You are on page 1of 18

The Virtues of Thrasymachus

Author(s): T. D. J. Chappell
Source: Phronesis, Vol. 38, No. 1 (1993), pp. 1-17
Published by: BRILL
Stable URL:
Accessed: 21-05-2015 02:09 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
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

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

This content downloaded from on Thu, 21 May 2015 02:09:00 UTC
All use subject to JSTOR Terms and Conditions

The Virtues of Thrasymachus

'We should at least consider the possibility that justice is not a virtue. This
suggestion was taken seriously by Socrates in The Republic, where it was
assumed by everyone that if Thrasymachus could establish his premise that injustice was more profitable than justice - his conclusion would
follow: that a man who had the strength to get away with injustice had
reason to follow this as the best way of life. It is a striking fact about modern
moral philosophy that no one sees any difficultly in accepting Thrasymachus' premise and rejecting his conclusion, and it is because Nietzsche's
position is at this point much closer to that of Plato that he is remote from
academic moralists of the present day.'
(Philippa Foot: 'Moral Beliefs', Proceedings of the Aristotelian Society,
59 (1958-1959), 99-100)
Thrasymachus' statement of an alternative to standard views about justice
in Republic Bk.I sets the challenge which Republic Bks. II-X must answer.
If this is not a serious challenge, if Thrasymachus' alternative view of justice
is not interesting, plausible or coherent, it is not clear why moral philosophers should bother with The Republic at all. Here I will offer an interpretation of Thrasymachus' alternative view of justice which does make
his view out to be interesting, and plausible, and coherent. My interpretation differs in one way or other from some very well known interpretations;
I hope it will become clear what, if anything, my interpretation achieves
that these others do not.

Consider the conflicts between these seven understandings of Thrasymachus:
Phronesis 1993. Vol. XXXVIII/J (AcceptedSeptember1992)


This content downloaded from on Thu, 21 May 2015 02:09:00 UTC
All use subject to JSTOR Terms and Conditions

1. Thrasymachus makes no clear point; on Plato's depiction he is merely
confused. (So Macguire, in Phronesis XVI (1971) 142-163.)
2. Thrasymachus is a revolutionary who wants to turn society upside down:
he rejects 'Conventional Justice' in favour of 'Natural Justice'.
(The entry on Thrasymachus in Pauly-Wissowa's Encyclopadie embraces (2): 'Die These rep.338c, daB das btXaLOV von Natur nichts anderes
sei als der Nutzen und Vorteil des Starkeren . . . entspricht der Destruktion des Rechtsgefuihles und der ethischen Normen'.
3. Thrasymachus is a Thucydidean cynic.
(Alasdair Maclntyre, in Whose Justice? Which Rationality? (London:
Duckworth, 1985), ascribes to Thrasymachus the 'Thucydidean' thesis
that 'Arete is one thing, practical intelligence quite another'.)
4. Thrasymachus' position is the same as Callicles' in the Gorgias.
(Shorey, the Loeb translator of The Republic, Vol. I, p.64: 'The actual
ruler or shepherd of the people . .. tends the flock only so that he may
shear it. All political experience and the career of successful tyrants. . ..
[Thrasymachus] thinks, confirms (sic) this view, which is that of Callicles
in the Gorgias'.)
5. Thrasymachus is a Nietzschean immoralist.
(So Shorey again, in his Loeb translation of The Republic Vol. I, p.x:
'Thrasymachus . .. affirms the immoralist thesis that justice is only the
advantage of the . .. stronger' (p.x) - a thesis which Shorey goes on to
call, not only 'Nietzschean', but 'sophistic', 'Machiavellian', and 'Hobbesian'.)
6. Thrasymachus believes that justice means obedience to the laws.
(So G.F. Hourani, Phronesis VII (1962), 110-120.)
7. Thrasymachus means to recommend injustice as a way of life.
(So, famously, G.B. Kerferd (Durham UniversityJournal, IX (1947-8),
19-27; Phronesis IX (1964-5), 12-16: 'Thrasymachus . .. makes it clear
that his own ideal is for everyone to seek his own interest, and he regards
justice as always involving the contrary, namely seeking another's interest, and injustice as always involving seeking one's own interest'. So
also Philippa Foot, to judge by her suggestion, above, that Thrasymachus' thesis is that 'a man who [has] the strength to get away with
injustice [has] reason to follow this as the best way of life'.)
It will already be clear that, of these seven views, I disagree most strongly
with (1). Yet even (1) seems, prima facie, quite plausible. After all, Plato
does make Thrasymachus say all of the following':
' All translationsfrom Plato in this paper are my own.


This content downloaded from on Thu, 21 May 2015 02:09:00 UTC
All use subject to JSTOR Terms and Conditions

A. qud yCE ?yO EvaM t6o 8xaLov oiUx akko Tt - T toIO XQErTiOVo;
upEQov - 'I say that justice is nothing other than the interest of the
stronger.' (338c)
B. TLOETat&e yE toiig v6'ovU ?xaoTNTIrXl rtleo; t6o CMtl tpovQOV
- 'Each kind of government makes laws in its own interest, and, by so
enacting, proclaims to its subjects that this is justice. . . (338e)
C. TOf)T' OVUV ?xrLuV, J

TCLTov EivwL &,xalov,


EV CaTtaOCttg TCtu Jto06Xr

Tro Tl'r XaOEOrviXULCg&Qx#;

'This then, my good sir, is what I say is one and the same justice in all
states: the interest of the government in power.' (338e)
D. i1 jiEv 6xaton3v1
xTo t6 &'xaLov CXXOTQLovayaOOv T4) 6VTL 'Justice and the just are in reality the other person's good.' (343c)
A-D all look incompatible. A will not square with B or C: it is not obviously
a necessary truth that 'the stronger' is always going to be the ruler, or (a
member of) the government in power. (Unless we define 'stronger' to mean
'more politically powerful'; but who says we have to do that?) Socrates
points out at an early stage that B and C are not consistent with each other,
either: those in power might be wrong about what was in their interest, and
so make laws intended to be in their own interests, but not actually in their
own interests (339e). As for D, this undermines all the other theses. If
justice is, with D, the other's good, then how can the stronger, or a
government, or the rulers, be just in pursuing their own interest - as A, B
and C require?
So this evidence could easily be taken to support Macguire's claim (1)
that Thrasymachus is simply lost in these perplexities. But in fact these
perplexities can be solved. This is the question to ask: Is Thrasymachus'
thesis a descriptive thesis about justice, or a prescriptive thesis?
For my purposes here, this distinction between the 'prescriptive' and the
'descriptive' might as well be exhaustive. It is not, of course: there are
plenty of other interesting things to do with words besides prescription and
description (cp. Wittgenstein, Philosophical Investigations 1,23). But I have
been unable to think how any of those other activities could be relevant to
the elucidation of Thrasymachus' thesis (though naturally I welcome suggestions). Therefore I will proceed upon the basis that, if Thrasymachus'
thesis about justice says anything worth hearing, and does not fit the one of
these two alternatives, then it must fit the other.
This is how I shall be using the distinction. If Thrasymachus' project
regarding the social practice called LxaLoru'v1 is simply to observe and

This content downloaded from on Thu, 21 May 2015 02:09:00 UTC
All use subject to JSTOR Terms and Conditions

describe it, I will say that he has a descriptive thesis about justice. Whereas,
if Thrasymachus' project regarding justice is to make reference to justice in
order to offer us reasons to behave or live in a certain way, I will say that
Thrasymachus has a prescriptive thesis about justice. I shall argue that
Thrasymachus does not have a prescriptive thesis about justice, and does
have a descriptive thesis about justice.
One important complication which will appear in this. Although Thrasymachus does not hold a prescriptive thesis about justice, he does hold a
prescriptive thesis about a character trait which is, very often though not
always, coextensive with the character trait of justice. This fact makes it
look at times almost as if Thrasymachus does hold a prescriptive thesis
about justice itself, rather than about the other character trait with which it
very often coincides. But I shall argue that this appearance is deceptive. if
Thrasymachus ever prescribes justice or just behaviour, he does so, as
and not essentially.
Aristotle would say, only xaxa oaufP3E0Pxo6;,
If Thrasymachus did hold a prescriptive thesis about justice, what form
could it take? Note first that neither Thrasymachus nor Socrates thinks, as
we are often inclined to think, that 'a just person' means pretty much the
same as 'a good person' and that the word 'justice' is, or can be, simply a
(rough) synonym for the phrase 'moral rightness'. Many modern moral
philosophers, for example Professor Hare, have tended to talk as if 'It
would be just to do F' were often or even usually a straightforward equivalent to 'It would be morally right to do F'. But however this may be for us,
it cannot have been so for Socrates and Thrasymachus, for the reason noted
by Mrs. Foot in my epigraph. The modern moral philosophers suppose that
someone who is told that it would be 'morally right' to do something cannot
go on to question this without making a mistake. It is, they say, merely
incoherent to ask 'But what reason do I have to do what is morally right?'.
If, tacitly or openly,they accept the rough-synonym view of the meaning of
'justice', they must presumably think that 'What reason do I have to do
what is just?' is similarly incoherent. But Thrasymachus and Socrates
apparently do not think it merely incoherent to ask 'But what reason do I
have to do what is just?'. However else Socrates may try to meet Thrasymachus' attack on justice, it is not by accusing him of this kind of incoherence. It follows, as Mrs. Foot notes, that Thrasymachus and Socrates must
mean something rather different from these modern theorists when they
talk of 'justice'. Their concept is more specific, less grandiosely all-embrac4

This content downloaded from on Thu, 21 May 2015 02:09:00 UTC
All use subject to JSTOR Terms and Conditions

ing, 'thicker' as Bernard Williams would say2. In particular, 'justice' for
them is not only a concept which might be used in the justification of
particular deeds or ways of living. It is a concept which itself stands in need
of justification, by reference to the more basic notion of that flourishing
human life to which justice or injustice may or may not be seen as contributing, and to question which is (perhaps) what they would find puzzling in
the way that modern theorists find it puzzling to question why we should do
what is 'morally right'.
It seems, then, that there are two prescriptive theses about justice, either
or both of which Thrasymachus might be offering us. (I) He may be
answering the question 'How should we behave?' by replying 'We should
behave in such and such a way, because it's just to behave like that'. Or (II)
He may be answering the more basic question 'But why should we be just?'
by replying 'Because it's a human excellence to be just'. Form I of the
prescriptive thesis will say that, in general, 'x is just' provides creatures like
us with a reason of some sort for doing x. Form II of the prescriptive thesis
will say that, in general, justice is a good thing to have in your character, a
character trait which is important or even essential to a flourishing human
life: in other words, a virtue.
Although I and II are logically distinct theses, there is no problem about
supposing Thrasymachus to be asserting both. Rather, as we shall see, there
is a problem about supposing him not to be asserting both theses of such a
pair. For, evidently, I prompts the question 'Why does "x is just" provide
creatures like us with a reason of some sort for doing x?' and, plausibly, II
answers this question. I is a thesis about one level of practical reasoning:
namely, our motivation for doing certain sorts of actions rather than others.
II is a thesis about another level of practical reasoning: namely, our motivation for pursuing certain sorts of developments in our characters rather than
others. Within the framework of an ethics of virtue, I and II are complementary theses.
I now argue that Thrasymachus does not argue either for I or for II, nor
(III, IV) for these theses in 'inverted' forms.
Suppose that Thrasymachus is arguing for I. Then he is recommending
certain forms of behaviour by reference to their being just. Why should we
do what is in the interest of the stronger, or obey the laws, or promote the
interest of the prevailing government, or promote the good of others?
Given I, the Thrasymachean answer will be: 'Because it's just to do so'.
2 V. Bernard Williams, Ethics and the Limits of Philosophy(London: Collins, 1985),
Chapter 7.


This content downloaded from on Thu, 21 May 2015 02:09:00 UTC
All use subject to JSTOR Terms and Conditions

The obvious problemfacingthis is simplythat Thrasymachus'remarks
about justice, if construedin this way as recommendationsabout how to
act, contradicteach other. This can be shown by comparingonly the first
and the last of Thrasymachus'four remarks,A and D3.
Suppose,with I, that 'x is just'givesus a reasonto do x, and, withA, that
justiceis whateveris in the interestof the stronger.Then it followsthat'x is
in the interestof the stronger'gives us a reasonto do x. But now suppose,
with D, that justice is also 'another'sgood'. By parityof reasoning,'y is
another'sgood' will also give us a reason to do y. But what if I am the
strongerin some situation,andmygoodconflictswithsomeotherperson's?
In that case, it maybe true boththat 'x is another'sgood', and that'not-xis
in the interestof the stronger'.In whichcase, apparently,I have reasonto
do both x and not-x; whichis absurd.
So Thrasymachusis not arguingforformI of the prescriptivethesis:that,
in general, 'xis just'providescreatureslike us witha reasonof some sortfor
doing x.
What about form II of the prescriptivethesis: that justice is a virtue, a
desirabletraitto have in yourcharacter?Thrasymachusis not arguingthis
either, for at 348c he explicitlydenies that justice is a virtue:
SOCRATES: Come now, what will you say about justice and injustice?
One of them, I suppose, is a virtueon yourview, and the other a vice?
THRASYMACHUS: Yes - why not?

SOCRATES:Justice, I take it, is the virtue, and injusticethe vice?
THRASYMACHUS: Is that likely, you sweet fool, when I say that injustice is profitable (XvoTEkEXv)but justice isn't?
SOCRATES:Then what do you mean?
THRASYMACHUS:The opposite.
SOCRATES:What, that justice is a vice?!
THRASYMACHUS:Well, no - but it is naive simplicity(;naviU

In my technique for refuting the idea that Thrasymachus'doctrine about justice is
primarilya prescriptiveone, and in my emphasis upon definition D of justice, I am, of
course, influenced by Kerferd'sclassictreatment(Durham UniversityJournal 1947-8,as
cited). However, (i) I spell the argumentout differentlyto Kerferd,who makes no use of
any distinction between levels of practicalreasoning;and (ii) my conclusion is different
from his. Unlike him I do not use this technique to argue that Thrasymachusthinks
injustice is a virtue, but (on the contrary)to argue that Thrasymachusthinksthat neither
injustice nor justice is either a virtue or a vice.


This content downloaded from on Thu, 21 May 2015 02:09:00 UTC
All use subject to JSTOR Terms and Conditions

SOCRATES: You mean then that injustice is duplicity (xaxo'OElav)?
THRASYMACHUS: No: injustice is practical intelligence (EviouXktcav).
This passage makes it quite clear that Thrasymachus rejects form II of the
prescriptive thesis just as emphatically as he rejects form I. Not only does he
deny that 'x is just' gives us a reason to do x. He also denies that human
beings, in general, have any reason to think that justice is a virtue, a
necessary part of the good life. Therefore, when Thrasymachus says (e.g.)
that 'Justice is the interest of the stronger', there are two meanings, corresponding to the two forms of the prescriptive thesis, which we certainly
cannot attach to these words. (I) We cannot construe him as meaning 'Do
what is in the interest of the stronger, because doing that is just (and a
flourishing human life necessarily involves justice)'. And (II) we cannot
construe him as meaning 'Do what is just, because doing that is doing what
is in the interest of the stronger (and a flourishing human life necessarily
involves doing what is in the interest of the stronger)'. For 'justice' - it has
turned out - is not a term of either level of Thrasymachean practical
reasoning. No Thrasymachean action is motivated by the thought that
'Doing this is doing a just action'; and the Thrasymachean agent has no
reason to try to develop in himself the character trait of justice, for that trait
has - on Thrasymachus' view - nothing to contribute to human flourishing.
This still leaves two other ways of arguing for a prescriptive reading of
Thrasymachus: namely the inverse forms of theses I and II. Why not say
that Thrasymachus means to argue (III) that 'x is unjust' states a reason for
us to do x? Or (IV) that injustice is a virtue?
A simple argument shows that, for the same reason for which he could
not hold I, Thrasymachus cannot consistently hold III either, 'X is just' (we
saw) could not state a coherent Thrasymacean reason for us to do x,
because Thrasymachus defined justice both as 'the interest of the stronger'
and as 'another's good'. So if I was the stronger, and doing x was just
because it was in my interest, but doing not-x was just because it was
another's good, then it would follow that I was committed to doing both x
and not-x. Pari passu: 'x is unjust' can't state a coherent Thrasymachean
reason for us to do x. For if I am the stronger, and doing not-x is unjust
because it is contrary to my interest, but doing x is unjust because it is
contrary to another's good, it follows that I am committed to doing both
not-x and x.
IV, the claim that Thrasymachus holds that injustice is a virtue, is (in
Socrates' phrase, 348e), fj&q(oT;QEc?)EQov,a rather tougher proposition. It
certainly admits of at least some defence from the text:

This content downloaded from on Thu, 21 May 2015 02:09:00 UTC
All use subject to JSTOR Terms and Conditions

SOCRATES: Thrasymachus, do the unjust seem to you to be intelligent
and good (aEyaOo0)?
THRASYMACHUS: Yes - those who are able to be completely unjust ...
348e 1-4:
SOCRATES: ... But this is what made me wonder: that you put injustice
in the class of virtue and wisdom, but justice in the opposite classes.
THRASYMACHUS: But that is just what I do.
SOCRATES: Now it is clear that you are going to say that [what is unjust] is
good (xakcov)and strong, and that you will render to it all the additions that
we rendered to what is just - since you have dared to put it [sc. what is
unjust] in [the class of] virtue and wisdom (rEL&bi yFexai Ev &'eETi aiT'o
xait oopqt eloPX&oAagOrEvat).
THRASYMACHUS: What a good prophet you are.
The first and simplest way to defend IV would be to say this: that if
Thrasymachus denies that justice is a virtue, then he must for that very
reason hold that justice is a vice and injustice a virtue. But at 348c, where
Thrasymachus certainly asserts that justice is not a virtue, he also denies
that justice is a vice:
SOCRATES: Then what do you mean?
THRASYMACHUS: The opposite.
SOCRATES: What, that justice is a vice?!
THRASYMACHUS: Well, no. . (oi)x...
So Thrasymachus does not think that justice is a vice, and injustice is a
virtue, simply because justice is not a virtue. But might he not hold that
justice is a vice and injustice a virtue on other grounds? A second, and more
subtle, line of argument for IV might point to what comes next at 348c:
THRASYMACHUS: Well, no - but it is naive simplicity (3navicyEvvatcav
SOCRATES. You mean then that injustice is duplicity (xaXoiOELav)?
THRASYMACHUS. No: injustice is practical intelligence (EvBouXkLav).
Here, it can be argued, Thrasymachus identifies justice and naive simplicity, and injustice and practical intelligence. Now he clearly holds that naive
simplicity is a vice, and that practical intelligence, as opposed to duplicity, is
a virtue. In which case Thrasymachus must, surely, hold that justice is a vice

This content downloaded from on Thu, 21 May 2015 02:09:00 UTC
All use subject to JSTOR Terms and Conditions

(namely, naive simplicity), and that injustice is a virtue (namely, practical
Thrasymachus nowhere actually affirms the conclusion which this second
argument attributes to him, that justice is a vice and injustice a virtue.
However, he does say that 'the unjust are intelligent (qwevLlioL)and good
(&yctOo')' (348dl-2); and he does say that he 'puts injustice in the class of
virtue and wisdom, but justice in the opposite classes' (348el-4; 348e7349a3). So can we not infer, without further ado, to the conclusion that he
thinks that justice is a vice and injustice a virtue?
No: for at least three reasons. First, to say that injustice is 'in the class of
virtue and wisdom' (ev CIwETij xat oopL'ct FQ?t), but that justice is 'in
the opposite classes' (Ev ToL; EvavUtot;), is not equivalent to saying
that injustice simply is virtue (or wisdom). If Plato wants Thrasymachus to
say that 'injustice is virtue', why does Thrasymachus not say precisely that,
rather than what he actually says: that 'injustice is in the class of virtue and
wisdom'? (And notice how he sticks to this double formulation. The unjust
are not good, they are 'intelligent and good'; injustice is not virtue or a
virtue, it is 'in the class of virtue and wisdom'.)
Conceding that Thrasymachus means it when he observes that injustice is
'in the class of virtue and wisdom' does not prevent us from denying that
injustice is itself a Thrasymachean virtue. For that observation might mean
rather that for Thrasymachus injustice is characteristic of the virtuous or
excellent man, without itself being one of the cardinal character traits that
make him a virtuous man. It might mean that his injustice is a secondary
feature of his character, an incidental consequence of his having some other
characteristic which is cardinal, essential to his excellence. (What other
characteristic? If we take 'virtue and wisdom' as a hendiadys, then we can
make room for the suggestion that the characteristic in question is 'the
virtue of wisdom'; where 'wisdom' is taken of course in a suitably Thrasymachean sense - i.e., as equivalent to 'practical intelligence'.)
The second reason is that, if Thrasymachus does hold IV, that justice is a
vice and injustice is a virtue, then it will be very odd if he does not also hold
III, that 'x is unjust' states a reason for doing x. If he holds IV without
holding III, then he supposes that there is a virtue of which he cannot say
how practical reasoning articulates it in action; but how can anything which
has no specifiable way of affecting action be a virtue? Yet we have already
seen why he cannot hold II1, any more than he can hold I: because given
Thrasymachus' own explicit remarks about justice, it yields contradictory
practical imperatives.


This content downloaded from on Thu, 21 May 2015 02:09:00 UTC
All use subject to JSTOR Terms and Conditions

The thirdpointis thatthissecondargumentdependscruciallyuponthere
being a strictidentitybetween injusticeand practicalintelligence.For it is
arguedthat (a) injusticeis (the same as) practicalintelligence;(b) practical
intelligenceis a Thrasymacheanvirtue;and therefore(c) injusticeis (the
same as) a Thrasymacheanvirtue.But Thrasymachusrejectsthisargument
becausehe rejects(a). For (a) to be trueis for thisto be true:anybehaviour
is practicallyintelligentbehaviourif andonly if it is unjustbehaviour:(Vx)
(Px <-> Ux). But this biconditionalis doublyfalsified,for Thrasymachus
clearlyholdsthattherecanbe bothunjustbehaviourwhichis not practically
intelligent behaviour, and practicallyintelligent behaviourwhich is not
unjust behaviour. This is a consequenceof Thrasymachus'remarks,for
exampleat 344c, aboutthe need for full or perfectinjusticeto be accompanied by power:
'Thus, Socrates, injustice is stronger, and more liberated, and more masterlythan
justice when it comes to the full (Nxavds yLyvot )..

Thrasymachusdoes not hold that any unjust act I do exemplifies my
practicalintelligence:only the unjustact whichis calculatedto matchmy
power. Acting in accordwith Thrasymacheanpracticalintelligence,therefore, cannot simply mean acting unjustly;ratherit must mean acting as
unjustly as you can get away with. It follows from this that practical
intelligenceand injusticeare, on Thrasymachus'view, not only not identical; in many situationsthey are actuallyinverselyproportionateto each
other. Glaucon'sstory of Gyges in Bk.II bringsthis point out well. Compare two men, equallyempowered,who both want to do what Gyges did,
but neitherof whom has Gyges' ring.The one who tries to do what Gyges
did withoutthe ringis the more unjust,but the one who restrainshimself,
knowinghe will never get awaywith it, is the more practicallyintelligent.
cannot be identified:hence injusticeis not shown to be a Thrasymachean
virtue just because practicalintelligenceis. This completes my argument
against prescriptivethesis IV, which was the last prescriptivethesis still
available.So, if it is coherentat all, Thrasymachus'
thesiscannot, in any of
these senses, be a prescriptivethesis.

Consider now the proffered alternative:that Thrasymachus'thesis is a


This content downloaded from on Thu, 21 May 2015 02:09:00 UTC
All use subject to JSTOR Terms and Conditions

In this case, Thrasymachus is not primarily concerned to tell us why, as a
matter of policy, we should choose justice (or for that matter injustice). His
main concern is to tell us what, as a matter of fact, justice is like; to observe
and describe the social practice called bxa1orC'v1, justice.
What is the nature of that social practice, according to Thrasymachus?
For him justice is neither a virtue nor a vice; it is a device. Consider in its
context what I labelled (4):
'You do not realise, Socrates, that justice and the just is in reality another person's
good (that is, it is the strongerand the rulingperson'sinterest) and one's own injury
- if one obeys and complies. But injustice is the opposite, and gives control over
those who are in truth simple and just (Tiv d;g a'kTJ06g EiMi0OX6VTe xai btxaiwv).
But they, since they are being controlled, do what is in the interestof the one who is
stronger.' (343c-d)

Thrasymachus believes that justice is a device in this sense: that it is a trick
played on the naive by the cunning, to make the naive give up any advantages they may have over the cunning. He gives us what are clearly
meant as descriptions of how this works in social practice:
'The just person always come out the worse in his encounters with the unjust
person. For one thing, suppose they co-operate together in some partnership.
When the partnership is dissolved, you will invariablyfind that the just person
comes away, not with more, but with less than the unjust person. Again, take their
relations with the city. Whenever there are taxes of some sort to be paid, the just
person contributes more than the average amount, and the unjust person less; but
whenever there are benefits to be claimed, the just person gains nothing, and the
unjust person gains a lot.' (343d)

For another example of what Thrasymachus means when he describes the
just person as E"Thr'I;,naive or simple, take the deed of the Anglo-Saxon
general Beorhtnoth, fighting the Vikings at the Battle of Maldon in 991
'The Northmen and the Englishwere . . . separatedby an armof the river;filled by
the incoming tide, it could only be crossed by a . . . causeway, difficult to force in
the face of a determined defence . . . But the Vikings knew, or so it would seem,
what mannerof a man they had to deal with: they asked for leave to cross the ford,
so that a fair fight could be joined. Beorhtnoth accepted the challenge and allowed
them to cross. This act of pride and misplaced chivalry proved fatal. Beorhtnoth
was slain and the English routed . . .'

(J.R.R. Tolkien's account: The Homecoming of Beorhtnoth(London: Unwin,
1975, p. 149-150)

Beorhtnoth's failure, in Thrasymachus' eyes, would not be to do with a fatal
flaw in his character of pride and misplaced chivalry', as Tolkien suggests

This content downloaded from on Thu, 21 May 2015 02:09:00 UTC
All use subject to JSTOR Terms and Conditions

(in line with the Maldonpoet's 'ofermode':v. TheBattleof Maldon,lines
85-954).For Thrasymachus,Beorhtnoth'sfatal flaw would be his lack of
?Fi4oukXa,practicalintelligence:his inability, or refusal, to see through
the social practiceof justice. He fails to see that the institutionof justice is
simply not somethingwhich humanshave any need of if they are to live
flourishinglives; on the contrary,to act accordingto justice is, in normal
cases, to choose to contributeto the flourishingof others' lives at the
expense of one's own flourishing. Justice as a social institution is, then, the
embodiment of a trick; a trick which the stronger play on the weaker, which
rulers play on their faithful subjects, and which the Vikings play on Beohrtnoth - with the very natural result, as Thrasymachus would see it, that
Beorhtnoth not only ceases to have a flourishing life, but ceases to live at

If this is Thrasymachus'descriptivethesis about justice, then there are
four interesting consequences. First, it becomes apparent that Thrasymachus is not an 'immoralist' - if by this it is meant that, in Shorey's words,
'Thrasymachus' "Umwertung aller Werte" reverses the normal application' of all moral terms. For (i) Thrasymachus does not reverse, simply hold
the mirror image of, any standard moral views at all. He may believe that
justice is not a virtue, but he does not, ipso facto, believe that justice is a
vice; nor that injustice is a virtue. And (ii) he does not express disagreement
with all standard moral views; he expresses disagreement with standard
moral views about justice. What his views are of the other virtues in Plato's
list-temperance, courage and wisdom - we do not hear. We certainly have
no reason to think that he would deny that courage and wisdom have at least
some importance for human flourishing.
Secondly: we have seen that Thrasymachus' thesis about justice is a
descriptive one, and not a prescriptive one. But we can now see that, quite
apart from his descriptions of the nature of justice, Thrasymachus does
have a general prescriptive thesis, a view about what human flourishing is,
and a more specific prescriptive thesis, a view about what character traits
enable one to flourish, both of which have a bearing on our attitude to
justice. Thrasymachus believes that human flourishing consists roughly in
this: in getting for oneself as much as possible of what are uncontroversially
agreed, in our society as much as in his, to be clear and obvious examples of
good things to have: money (343e), property and valuables like treasure
(344b), and power over others to bend them to my will (344b). Now one
'Text and translationof The Battleof Maldoncan be found in R. Hamer, ed., A Choice
of Anglo-Saxon Verse(London: Faber, 1970).


This content downloaded from on Thu, 21 May 2015 02:09:00 UTC
All use subject to JSTOR Terms and Conditions

character trait which we need to obtain this sort of good life is practical
intelligence. For practical intelligence involves knowning how to use the
device justice; and so what is prescribed for us about justice is that we
should use it, as the device it truly is, to help us achieve just these sorts of
good things. But this recommendation, that by putting practical intelligence to work we use justice as a device, is not a prescriptive thesis about
justice, but about practical intelligence.
Thirdly, and consequent to this, Thrasymachus' stated or implicit view of
human flourishing may even suggest that he has a list of four cardinal virtues
to rival Plato's. At 344c he mentions three of these virtues: 'Thus, Socrates,
when it comes to the full, injustice is stronger and more unrestrained and
more imperious than injustice . . '. Why is injustice to be preferred to
justice? As we have seen, not because it is itself a virtue; rather because it is,
in general and as a rule, more in accordance with the Thrasymachean
virtues of Strength ('ioxtU3),
of Unrestraint(XEvuOEQcta;
for the justification of my translation cp. Gorgias 492c), and of Imperiousness (bEoloWhat is the fourth Thrasymachean virtue? It is not Injustice, for
reasons we have already seen. It is rather that quality of mind which,
Thrasymachus says, discerns the realities of the social institution of justice,
and of which he sees unjust behaviour as being, normally, a manifestation:
namely Practical Intelligence, cv3ouVXca.
The fourth consequetice of this descriptive understanding of Thrasymachus' thesis about justice is that we can now reconcile his remarks about
justice, A-D, with which we began, and also sort out the sheep from the
goats among the commentators' opinions about Thrasymachus, 1-7.
First, then, his observation that justice is 'another's good' (D) means that
going in for justice entails doing like Beorhtnoth, entails being persuaded
or duped by the other person into giving up all your advantages over her. So
when Thrasymachus says that justice is 'the interest of the government in
power' (C), he will mean that whatever government is in power, it will,
wittingly or not, be playing exactly this confidence trick of justice on its
subjects. In persuading them or legislating for them to act justly, it will be
causing them to do what is in fact in its own interest. If the Thrasymachean
account of virtue is right, perhaps the citizens would do much better for
themselves, and for their own interests, if they did not generally act justly by
(to take one prominent example) 'obeying the laws' (B). But they are being
deluded by their rulers into thinking that they have some good reason to act

This content downloaded from on Thu, 21 May 2015 02:09:00 UTC
All use subject to JSTOR Terms and Conditions

justly. And in being so deluded, they are in truth acting as weaklings, not in
their own interest, but 'in the interest of the stronger' (A). Here 'the
stronger' could mean either 'those who (already) have the strength of
character to see through the trick of justice', or 'those who by seeing
through the trick of justice, and seeing how to use it as a device or weapon,
have become the stronger'; or perhaps it could refer to both groups.
As for the commentators: (1) Evidently we do not have to think, with
Maguire, that Plato's Thrasymachus is just confused. (2) So far from his
being a political revolutionary, Thrasymachus' view is that all political
structures tend to have the same effect (338b). The only way in which they
influence human flourishing is that those at the top of political structures
flourish, and those at the bottom do not. The ideal situation, because it
maximises one's power over others, is to be a tyrant (344a). That he
believes this, however, does not mean that Thrasymachus is arguing that
such a situation exemplifies 'Natural Justice' (a good thing) as opposed to
'Conventional Justice' (a bad thing), or that a tyrannical constitution is the
best one. Thrasymachus' question is: 'Best for whom?'. Your justice is good
for me and bad for you, and my justice is good for you and bad for me; but
there is for Thrasymachus no interesting sense in which any old justice,
yours or mine indifferently, is either good or bad without qualification.
(That is the point of his rejection of the prescriptive theses, and of the
assertion that justice is neither a virtue or a vice, but a device.) So likewise,
a tyranny is the best constitution for the ruler and the worst for the ruled; but
there is no sense in Thrasymachus' eyes, in calling tyranny a good or bad
state of affairs without reference to some person's good or bad. In short
Thrasymachus is neither a revolutionary nor a fascist; he's an opportunist.
(3) No doubt Thrasymachus's remarks do bear comparison with, e.g., the
speeches of the Athenian envoys to Melos (Thucydides 5.84-116) - speeches which express just the kind of belief in 'Naturrecht' which Pauly-Wissowa foists on Thrasymachus. (So for example Thuc. 5.89: 'Justice, in human
affairs, is judged according to the balance of compulsion: those with an
advantage do whatever they are able to, those at a disadvantage suffer
whatever they have to'.) On the other hand, Maclntyre seems wrong to say
that Thrasymachus would agree with the 'Thucydidean' thesis that aQETin is
one thing, practical intelligence quite another'. For of course Thrasymachus rejects Plato's conceptions of &QETf and of practical intelligence,
and so must also reject the Platonic ways of connecting them. But Thrasymachus' reason for rejecting these Platonic views is precisely that Thrasyand practical intelligence
machus has his own conceptions of what a'QEnT


This content downloaded from on Thu, 21 May 2015 02:09:00 UTC
All use subject to JSTOR Terms and Conditions

are. But thisfact givesus no reasonto thinkthatThrasymachuscannothave
his own way of connectingpracticalintelligenceandazQvri.Rather,if I am
right, then Thrasymachus'conceptionof practicalintelligenceis in fact a
crucialcomponent of his conceptionof a'QETij.
(4) In the Gorgiaswe find Calliclesassertingseveraldoctrinesfor which
my argumentsuggeststhere is no evidencein Thrasymachus'remarks.For
one thing, Callicles'argument(unlikeThrasymachus')does depend upon
an opposition of 'Naturrecht'and 'ConventionalJustice' (To xa-t'a qprv
xakXOvxai bLxctLov, Gorgias 491e; Tlan7aUC'
492c). For another,Callicles'attackis not merelyon justice'spretensionsto
be a virtue, but also, explicitly, on temperance's:'Since [the many] are
unableto arriveat the fulfillmentof theirpleasures,they praisetemperance
and justice throughtheirown unmanliness'(Gorgias492b). Thirdly,Callicles says that 'Luxuryandintemperanceandunrestraint,if they are backed
with force, are virtue and happiness; (492c: xTuqpixati axokaoIa xai
eav etLxouQIav exi', Tofx' Ot'TvCLxQEIJ
batLovLa).This may suggest that Callicles' list of virtues differs
Thrasymachus'list; althoughpossibly,if one bearsin mindthe evidenceof
and Q6vLoL;)
492a, where courage and practicalwisdom (avbQQIEa
mentioned,they can be harmonised.The ideawouldbe thatCallicles'list of
four cardinalvirtues runs thus: IQVi / &xoXaoLt / XEU0EuOE, E'ItxoiIt can, therefore, be correlated one-one,
QLa,cavbELLa, cpQOVrjoL;.
though not necessarilyidentified, with Thrasymachus'list of cardinalvirThis
6Fecso7Uta, Eci4ouMXa.
tues, which runs thus: 'kEU0E(`a,
way of seeing thingswould have the furtherconsequencethat there is an
interestingsense in which Calliclesis an immoralistand Thrasymachusis
not. For one of the namesof one of Callicles'virtuesis &xokaoLta,
normallyconsidereda vice. So Calliclesdoes stand at least one standard
virtueon its head: somethingwhichThrasymachus,at least in what he says
explicitly,does not do.
(5) Thrasymachuswouldhave agreedwithNietzschethatonly weaklings
go in for justice, andthatthe trueoppositionin valuesis between'good'and
'bad', not between 'good' and 'wicked'.Again, Thrasymachuswould have
found much to admire in the characterof the Nietzschean super-man;
Nietzsche's Ubermensch and Thrasymachus' XQEITTwvhave much in com-

On the other hand, consider Nietzsche's words in the section of Also
SprachZarathustra,entitled 'Of the Virtue that makes small':
'I have found this hypocrisythe worstamongthem:that even those who


This content downloaded from on Thu, 21 May 2015 02:09:00 UTC
All use subject to JSTOR Terms and Conditions

command affect the virtues of those who obey. "I serve, you serve, we
serve" - so here even the hypocrisy of the rulers intones - and alas, if the
first ruler is only the first servant!'
Nietzsche deplores the 'hypocrisy' of the ruler in pretending to the virtues
of his subjects (like the Pope, who is Servus servorum Dei). He deplores
this kind of behaviour because he thinks that it stunts and warps the
development of the Ubermensch. Thrasymachus, by contrast, would applaud this sort of hypocrisy as a piece of riU3ouXta;as Glaucon says in his

recapitulationof Thrasymachus'views, EOX4TI yt

CEbxLa bO6XELV 8&-

xaLov [01 OVTa, 'the height of injustice is to seem just without being just'
(361a). The person in this situation, in fact, gets the best of both worlds: for
she gets all the good repute and honour of justice, without having to suffer
any of its disadvantages. Thrasymachus would no doubt agree with
Nietzsche that, if one is foolish enough to think justice a virtue, then one can
damage one's soul. But provided justice is recognised to be a device and not
a virtue, and used accordingly, it can be a very useful thing to the superior
person on Thrasymachus' view. In short: Nietzsche thinks that the so-called
virtue of Justice, and indeed all 'slave morality', is a device of the weak for
preventing the strong from getting too great an advantage over them;
whereas Thrasymachus, on the contrary, thinks that Justice is a device of
the strong for keeping the weak in their place.
(6) Houranian conventionalism is disposed of by my remarks about the
compatibility of A-D in Section I. And lastly (7) the Kerferd/Foot view,
that Thrasymachus argued from the premise 'that injustice was more profitable than justice' to the conclusion 'that a man who had the strength to get
away with injustice had reason to follow this as the best way of life', is also,
if I am right, to be rejected. For my view has the consequence that injustice,
as such, will not be Thrasymachus' 'best way of life'; for considerations of
justice and injustice will not feature at all in Thrasymachean practical
reasoning, or not at least (to borrow a phrase from Aristotle) as supplying
'premisses of the good'. Hence, though it might on my interpretation of
Thrasymachus' view be true 'that a man who had the strength to get away
with injustice had reason to follow this as the best way of life', such a
Thrasymachean agent's reason to live some form of the unjust life could not
be: 'Because this life is unjust'. The injustice of his preferred life is not
central, but incidental, to the practical reasoning on account of which he
chooses to live that way. For it is true that the person of Thrasymachean
virtue does, typically but not always, do what is unjust; but he does not do
what is unjust under the description 'act of injustice', but under the description 'act of ci4ouXv(a'. His reasons for living like that would be given, not

This content downloaded from on Thu, 21 May 2015 02:09:00 UTC
All use subject to JSTOR Terms and Conditions

by reference to a supposed virtue of injustice; but by reference, firstly, to
the Thrasymachean conception of the flourishing human life which I have
tried to develop, and, secondly, to the human character traits (especially
'ri4ouia) by which the good life so conceived is to be rendered attainable: which is to say, by reference to the virtues of Thrasymachus5.
Wolfson College, Oxford

' I am grateful to Malcolm Schofield and Roger Crisp for written comments on earlier
drafts. In conversation, Steven Everson and David Charles gave useful criticisms.
Elizabeth Telfer, RichardStalley and MaryHaight made valuable points when I presented one form of the paper at Glasgow University, as did Roger Trigg at Warwick
University. I am also indebted to my pupils David Kensinger and James MacLain of
Williams College, Massachussetts, U.S.A., for obliging me, in Trinity Term 1992, to
think harderabout Thrasymachus.


This content downloaded from on Thu, 21 May 2015 02:09:00 UTC
All use subject to JSTOR Terms and Conditions