAdeimantus,while arguing hatjustice is an intrinsicevil

for the agent, defendThrasymachus' iew thatjustice is another'sgood

(367c; cf. 343c), i.e. it is a good for the "patient" f the action.Suppose

that Glaucon and Adeimantus,n saying that A's just action is the good

of someone otherthanA, meantmerelythat A's just action has beneficial

consequences in Plato'ssense)for thepatientof the action.Then theview

of Glaucon and Adeimantuswould be that, normally,the just agent's

actionproducesexactly the same kinds of consequences namely, goodones - for both the agentand the patient.For,as noted above, all partiesto theargument gree that,normally, heagent's ust actionproducesgood

consequencesfor the agent. But it is clear that the view thatjustice isanother'sgood means that in the way in whichjustice is another'sgood

it is preciselynot the agent'sgood. The only way in which the view set

out by GlauconandAdeimantusholds that ustice is not the agent's goodis that it is not good in itself for the agent. So when that view assertsthat

justice is the patient'sgood it means thatjustice is good in itself for the

patient.Therefore,he view of Glauconand Adeimantusmust be that ust action

is an intrinsic evil for the agent and an intrinsicgood for the patient.Hence, the same thing may be an intrinsicgood for A and an intrinsicevil for B.

Since the disagreementbetween the many and Socrates concernsthevalue of just action for the agent, the many put it into the thirdclass of

goods becausethey believe thatit is an intrinsicevil for the agentof the

action even though t is also an intrinsicgood for thepatientof the action.

But sincejusticeis an intrinsicgood fromthepointof view of the patient,it cannotbelong to the thirdclass of goods and so must,from thatpointof view, belong to the first or second class of goods. Hence, the same

thingmay fall into differentcategoriesof good.Of course,Plato will go on to arguethat usticeis an intrinsicgood for

the agent.Since all presumably greethatjustice is an intrinsicgood for

the patientof the action,the samethingcan be an intrinsicgood for morethan one person,here the agentand the patientof the action.If, as Plato

holds, what is good for a personis based on what is in fact in that per-

son's self-interest,and if the self-interestof A is not identical with the

self-interestof B, then it must be the case that the same thingcan be anintrinsicgood for differentreasons.In the case of just action,it is clearhow this will be so. The claim thatjustice is the intrinsicgood of the




patientwill be based on the fact that the patientreceives some good fromthe action, money, honor or whatever.The claim that just action is an

intrinsicgood for the agentwill based on the fact that it has certaincon-

sequencesfor the soul of the agent,viz. it promotes ustice in the agent's

soul (588b-59le with 612a-b). So different ypes of causal consequence

are relevantto determiningwhetherthe action is an intrinsicgood.4'

UniversityCollege London

4' I thank the Arts and HumanitiesResearch Board for supportingmy researchfor

this paper. I also thank ChristopherRowe for his comments on an earlier version of

the paper.

plato's division of goods in the republic

Documents