Phaedo of Elis and Plato on the Soul1

GEORGE BOYS -STONES

ABSTRACT
Phaedo of Elis was well-known as a writer of Socratic dialogues, and it seems
inconceivable that Plato could have been innocent of intertextuality when, excus-
ing himself on the grounds of illness, he made him the narrator of one of his
own: the Phaedo. In fact the psychological model outlined by Socrates in this
dialogue converges with the evidence we have (especially from fragments of the
Zopyrus) for PhaedoÕs own beliefs about the soul. SpeciŽ cally, Phaedo seems to
have thought that non-rational desires were ineliminable epiphenomena of the
body, that reason was something distinct, and that the purpose of philosophy was
its ÔcureÕ and ÔpuriŽ cationÕ. If PlatoÕs intention with the Phaedo is to assert the
separability and immortality of reason (whatever one might think about desire
and pleasure), then Phaedo provides a useful standpoint for him. In particular,
Phaedo has arguments that are useful against the Ôharmony-theoristsÕ (and are the
more useful rhetorically speaking since it is only over the independence of rea-
son that Phaedo disagrees with them). At the same time as allying himself with
Phaedo, however, Plato is able to improve on him by adding to the demonstra-
tion that reason is independent a proof that it is actually immortal.

With a growing acknowledgement that the study of PlatoÕs philosophy

cannot proceed without a sensitivity to the manner of its presentation has
come a growing interest in the characters who people the dialogues. 2 The
assumption – surely right – is that PlatoÕs choice of social milieu, of inter-
locutor, in the case of framed dialogues of the narrator too, must be rel-
evant one way or another to the philosophical subject matter in hand. One
Ž gure in all of this has attracted particular attention over the years, namely
Phaedo of Elis: narrator, interlocutor, and eponym of the Phaedo.3 There

Accepted August 2003

1
This article comes out of work I have been doing for a forthcoming collaborative
study of physiognomy in the ancient world under the general editorship of Simon
Swain (Physiognomy: an Interdisciplinary Study from Graeco-Roman Antiquity to Islam).
My thanks to Christopher Rowe and David Sedley for their critical comments on ear-
lier drafts.
2
See now D. NailsÕ invaluable resource, The People of Plato. A Prosopography of
Plato and Other Socratics (Indianopolis, 2002).
3
Testimonia and fragments of Phaedo in G. Giannantoni (ed.), Socratis et Socraticorum
Reliquiae [SSR] (Naples, 1990), IIIA, with discussion at vol. 4, 115-27. Giannantoni
is rigorous in excluding texts that do not name Phaedo; a more generous collection of
testimonia for the Zopyrus is presented and discussed in L. Rossetti, ÔRicerche sui

© Koninklijke Brill NV, Leiden, 2004 Phronesis XLIX/1

Also available online – www.brill.nl
2 GEORGE BOYS-STONES

seem to be several reasons for this level of interest in him. First, the
Phaedo is one of PlatoÕs most poignant dialogues: the dramatic context
evoked around the impending execution of Socrates asserts itself with
unusual force. Secondly, PhaedoÕs role as the ostensible narrator of the
dialogue is given an unusual emphasis by Plato. The Phaedo not only
opens with an emphatic assertion of PhaedoÕs right to narrate as a wit-
ness of SocratesÕ last hours (aétñw, Î FaÝdvn, pareg¡nou . . .; aétñw, Î
ƒEx¡kratew), but also, extraordinarily, makes a point of explaining why
Plato could not narrate events himself: he was off sick at the time (Pl‹tvn
d¢ oämai ±sy¡nei: 59b10). Finally, it happens that evidence about PhaedoÕs
background from outside the Phaedo provides a ready answer to the ques-
tion of why, from a dramatic point of view, Plato should have chosen to
speak through Phaedo rather than one of the other Socratics who were
supposed to have been present with him in SocratesÕ cell. Phaedo, we are
told, had been a prisoner of war, and made to work as a prostitute. The
analogy with the soul as Socrates describes it in the Phaedo is not hard
to see: for it too, during life, is imprisoned, trapped in polluting service
to carnality. And just as the soul is eventually puriŽ ed and released from
attachment to corporeality through the practice of philosophy, so Phaedo
was liberated from his enslavement at the instigation of Socrates; became,
indeed, a philosopher himself, and the founder of his own school at Elis.4

ÒDialoghi SocraticiÓ di Fedone e di EuclideÕ, Hermes 108 (1980), 183-200, at 183-98.

With most commentators (including Giannantoni: SSR 4.126), I assume as the only
serious hypothesis available for their provenance that the fragments describing the
famous encounter between Zopyrus and Socrates (frr. 6-11 Rossetti) ultimately derive
from PhaedoÕs Zopyrus – although it is true that Phaedo is mentioned in none of them.
On Phaedo generally (including earlier suggestions that we might reconstruct something
of his views from Plato), cf. K. von Fritz, ÔPhaidon (3)Õ, RE xix.2 (1938), 1538-42;
L. Rossetti, ÔTherapeia in the Minor SocraticsÕ, Theta-Pi 3 (1974), 145-57; ÔÒSocraticaÓ
in Fedone di ElideÕ, Studi urbinati (Ser. B) 47 (1973), 364-81, a revised version of
which appears in his Aspetti della letteratura socratica antica at 121-53 (cf. 133-4 for
the suggestion that Phaedo 88e-89a represents PhaedoÕs own view of Socrates);
H. Toole, ÔEÞw poÝon Svkratikòn pr¡pei nŒ Žpodñyú ² dusmorfÛa toè Svkr‹touw. PerÜ
t°w xronolog®sevw tÇn ¦rgvn tÇn SvkratikÇnÕ, Athena 75 (1974-5), 303-17 (esp.
306-7 for the suggestion that Phaedo lies behind PlatoÕs portrait of Socrates in the Sym-
posium); M. Montuori, ÔSu Fedone di ElideÕ, Atti della Accademia Pontaniana 25 (1976),
27-40; S. Duáanić, ÔPhaedoÕs Enslavement and LiberationÕ, Illinois Classical Studies
19 (1993), 83-97 (arguing at 96-7 that the Phaedo has been coloured by a critical atti-
tude to Spartan aggression that Duáanić attributes to Phaedo himself); C. H. Kahn,
Plato and the Socratic Dialogue (Cambridge, 1996), 11-12; Nails, People of Plato 231.
4
So e.g. K. Dorter, PlatoÕs Phaedo: An Interpretation (Toronto / Buffalo / London,
1982), 9-10, 89. For the various accounts of PhaedoÕs life, see SSR IIIA 1-3. E. I.
McQueen and C. J. Rowe have shown that his capture in war is at least historically
 PHAEDO OF ELIS AND PLATO ON THE SOUL 3

Given the ostensible psychological model of the Phaedo, then, it is pos-

sible to start thinking of reasons why PhaedoÕs biography might make him
an appropriate narrator. But it cannot be satisfactory to leave things like
this. For one thing, it is quite possible that ancient biographies of Phaedo
were themselves embroidered in the light of his role in the Phaedo (or
indeed in the light of what I shall go on to suggest was his own philos-
ophy). For another, the ÔdramaticÕ explanation of PhaedoÕs presence in the
Phaedo makes no mention of the one thing we know for sure: that Phaedo
was himself a philosopher of substance and, more than this, a writer of
his own Socratic dialogues. (Two of these, the Zopyrus and the Simon,
preserved his fame throughout antiquity.) 5 This has some claim to be the
more striking fact about his presence in the Phaedo – for how could Plato

possible: ÔPhaedo, Socrates and the Chronology of the Spartan War with ElisÕ,
MŽ thexis 2 (1989), 1-18 (cf. also Duáanić, ÔPhaedoÕs Enslavement and LiberationÕ).
Whether it is plausible that he was prostituted is another question (it is denied by
Montuori: ÔSu Fedone di ElideÕ, 36-40; ÔDi Fedone di Elide e di Sir Kenneth DoverÕ,
Corolla Londoniensis 2 (1982), 119-22). Other explanations for PlatoÕs choice of
Phaedo as narrator of the Phaedo have been suggested. In the view of L. Parmentier,
Plato was paying a simple homage to a dead friend (ÔLÕ‰ge de PhŽ don dÕElisÕ, Bulletin
de lÕAssociation Guillaume BudŽ 10 (1926), 22-4 at 23). W. D. Geddes (Plato, Phaedo
(2nd edn.: London, 1885), xiii-xiv) suggests that Phaedo was chosen as being known
for having the right balance of artistic sensitivity and philosophical acumen for the
occasion. Giannantoni, by contrast, thinks that the choice of Phaedo is motivated pre-
cisely by his insigniŽ cance for the circle of Socrates: by choosing Phaedo as his nar-
rator, Plato ensures that our view of Socrates at such an important moment will not
be clouded by association with a Euclides, Antisthenes, or Aristippus (SSR 4.119).
There is always the possibility that Phaedo was in fact PlatoÕs source for events on
that day: one would not have to assume just because of this that the discussions in
the dialogue were mere transcripts. Cf. R. Hackforth, PlatoÕs Phaedo (Cambridge,
1955), 13; and the cautious remarks of J. R. Baron on the last words of Socrates: ÔOn
Separating the Socratic from the Platonic in Phaedo 118Õ, Classical Philology 70
(1975), 268-9. D. N. Sedley is unusual in looking to PhaedoÕs philosophical position
for an explanation of his presence, observing a Ôphilosophical kinshipÕ between the
Zopyrus and the Phaedo: ÔThe Dramatis Personae of the PhaedoÕ in T. Smiley (ed.),
Philosophical Dialogues. Plato, Hume, Wittgenstein. Dawes Hicks Lectures on Philosophy
= Proceedings of the British Academy 85 (Oxford, 1995), 3-26, esp. 8-9.
5
Out of a longish list of dialogues attributed to him (DL 2.105; Suda s.v. FaÛdvn),
Diogenes Laertius (loc. cit.) thinks these two genuine. The Simon seems to be behind
the invective of one of the ÔLetters of AristippusÕ (no. 13 = SSR IVA 224; cf. Letter
12 (part) = SSR IVA 223, with K. von Fritz, ÔPhaidon von Elis und der 12. und 13.
SokratikerbriefÕ, Philologus 90 (1935), 240-4). Since these letters are later composi-
tions than they pretend to be (cf. e.g. Giannantoni, SSR 4.165-8), this indicates an
interest in the Simon somewhat later than the generation after Socrates. The more dis-
tinguished sources of testimonia for Phaedo include Cicero, Seneca and (into the 4th
century AD) the Emperor Julian.
4 GEORGE BOYS-STONES

be innocent of intertextuality when he writes a dialogue narrated by

another (rival? collaborative?) writer of dialogues? 6 It could be that we
should not just be looking to explain PlatoÕs choice of Phaedo in terms
of what is distinctive about the dialogue he narrates, but that we might
also (conversely) explain what is distinctive about the Phaedo in terms of
PhaedoÕs presence as its narrator.

The Phaedo seems to stand apart from other Platonic texts in the psy-
chological model with which it works. It stands apart, in particular,
through its treatment of desire – desire, that is, for corporeal stimulation
or satisfaction. According to the ÔstandardÕ Platonic account, this sort of
desire forms a distinct ÔpartÕ of the soul, of which another part is reason.
Like reason (with which it may con ict), such desire is a psychological
determinant of action. What makes this a plausible account of desire is,
Ž rst, the very fact that it is one source of impulse for a body whose life
and activity depends on the presence of the soul; and, secondly, the fact
that the pleasure which is posited by desire as the end of human activity
is itself something that registers in the soul. 7 In the Phaedo, however, Plato
appears to be trying something different. According to the Socrates of the
Phaedo, desire is not of the soul at all, but of the body. It has an impact
on the soul (which in essence is pure reason); but as an external distrac-
tion to it, not as a wayward part of it. The idea seems to be that, once
animated by the directive presence of reason, the needs and the satisfac-
tion of the body assert themselves as appropriate objects of reasonÕs care.
In many cases, reason (brought to forgetfulness of its proper, divine sphere
at the moment of incarnation) actually goes so far as to identify its own
interests with those of the body. Nevertheless, the body, and the desires
that come from it, are properly alien to the soul, which stands to them as
a guard to his post (cf. 62b), or a man to his cloak (cf. 87b-e) – or a con-
demned prisoner to his cell, or, if you like, a noble P.O.W. to his igno-
minious bordello.
It would be wrong to deny the familial resemblance between the
Phaedo and other dialogues in which Plato discusses the character of the
soul. In particular, Plato never denies the primacy of reason;8 and if, in

6
The Phaedo is not the only case of this: the Theaetetus is narrated by Euclides,
founder of the Megarian school (cf. DL 2.108 for his dialogues).
7
Compare esp. Philebus 21a-d (for the mental dimension to pleasure); 35cd (for
the location of desire and impulse in the soul).
8
T. Johansen notes references to our rational nature as ÔoriginalÕ in the Republic
(611d) and Timaeus (42d, 90d): ÔBody, Soul, and Tripartition in PlatoÕs TimaeusÕ,
Oxford Studies in Ancient Philosophy 19 (2000), 87-111 at 109 with n. 34.
 PHAEDO OF ELIS AND PLATO ON THE SOUL 5

the Gorgias and Phaedrus, desire is so far from being alien to the soul
that it seems to be an essential and eternal component of it, in the Republic
and Timaeus Plato takes what might be thought of as the middle ground
between this view and that of the Phaedo. According to these dialogues,
desire is of the soul, indeed, but as accident not essence, so that it becomes
a ÔmortalÕ accompaniment to immortal reason which might (in images con-
vergent with the dominant theme of the Phaedo) eventually be ÔpurgedÕ
of it.9
Just as importantly, Plato never denies the crucial role played by the
body in shaping desire, or the irrational soul more generally. 10 Even if the
desire for pleasure springs from the soul, the body, as the means by which
the pleasure is attained, naturally has a signiŽ cant input into the shape
taken by an individualÕs desires. In exploring this aspect of the question,
Plato sometimes sails quite close to the position of the Phaedo – the posi-
tion that desires spring from the body in the Ž rst place. It has recently
been argued for the Timaeus in particular that Plato sees the character of
the soul there in reductionist terms, as ÔfollowingÕ the temperament of the
body; as a straightforward function of the bodyÕs physiological state in
terms very similar to those of the Phaedo. This is how Plato can say, in
the Timaeus, that no-one errs willingly: vice is a result of bodily disease. 11
It seems to me, however, that this cannot be quite right – and that Plato
never (i.e. outside the Phaedo) commits himself to anything stronger than
the claim that the body is one in uence on the character of the irrational

9
Republic esp. 10, 611b-612a; Timaeus esp. 41d, 69cd (42b for the escape of the
just soul from incarnation).
10
Themes in the dialogues which re ect PlatoÕs interest in the scope of the bodyÕs
effect on the character of the soul include speculation about the psychological impli-
cations of the physical environment (e.g. for the character of the Greeks at Timaeus
24c; for the Atlantans at Critias 111e; for Northern races such as Thracians and
Scythians at Republic 435e; cf. also Laws 747de). Or, again, his discussions of the
inheritance of character (e.g. Charmides 157d-158b; Cratylus 394a; hence also the pos-
sibility of breeding for good character: Republic 375-6, 459; Politicus 310 and Laws
773ab; cf. Critias 121b). Neither theme contradicts what I shall go on to argue, namely
that PlatoÕs standard position is that the body does not determine character: both,
rather, operate on the assumption that the nature of the body might predispose some-
one lacking the appropriate control of reason to acquire a certain sort of character.
11
Timaeus 86e. See Christopher Gill, ÔThe BodyÕs Fault? PlatoÕs Timaeus on
Psychic IllnessÕ in M. R. Wright (ed.), Reason and Necessity. Essays on PlatoÕs Timaeus
(London, 2000), 59-84. Gill invokes Galen to his aid; and the language of Ôfollowing
[bodily] temperamentÕ (§pesyai kr‹sesi) is taken from GalenÕs reductionist interpre-
tation of Plato in his QAM (Quod animi mores; or, to give it its full title and in Greek:
†Oti taÝw toè sÅmatow kr‹sesin aß t°w cux°w dun‹meiw §pontai).
6 GEORGE BOYS-STONES

soul. He never in fact says that the body determines oneÕs desires or incli-
nations. The Timaeus, in particular, makes it very clear that reason and
philosophy are forces which counter-balance the in uence of physical
state: a person becomes bad because of a bad state of body and an
Ôupbringing without educationÕ (86e); or where a poor state of body com-
bines with a poor government and poor parenting (87b); the route to hap-
piness involves both physical and intellectual training (88bc).12 There is
no overwhelming reason, whatever the state of a personÕs body, why their
psychology should be marred by bad desires – so long as the natural restraints
of reason are in place. (The reason in question might be oneÕs own or that
of oneÕs parents or society: it makes no difference.)13 In a fully natural
society, nobody would have a bad character at all; and this is not a ques-
tion of the needs or temperament of the body. Desire (and the irrational
soul more generally) remains distinct from the body, and under no com-
pulsion to ÔfollowÕ it.
This, then, is where the model presented in the Phaedo is unique. It is,
to be sure, possible to argue that the Phaedo does not give us a license
to think that Plato changed his mind over the nature of the soul. It is
entirely possible, even probable, that the model we are presented with here
is ultimately intended to be read as emphasising certain features of his
psychological beliefs at the expense of others without actually implying
inconsistency with the ÔstandardÕ view.14 In any case, nothing in what I

12
Gill recognises (ÔThe BodyÕs Fault?Õ, e.g. 60), but plays down (61) the signi-
Ž cance of educational and political in uences on psychological development, partly
because he assumes that oneÕs mental capacities are determined by the body as well.
This itself seems to me mistaken. It is true that Ômadness and ignoranceÕ can be
explained by physical disease (86bc); but the point is that the mindÕs natural activity
here is disturbed by an unusual degree of turmoil in the body, not that it is in gen-
eral a function of physiological state. Cf. e.g. F. M. Cornford, PlatoÕs Cosmology
(London, 1937), 346-9; D. J. Zeyl, Plato, Timaeus (Indianapolis, 2000), lxxxv-lxxxvi.
13
It should be emphasised that both doing and being receptive to philosophy are
entirely natural human functions. (For the place of philosophical education in oneÕs
development, see 44bc; cf. 88bc.) Indeed, both are inscribed in the body, every bit as
much as the tendency towards irrational vice: so manÕs philosophical destiny (cf. 42ab)
is an explanatory factor behind, for example, the structure of the sense-organs (47b-e),
the mouth (75e), and even the gut (72e-73a). Cf. Johansen, ÔBody, Soul, and Tripartition
in PlatoÕs TimaeusÕ, 109-10; C. Steel, ÔThe Moral Purpose of the Human Body. A
Reading of Timaeus 69-72Õ, Phronesis 46 (2001), 105-28. I take it that all of this
allows us to say that even someone who was ÔconstitutionallyÕ mad or ignorant should
in the natural course of things be under the care of others, whose reason would be
substitute for his own in counterbalancing the effects of excessive physical disorder.
14
A common way of doing this is to say that the soul manifests its nature differ-
 PHAEDO OF ELIS AND PLATO ON THE SOUL 7

shall go on to argue offers any sort of challenge to this position (I do not

assume that Plato held or wished to be thought of as holding a different
position on the soul in the Phaedo). Nevertheless, it is important to recog-
nise that there are differences at least in the presentation of PlatoÕs psy-
chology in this dialogue: even if one assumes that that is all they are, they
are still striking enough to need explanation. On a straightforward read-
ing of the Phaedo, desires are presented as functions of the body, and
nothing else; desires can be resisted, but not, during life, eliminated. There
is no scope for ÔharmonisingÕ them with reason or subduing them by it,
because there is, more generally, no possibility for uprooting them from
the body. And this suggests a further divergence from the ÔstandardÕ
Platonic model. For according to the standard model, in which desires are
properly part of the psyche, oneÕs natural character can be worked on and
improved: desire can come under the in uence of reason and be trained
to a better state. But as far as the Phaedo is concerned (at least on a
straightforward reading of it), oneÕs natural character is ineliminably inscribed
in oneÕs body. Of course reason can resist: desire does not determine
behaviour. 15 But reason cannot eradicate or (within broad limits, perhaps)
restrain inclination. In these terms, a person has control of their behav-
iour, but not their character or ÔnatureÕ. And I put it in these terms because
it seems that this might have been exactly what Phaedo of Elis thought.

Our evidence for PhaedoÕs views about human psychology comes from
the fragments of his lost work, the Zopyrus, and in particular from what
seems to have been the central episode of that dialogue in which a visi-
tor to Athens named Zopyrus was prevailed upon to demonstrate on

ently in its incarnate and discarnate states. In particular, if ÔdesireÕ is thought of as

the operation of the soul in respect of the body, it can be considered an essential and
immortal capacity of the soul (which would explain the Gorgias and Phaedrus), yet
ÔmortalÕ in its actualisation, since this does not outlast its connection with the body.
The Phaedo, then, focusses on desire qua mortal and insofar as it is related to the
activity of the body. So e.g. R. D. Archer-Hind, The Phaedo of Plato (New York, 1973),
27-31; cf. R. S. Bluck, PlatoÕs Phaedo (London, 1955), 2-5; L. P. Gerson, ÔA Note
on Tripartition and Immortality in PlatoÕ, Apeiron 20 (1987), 81-96; C. J. Rowe, Plato,
Phaedo (Cambridge, 1993), 9; Johansen, ÔBody, Soul, and Tripartition in PlatoÕs TimaeusÕ.
15
Despite the apparently compelling nature of desires which leave no room for phi-
losophy (66b-d), and the unavoidable demands of pleasure and pain which ÔrivetÕ the
soul to the body (83b-d), reason is capable of maintaining control – in the Ž rst place
precisely by avoiding situations of intense pleasure and pain which might impede its
own activity (83b).
8 GEORGE BOYS-STONES

Socrates his claim that he could divine a manÕs character from his phys-
ical appearance. 16
Our fragments differ in their report of the details, but concur in the
general thrust of what happened. Confronted with Socrates, Zopyrus
announced that he was a man possessed of Ômany vicesÕ (fr. 7 Rossetti);
the thickness of his neck indicated that he was Ôstupid and dullÕ (fr. 6
Rossetti); his eyes showed him to be either a womaniser (frr. 6, 9 Rossetti;
cf. 8), or perhaps a pederast (fr. 11 Rossetti, from Cassian, who is pur-
porting to quote). In either case, SocratesÕ companions, and Alcibiades
in particular, had reason to laugh (frr. 6, 8-10). Socrates no doubt in
PhaedoÕs work as much as in PlatoÕs was a paragon of virtue; not stupid
but the wisest man alive (if it were Plato, one might think of the oracle
reported at Apology 21a); not licentious, but preternaturally abstinent (if
it were Plato, one would hear AlcibiadesÕ laugh and think of the Symposium).
ZopyrusÕ diagnosis must be wrong: his false claims to knowledge ex-
ploded. But Phaedo has a surprise in store. The onlookers laugh at Zopyrus,
and the reader laughs with them; but Socrates tells us all to stop: ÔThis is
how I am,Õ he said (or something like it; see further below); Ôbut through
the practice of philosophy I have become better than my nature.Õ
Is Socrates merely being ironic here? To answer this question we need
context; and it so happens that the one other fragment of the Zopyrus we
have might provide it. We know that someone in the Zopyrus told the fol-
lowing story (fr. 1 Rossetti = SSR IIIA 11):
They say, Socrates, that the youngest son of the King made a pet of a lion
cub . . . And it seems to me that it was because the lion was brought up with the
child that it followed him wherever he went even when he was a young man, so
that the Persians said that it was besotted with the boy.

The story is preserved for no better reason than that the grammarian
Theon thought it a happy illustration of the change from indirect to direct
speech in narrative. It is lacunose; we do not know who told it (though

16
The premise of the Zopyrus might, then, be compared with the starting-point for
some of PlatoÕs dialogues: in the Protagoras too, for example, or the Ion, a foreigner
arrives in Athens with a claim to special expertise. That Zopyrus was a foreigner is
clear from fr. 9 Rossetti (ÔWhen he [sc. Socrates] was alive, a man called Zopyrus
came to Athens . . .Õ). It has been suggested that Zopyrus was, more speciŽ cally, a
Persian, partly because of his name (cf. e.g. Herodotus 3.153-60) and partly through
the circumstantial detail that someone in the dialogue told a story concerning a Persian
prince (quoted below in the text; for the argument, cf. Rossetti, Aspetti della lettera-
tura socratica antica 145-6). Neither piece of evidence is unassailable, however: the
latter is merely circumstantial (and cf. next note); and we know that the name Zopyrus
was not conŽ ned to Persians (Plato, Alcibiades 122ab and below note 20).
 PHAEDO OF ELIS AND PLATO ON THE SOUL 9

Zopyrus seems the best guess on the face of it);17 and, most importantly,
we have nothing at all to suggest where and how the telling of this story
stood in relation to the physiognomical episode. But despite all this, it is
hard to ignore the powerful thematic connections linking this narrative to
the physiognomical episode. In both cases, the ostensible moral is, or
involves, the possibility that a creatureÕs nature can be changed – in par-
ticular that savage and brutal inclinations can be tamed. In the one, we
learn that someone as stupid and licentious as Socrates can become a
chaste philosopher; in the other, that a member of the wildest species of
animal can become as tame and broken as a besotted lover.
Evidence that just such a moral is one that his readership might have
expected from Phaedo comes from the Emperor Julian (Epistle 50, 445A
= SSR IIIA 2):
Phaedo . . . supposed that nothing was beyond the cure of philosophy, but that
everyone can be cured of any kind of life through it – of their behaviour, desires,
everything, in a word, of the sort. If it helped only the well born and well
brought-up there would be nothing amazing about what it did, but if it brings
people in such a bad state to the light, it seems to me surpassing wonderful.

There is no way of telling on which of PhaedoÕs works Julian based his

assessment;18 but it does seem to be the case that the Zopyrus offered at
least one candidate to whom such a moral would be particularly appro-
priate. For there is one further link between the two episodes of the
Zopyrus I have been considering: the person of Alcibiades. According to
frr. 6 and 8 Rossetti, it was Alcibiades in particular who laughed at
ZopyrusÕ diagnosis of Socrates; and, although we have no information that
he was there to hear the story of the lion cub, we might remember
ourselves that Alcibiades compared himself and was compared by others
to a lion. 19 The possibility of this connection is surely strengthened by

17
But the reference to Persia could as well be in deference to ZopyrusÕ presence
as an indication that he was the narrator; and then again, it could be incidental (since
the identiŽ cation of ZopyrusÕ nationality itself rests in part on the reference to Persia
here). Socrates himself is not ruled as the narrator of the story by the fact that it is
addressed to him: the Platonic Socrates, anyway, is quite capable of relating narra-
tives as told to him, or discussions he has had with others (as with Diotima at
Symposium 201d ff., for example) in which he himself is addressed.
18
The Zopyrus is a possibility (so von Fritz, ÔPhaidonÕ, 1540; Kahn, Plato and the
Socratic Dialogue, 12); but Nails pessimistically wonders (The People of Plato 231)
whether he has read any genuine works of Phaedo at all.
19
AlcibiadesÕ leonine nature is suggested by Plato (Alcibiades 122e-123a: cf.
N. Denyer, Plato, Alcibiades (Cambridge, 2001), 186), Aristophanes (Frogs 1431-2),
and Alcibiades himself at Plutarch, Alcibiades 2.2.
10 GEORGE BOYS-STONES

the fact that Alcibiades had a tutor who shared his name with the dia-
logueÕs eponym: Zopyrus. Indeed, it might not be too fanciful to suggest
that the Ž rst readers of the Zopyrus were supposed to assume from the
title that they were purchasing yet another dialogue about Alcibiades and
his education. 20
Whatever the truth of the matter, the important conclusion for now is
that such evidence as we have for the Zopyrus aside from the physiog-
nomical episode hints at a remarkable degree of convergence with a read-
ing of the physiognomical episode that treats it seriously, not ironically;
that at least one of its themes was the transforming power of philosophy.
It starts, in other words, to look as if the Zopyrus as a whole is best served
if Zopyrus really did get Socrates right; and Socrates (as written by
Phaedo) was not being ironic in defending him and confessing to a wicked
nature. The point is his reform through philosophy. But if this should be
accepted, then we can surely go further and ask by what mechanism
Phaedo might have explained all this. What might Phaedo have believed
about the soul to lead him to the conclusion that ÔnaturalÕ character man-
ifested itself in physical appearance, but was the kind of thing which phi-
losophy could overcome?
There is, of course, no reason at all to suppose that Phaedo ascribed to
Zopyrus a theoretical view of the soulÕs relationship with the body, or that
Socrates was supposed to be in agreement with him about this. In fact,
the dynamic of the dialogue would be better explained if Zopyrus had no
theory at all. If PhaedoÕs dialogues were anything like PlatoÕs, it is a fair
bet that Socrates spent a good deal of his time talking precisely to peo-
ple whose abilities ran ahead of their capacity to give them a theoretical
underpinning – whose ÔskillsÕ and ÔvirtuesÕ were empirical, where they
should have been knowledge-based. 21 Perhaps Zopyrus was like this: a

20
Rossetti supposes that AlcibiadesÕ tutor (for whom, see Plato, Alcibiades 122ab)
was the dialogueÕs eponym (ÔÒSocraticaÓ in Fedone di ElideÕ, 371; more cautiously at
Aspetti della letteratura socratica antica 145; cf. also Nails, The People of Plato 305
s.v. ÔZopyrusÕ). But this seems unlikely if PhaedoÕs Zopyrus was a stranger in Athens
at the time of his encounter with Socrates – and especially if he really was a Persian
(cf. note 16): AlcibiadesÕ tutor was Thracian (Alcibiades 122b). For Alcibiades as a
stock Ž gure of the Socratic dialogue, cf. Denyer, Plato, Alcibiades 5, noting that Aeschines
(DL 2.61), Antisthenes (ib.; also DL 6.18), and Euclides (DL 2.108) as well as Plato
and Phaedo himself (Suda s.v. FaÛdvn) are all credited with dialogues named after
Alcibiades.
21
The bravery of Laches, despite his inability to deŽ ne bravery in the Laches,
would be a good example. Cf. for the theoretical point the two types of physician at
Plato, Laws 720ab, who share the same title Ôwhether they are free men – or whether
 PHAEDO OF ELIS AND PLATO ON THE SOUL 11

capable enough physiognomist but, like those sophists in Plato whose dis-
course was also about the soul (cf. Phaedrus 271cd), lacking in an under-
standing of the virtues and vices of the souls he judged. Knowledge about
the soul, then, is what Zopyrus leaves for PhaedoÕs Socrates to explore;
and Socrates for his part need not be expected to reject physiognomy (as
PlatoÕs Socrates does not exactly reject rhetoric), but he might want to
show how it is only a worthwhile pursuit if it can be made philosophical,
grounded on knowledge about the soul. Perhaps this is why he stopped
his companions from laughing.
But what, then, did PhaedoÕs Socrates think about the soul? The ques-
tion is complicated by the fact that our sources differ over what, exactly,
he replied to his companionsÕ mirth at ZopyrusÕ diagnosis. In particular,
there is a difference over whether Socrates changed his natural character
(and so, whether oneÕs nature is changeable) or whether he rather acted
in despite of it (so that oneÕs ÔnatureÕ turns out to be something immutable
but non-determinative).
Cicero certainly talks as if he saw the former model in his source.
According to him, Socrates admits to having been born with the vices
identiŽ ed by Zopyrus, but states that he managed to rid himself of them
through the practice of philosophy:
[Of the vices ascribed to Socrates by Zopyrus:] It is possible that they were born
from natural causes; but it is not due to the power of natural causes that they
were rooted out and altogether removed so that he himself was called away from
those vices to which he had been prone (fr. 6 Rossetti = Cicero, de fato 10).
[Zopyrus was defended by Socrates . . .] who said that, although those vices had
been implanted in him, he had cast them out of himself by reason (fr. 7 Rossetti
= Cicero, TD 4.80).

The fact that Zopyrus was able to get at the nature with which Socrates
was born would suggest that the body is somehow or other a force that
predisposes oneÕs irrational nature to develop in a certain way, or lays
down its ÔdefaultÕ; but it cannot determine oneÕs nature, which is open to
the healing in uence of reason as well. Cicero, in fact, ascribes to Phaedo
something not unlike the psychology I argued was championed by Plato
in the Timaeus – with the difference only that the irrational soul (or its
equivalent) starts, not as a blank sheet subject to the in uences of body

they are slaves and acquire their skill by following their mastersÕ orders, and by obser-
vation and experience. They do not acquire it from [an understanding of] nature, like
the free men, who learn it themselves and teach it this way to their own sons.Õ
12 GEORGE BOYS-STONES

and reason, but as a comprehensive sketch of the body subject to the era-
sure and correction of reason later on.
Cicero, however, is at odds with the majority of our fragments; and,
while a head-count is no way to make a judgement in these cases (we
have no way of knowing whether or when our fragments are based on a
reading of the original, and no evidence for their source if they are not),
the principle of cui bono? speaks against him too. For one thing, the bipar-
tite psychology implied by Cicero may come close to his own preference
in the area.22 For another, such a psychology provides him with the most
robust stand-point in the particular polemical context from which his evi-
dence comes.23 It seems more plausible all round, then, to prefer the tes-
timony of the remaining fragments, which all imply that Socrates some-
how retained the evil nature identiŽ ed by Zopyrus, but did not let that
interfere with his life:
fr. 8 (= Scholia to Persius 4.24): Ôsum quidem libidinosus; sed meum est ipsam
libidinem vincereÕ
fr. 9 (= ps.-Plutarch, perÜ ƒAsk®sevw, versio Syriaca f. 179):24 ÔIn Wirchlichkeit
hat dieser Mann nicht gelogen, denn von Natur neige ich sehr zur Begierde [sc.
nach den Weibern], durch angewendete Sorgfalt aber bin ich, wie ihr mich kennt.Õ
fr. 10 (= Alexander, de fato 6): ·n gŒr n toioètow, ÷son ¤pÜ t» fæsei, eÞ m¯
diŒ t¯n ¤k filosofÛaw skhsin ŽmeÛnvn fæsevw ¤g¡neto
fr. 11 (Cassian, Collationes 13.5.3: note the suggestion that he is quoting the orig-
inal): «eÞmÜ g‹r, ¤p¡xv d¡», id est: Ô. . . etenim sum, sed contineoÕ.

22
Cf. S. Bobzien, Determinism and Freedom in Stoic Philosophy (Oxford, 1998),
298. We know that Cicero shared CarneadesÕ inclination towards a more or less Peripatetic
ethics (e.g. de ofŽ ciis 1.2 with Academica 2.139 and M. Ducos at R. Goulet (ed.),
Dictionnaire des philosophes antiques (Paris, 1994), A28 for the ÔCalliphonÕ men-
tioned there): this may well have been associated with an inclination towards a more
or less Peripatetic psychology.
23
Cicero, in de fato 10, is arguing against the StoicsÕ claim that a personÕs actions
are fated because determined by their nature. His response is not to deny the link
between oneÕs actions and oneÕs nature. (If he did so, the Stoics would complain that
he had removed the guarantee that one is responsible for what one does.) Rather, he
argues that we avoid the snares of fate because we have the freedom through reason
to make our natures what we wish: witness Zopyrus and Socrates. See again Bobzien,
Determinism and Freedom 297-8.
24
The translation (from the Syriac in which this work is preserved) is from
J. Gildemeister & F. BŸcheler, ÔPseudo-Plutarchos perÜ Žsk®sevwÕ, Rheinisches Museum
27 (1872), 520-38.
 PHAEDO OF ELIS AND PLATO ON THE SOUL 13

ÔI amÕ, says Socrates (frr. 8, 11); ÔI do inclineÕ (fr. 9); Socrates has become
better than a nature which he nevertheless retains (fr. 10).
The evidence, I am suggesting then, attributes the following claims to
Phaedo: (1) that each person has a ÔnatureÕ which encompasses their irra-
tional impulses; (2) that this ÔnatureÕ is related to the body in such a way
that an expert in the matter could deduce the former from the appearance
of the latter; and (3) that oneÕs ÔnatureÕ does not determine behaviour. If
we assume, as seems likely, that what determines behaviour in cases
where nature does not is reason, then Phaedo is working with a bipartite
model of behaviour familiar enough from Plato and Aristotle. Where
Phaedo now seems to differ, however, is in the claim that irrational urges
are no more susceptible to training or rehabituation than the set of the
eyes or the shape of the neck. How could he claim this?
One possibility is that Phaedo believed something a little bit like Plato:
at least that there are rational and irrational parts to the soul; but that he
believed in addition (and unlike Plato) that the irrational part automati-
cally throws its lot in with the body and remains throughout deaf to rea-
son. Possible but, it seems to me, unlikely: such a model has no parallels
in antiquity; and one might wonder in a case where the irrational was so
fully determined by the body what advantage there might be in claiming
that it was different from the body at all.
Another possibility, then, is that Phaedo held something like an Ôemer-
gentistÕ view of the soul. The idea would be that ÔpsychologicalÕ func-
tions (including desire and reason) somehow supervene on physiological
activity, but that reason acquires in its turn a causal efŽ cacy which is inde-
pendent of the body. Philosophically, this is undoubtedly a more attrac-
tive view; and I think it cannot be positively ruled out for Phaedo. But it
also has historical problems to contend with: for the only (other) evidence
for emergentist theories of soul in antiquity suggests that they post-date
Aristotle and, more than this, makes it look as if they were inspired by
him.25 The form of psycholog ical reductionism known to Plato and

25
Dicaearchus, a pupil of AristotleÕs might be one early example: R. W. Sharples,
anyway, ascribes an emergentist view of the soul to him (ÔDicaearchus on the Soul
and on DivinationÕ in W. W. Forenbaugh and E. SchŸtrumpf (edd.), Dicaearchus of
Messana. Text, Translation and Discussion (New Brunswick, 2001), 143-73; though
see against this V. Caston, ÔDicaearchusÕ Philosophy of MindÕ, ib. 175-93; and cf.
H. B. Gottschalk, ÔSoul as HarmoniaÕ, Phronesis 16 (1971), 179-98). For the emer-
gentist position of a later Aristotelian, Alexander of Aphrodisias, see CastonÕs ÔEpipheno-
menalisms, Ancient and ModernÕ, The Philosophical Review 106 (1997), 309-54, esp.
347-9.
14 GEORGE BOYS-STONES

Aristotle, on the other hand, seems to have relied on a simple identiŽ cation
of the soul with a ÔharmonyÕ of physical elements; to have been a form
of reductionism which left no room for the independent activity of reason
so essential for Phaedo. 26
If this represents the historical situation fairly, then we are left with a
third possibility, which falls somewhere between the last two and has, it
seems to me, the best measure of philosophical and historical plausibility.
It could be that Phaedo believed in an independent, rational soul on the
one hand, and explained desires on the other as physiological (epi)phe-
nomena. Their corporeal roots would explain why physical appearance
can be used as a guide to their character (oneÕs ÔnatureÕ); and also why
reason could have no effect on them: one could no more change oneÕs
ÔnatureÕ through reason than one could improve oneÕs physical appearance
by thinking about it. What reason can do, however, is to take charge: a
person can make a rational choice to organise their life in any way they
see Ž t, despite the predispositions written into their physiology: reason is
precisely not determined by the body in which it resides. And it seems to
me that this possibility gains credibility precisely through its convergence
with the ostensible position of the Socrates of the Phaedo.

My suggestion, then, is that the distinctive features of the psychology of

the Phaedo (allowing, as I said above, that these might be differences merely
of presentation) can be explained if we assume that it is PlatoÕs intention
to invoke the philosophy of Phaedo of Elis. What is distinctive about
PlatoÕs psychology as presented in the Phaedo is precisely what was dis-
tinctive about the psychology of the historical Phaedo. If this is right, though,
there is an obvious question: why might he have wished to do this? What
end would an invocation of PhaedoÕs psychology serve for him?
In a certain way, this question might be thought to be the same as the
question that, whether I am right or wrong about PhaedoÕs presence in the
Phaedo, faces all commentators on the dialogue: why does Plato present
a model of the soul in that dialogue that seems so different in important

26
For Plato, see Phaedo 85e-86d (where such a theory is outlined by Simmias) and
91b-95a (where it is refuted by Socrates). For Aristotle, see de anima 1.4, 407b27-
408a30; also fr. 45 Rose3 (from his Eudemus). Arguments ex silentio are never ideal;
but if Phaedo had been an emergentist, and Plato knew it, the decision to make Phaedo
the narrator of the Phaedo would have been very strange indeed. The attack on reduc-
tionism so important for establishing the immortality of the soul would be fatally
undermined by the constant reminder that Phaedo himself held an alternative form of
epiphenomenalism less vulnerable to much of SocratesÕ argument.
 PHAEDO OF ELIS AND PLATO ON THE SOUL 15

respects from his discussions elsewhere? The answer often given to this
question is that the differences are more of presentation than of substance:
that Plato here wishes to examine the soul from a certain point of view,
wishes for example to focus on reason as the soulÕs essential attribute.27
But if this is the right answer, then it already provides an explanation for
why he might adopt PhaedoÕs viewpoint: it is precisely by his engagement
with PhaedoÕs psychology that Plato adopts the appropriate point of view
(the appropriate focus on reason). Of course, Plato might have focussed
his discussion in the appropriate way without invoking Phaedo; but by
putting his discussion in PhaedoÕs hands, so to speak, he can remind the
reader that the discussion of the soul in the Phaedo is precisely one per-
spective: a ÔPhaedonianÕ perspective, not the whole Platonic story.
But whether this is right or wrong, there is something else to consider
here as well. The Phaedo is not just an exploration of the soul (from what-
ever perspective); it is, more speciŽ cally, a discussion of its immortality,
and part of the reason why Plato presents the psychological model that he
does (or in the way that he does) must be that he thinks it allows him to
argue the soulÕs immortality more clearly or more securely. One might
think, for example, that by associating desire with the body and identify-
ing ÔsoulÕ with reason, Plato leaves himself free to argue for what really
matters, namely the immortality of reason, without getting bogged down
in objections that someone might raise against arguments which implied
a commitment to the immortality of ÔphysicalÕ desire as well.28 In other
words, the position that Plato adopts in the Phaedo allows him a clear run
at showing in the strongest possible terms that the minimum one would
have to believe about the soul is that reason at least is separable from the
body and not liable to dissolution. But such a position is not only one that
might be most easily be made from a psychological perspective like that
of Phaedo; it could be that we have as a matter of fact already seen the
Ž rst step in the argument towards it in PhaedoÕs Zopyrus. For one of the
main points of the Zopyrus was SocratesÕ assertion that reason always

27
See again note 14 above.
28
There will be time elsewhere for Plato to make clear his views about the psy-
chological status of non-rational impulse. In fact, the need for a further discussion of
pleasure in particular is cued, perhaps, in SocratesÕ re ection at the beginning of the
Phaedo that pleasure and pain always come together (60bc; cf. perhaps, as an alter-
native – or additional – reading to the one suggested in note 35 below, PhaedoÕs Ômix-
tureÕ of pleasure and pain at 59a). The re ection seems to have little signiŽ cance for
the immediate discussion, but lies at the heart of the analysis of pleasure in the Gorgias
(496c-497a) and, especially, Philebus (31b ff.).
16 GEORGE BOYS-STONES

remains in control of oneÕs behaviour and actions, whatever oneÕs irra-

tional ÔnatureÕ. According to Phaedo in the Zopyrus, in other words, rea-
son could con ict with and overcome non-rational impulses. But this, the
moral of the Zopyrus, turns up in the Phaedo as one of the most com-
pelling arguments moved against SimmiasÕ theory that the soul might be
no more than an epiphenomenon – the ÔharmonyÕ – of the body. So at
94bc (but see all of 94b-95a):
ÔWell then,Õ he [Socrates] said, Ôof all the things in a man do you say that it is
anything other than the soul, especially if it is wise, that rules him?Õ
ÔNo, I donÕtÕ [said Simmias].
ÔDoes it acquiesce in the passions of the body, or oppose them as well? I mean,
for example, when there is heat and thirst in us, can it drag us in the opposite
direction, to abstain from drinking; and when there is hunger in us, to abstain
from eating? And in thousands of other matters we see the soul opposing the
passions of the body, donÕt we?Õ
ÔAbsolutely.Õ
ÔBut didnÕt we agree earlier that, if the soul is a ÔharmonyÕ, it would not sound
in opposition to the way its elements were stretched and relaxed and plucked, or
however affected, but would follow them and never take the lead?Õ
ÔOf course we agreed,Õ he said.

The soul, insofar as it can take the lead – and Ôespecially if it is wiseÕ,
i.e. especially insofar as reason is engaged – can con ict with and over-
come bodily urges; hence the soul (especially reason) cannot, after all, be
a ÔharmonyÕ of physical elements. Note, by the way, that this argument
does not ultimately have to rely on a belief that the desires themselves
are bodily (though I have argued that Phaedo himself happens to have
thought that): the phenomenon of psychological con ict in general would
by the same line of reasoning demonstrate that at least part, but perhaps
the whole of the soul is distinct from the body. But it is clearer as an
argument against the harmony-theorists if one phrases it as if from such
a position; and, as conceding more to the harmony-theorists (namely, that
desires at least are functions of bodily state), is arguably more rhetorically
effective against them.
Someone resistant to my suggestion that Plato deliberately invokes
PhaedoÕs psychology in the Phaedo (or adopts a Phaedonic perspective in
his own exploration of the soul there) will naturally suppose that the co-
incidence between PhaedoÕs own assertions about reasonÕs relationship to
desire in the Zopyrus and the argument against Simmias in the Phaedo is
no more than that – coincidence. But it is not merely wishful thinking that
leads me to make the comparison. There is a powerful suggestion in the
 PHAEDO OF ELIS AND PLATO ON THE SOUL 17

text itself that Plato might have had Phaedo in mind during his attack on
the harmony-theory mooted by Simmias. My argument so far has dealt
with two Phaedos: the historical Phaedo, philosopher of Elis, and the Platonic
character, Phaedo, who narrates the Phaedo and SocratesÕ arguments in
that dialogue. But there is a third Phaedo to be reckoned with here as
well. For Phaedo is also an interlocutor in the Platonic dialogue he nar-
rates, and Phaedo the interlocutor becomes important precisely as Socrates
is about to make his Ž nal assault on the crucial objections to the hypoth-
esis of the soulÕs immortality made by Simmias and Cebes.
PhaedoÕs appearance within the frame of the Phaedo can be easily sum-
marised – they are not, as it happens, very many. We learn, Ž rst of all
and early on, that on the day of SocratesÕ execution he was suffering, not
pity for Socrates, but a strange ÔmixtureÕ of pleasure and pain (58e-59a).
His next appearance is 30 Stephanus pages later, when we are told that
although, along with the others, he had been convinced by SocratesÕ ear-
lier arguments for the immortality of the soul, he was ÔunpleasurablyÕ dis-
turbed by the objections of Cebes and Simmias and thrown into doubt
again (88c). Socrates teases him for his long hair, and correctly guesses
that Phaedo was expecting to cut it in mourning for him (89ab). But
Socrates thinks that he should cheer up: the arguments can be defeated,
that there will be nothing to mourn for. Indeed, Socrates pledges to help
Phaedo defeat them: Phaedo will be Heracles, Socrates his Iolaus (89c).29
Phaedo reverses the roles (he will be Iolaus; Socrates should be Heracles);

29
The allusion is to HeraclesÕ encounter with the Lernaean Hydra, during which
he was attacked as well by a giant crab, and required the assistance of Iolaus. For the
story, see e.g. Apollodorus, Library 2.5.2. It is worth speculating whether the appeal
to ÔHeraclesÕ here might have particular resonance for readers of Phaedo: it seems,
anyway, that the eponymous cobbler of his Simon ÔrefutedÕ the Encomium of Heracles
by Prodicus (SSR IVA 224.1-4; for ProdicusÕ Encomium, cf. Xenophon, Memorabilia
2.1.21-34). But this is not the only occasion on which Plato alludes to the myth: for
his richly suggestive use of it at Euthydemus 297b-d, see R. Jackson, ÔSocratesÕ Iolaos:
Myth and Eristic in PlatoÕs EuthydemusÕ, Classical Quarterly 40 (1990), 378-95.
R. Burger (The Phaedo: A Platonic Labyrinth (New Haven, 1984), 159-60) is unusual
among commentators in trying to explain the image as it occurs in the Phaedo. His
suggestion is that Phaedo qua narrator fulŽ ls the role of Iolaus through his hesitancy
in reporting Ôwhat should presumably constitute the philosophical peak of the dia-
logueÕ. The idea is that he thereby exempliŽ es, Ôhowever unwittingly, the inevitable
ÒimpurityÓ of the procedure of hypothetical reasoningÕ and Ôshows himself to be a
most appropriate Iolaus to SocratesÕ Heracles in the battle for the salvation of the
logosÕ. This explanation seems rather forced, however: in making PhaedoÕs labour the
Ôsalvation of the logosÕ rather than that of the soul; in making the narrator of SocratesÕ
words his assistant as such; and in making his assistance so negative.
18 GEORGE BOYS-STONES

Socrates says that it will be all the same, and goes on to address to Phaedo
his remarks about ÔmisologyÕ. Once again, Phaedo drops out of the nar-
rative until very near the end (at 117c), when he breaks down in tears –
not, he says, for the fate of Socrates, but for his own misfortune in losing
such a friend.
That Phaedo the interlocutor has such a small role might not seem too
disturbing at Ž rst. It might be assumed that he is mentioned at all only to
remind us that Phaedo the narrator was present at the events he is nar-
rating; an assertion of his right to narrate equivalent in its own way to the
repeated aétñw [sc. paregenñmhn] with which the dialogue opens. In this
case, we would not expect him to intrude himself into the conversation
more than necessary. But there is, in fact, a good reason to think that
PhaedoÕs relative lack of involvement as an interlocutor in the discussions
about the soul has a positive signiŽ cance. The reason is that is that his
relative silence problematises his one small moment of glory. For right at
the heart of the dialogue, after the crucial challenges by Simmias and
Cebes, just before the argumentative climax which is their refutation, Socrates
appoints Phaedo as the Heracles who will tackle them (89c):
ÔIf I were you and the argument  ed me I would swear an oath like the Argives,
not to allow my hair to grow before I had fought and defeated the argument of
Simmias and Cebes.Õ
[Phaedo replies:] ÔBut,Õ I said, Ôeven Heracles is said not to have been able to
deal with two.Õ
ÔThen, while there is still light, call me to your aid as Iolaus,Õ he said.
ÔI call you to my aid, then,Õ I said: Ônot as if I were Heracles, but as if I were
Iolaus calling Heracles.Õ
ÔIt wonÕt make any difference,Õ he said.

It is important that Phaedo himself accepts the image thus elaborated by

Socrates, though he reverses the roles (he will be Iolaus and aid the Heraclean
Socrates): for all the modesty of the transposition, he accepts, thereby, that
he has a crucial role to play. This is important because the truth of the
matter is that Phaedo does nothing at all: he nods agreement to SocratesÕ
discourse on misology which follows, and then shrinks again into the
background until his tearful outburst at 117c. He does not lead the charge
against Simmias and Cebes – in fact Socrates explicitly marches against
their combined argument in the singular (oêtvsÜ ¦rxomai ¤pÜ tòn lñgon:
91b). Nor does he lift a Ž nger to help (the Ž rst person plurals with which
Socrates surveys the progress of the argument at 95a are inclusive of
Cebes and Simmias as his interlocutors there). So why does Socrates
 PHAEDO OF ELIS AND PLATO ON THE SOUL 19

invoke the comparison? And why does Phaedo accept it, albeit with a
modiŽ cation of parts?
The answer is not in the text: Phaedo does nothing at all. So perhaps
Socrates is looking to a future that lies outside of the text? Phaedo must
have been a young man at the time of SocratesÕ death; it has been sug-
gested that the long hair remarked on by Socrates in the immediately pre-
ceding passage is meant, one way or another, as an indication of the fact.30
Certainly, he is not within the text ascribed a ÔmatureÕ philosophical posi-
tion: he is, for example, convinced and then thrown into doubt again about
the immortality of the soul. Now Plato quite often (presumably rather
more often than we can tell) plays with his readerÕs knowledge of what
was to become later on of characters in his dialogues – the historical fate
of some;31 the philosophical fate of others.32 Consider, in particular, the
difference it makes to our reading of SocratesÕ comments on Isocrates at
Phaedrus 278e-279a that we possess so much of his work. Perhaps, then,
the suggestion that Phaedo will perform Heraclean deeds in support of the
position Socrates goes on the develop against Simmias and Cebes is meant
to make us look forward to PhaedoÕs future achievements. If Phaedo was,
in his own philosophical career, known precisely for the development of
arguments which could be used against positions such as that of Simmias
(the position that the soul, reason and all, was an epiphenomenon of the

30
It has been taken to indicate that Phaedo was still a boy (e.g. Burnet, Plato,
Phaedo, ad 89b2; R. S. Bluck, Plato, Phaedo (London, 1955), 34); or, since Socrates
Ôused to teaseÕ him for the length of his hair, it might indicate that he had, by Athenian
conventions, outgrown the style, and was a young man. So L. Robin (Platon, PhŽ don
(Paris, 1926), p. x) followed by Rossetti (Aspetti della letteratura socratica antica
122-6), and Rowe (Plato, Phaedo, 212 ad 89b3-4). (McQueen and Rowe argue in fact
that Phaedo must have been around 20-22 at the time of SocratesÕ death: ÔPhaedo,
Socrates and the Chronology of the Spartan War with ElisÕ, 2 n. 7 with 14 n. 65.) It
should be noted that not everyone thinks the hair signiŽ cant (Giannantoni, SSR 4.119),
and of those who do, not everyone reads it as a sign of PhaedoÕs age. Some see pro-
Spartan afŽ liation in it (e.g. Parmentier, ÔLÕ‰ge de PhŽ don dÕElisÕ 22-3; Montuori, ÔSu
Fedone di ElideÕ 35-6; Nails, People of Plato 231); J. Davidson (Courtesans and
Fishcakes: The Consuming Passion of Ancient Athens (New York, 1998), 332 n. 56)
suggests an allusion to PhaedoÕs time as a prostitute.
31
E.g. on Cephalus, M. Gifford, ÔDramatic Dialectic in Republic Book 1Õ, Oxford
Studies in Ancient Philosophy 20 (2001), 35-106 esp. 52-8.
32
There are ways of twisting the trope too: the Ômight have beenÕ of Theaetetus
(Theaetetus 142c), for example, is already negated by his impending death (142b); the
promise of the youthful Socrates in the Parmenides, on the other hand, seems alto-
gether too great, since he seems to have PlatoÕs theory of forms well on the way to
completion.
20 GEORGE BOYS-STONES

body), then SocratesÕ suggestion that they will combine efforts to defeat
Simmias and Cebes starts to make a lot of sense: it is, as it were, an indi-
cation that the arguments that follow are not just SocratesÕ arguments, but
rely on or include positions developed by (the historical) Phaedo as well.
As I have already suggested, PhaedoÕs contribution would be precisely to
have shown the fallacy inherent in a position like that of Simmias by
which we might be led by the plausible suggestion that some psycholog-
ical functions are dependent on and cannot outlive the body (namely, func-
tions such as those desires whose object of care is the body) to the more
general claim that the whole soul is dependent on the body. What Phaedo
has, on my reconstruction, established is that, even if one supposes desires
to be corporeal, reason can still be thought of as independent of the body;
and, more than this, that the fact that rational choice might actually
con ict with and override desires and inclinations generated by the body
shows that it must be thought of as such. This, as I noted, is the point of
the encounter between Socrates and Zopyrus; and it is the argument
against Simmias given at Phaedo 94b-95a.
None of this, of course, starts to answer Cebes, who worried that, even
if Simmias was wrong and one could prove that the soul is independent
of and outlasts the body, one has nevertheless not yet proved that the soul
is immortal. (The soul might be like a man who outlasts a series of cloaks,
but nevertheless dies in the end: 86e-88b.) But this is quite consistent with
the idea that Phaedo is invoked as SocratesÕ collaborator. If Phaedo could
provide arguments that were valid against Simmias, we have no evidence
that his own work could be taken to furnish arguments against Cebes –
no evidence, in fact, that Phaedo addressed the immortality of the soul as
such (however inclined one might be to assume that he believed in it).
Neither is it possible that the argument we get against Cebes in the Phaedo
(at 95a-107a) was drawn from Phaedo – not least because it is premised
on a characteristically Platonic theory of forms.33 This, then, is why
Phaedo needs an Iolaus (or is an Iolaus in need of a Heracles): it is only
by the combination of his argument with arguments supplied through Socrates
by Plato that the combined threat of Simmias and Cebes is Ž nally
defeated.

33
Characteristically Platonic theory also plays a role in one of the arguments
brought against Simmias, viz the appeal to the theory of recollection (91e-92e: the
soul could hardly remember knowledge it acquired before entering the body if the soul
were an epiphenomenon of the body). This however, is only one of the arguments
against Simmias: a Platonic addition, I am suggesting, to an argument found also in
Phaedo.
 PHAEDO OF ELIS AND PLATO ON THE SOUL 21

Phaedo, in short, is a useful ally for Socrates in the refutation of

Simmias and Cebes if he was known to PlatoÕs readers as a champion of
a radical distinction between the irrational and the rational in discussions
of the relationship between the soul and the body – and all the more use-
ful as an ally if I was right earlier to suggest that he accepted the force
of a Ôharmony-theoryÕ of soul, at least to the extent of accepting that
desires, and irrational impulse more generally, were physiological epiphe-
nomena. On the back of his arguments, Socrates can add his Ž nal proof
of the soulÕs immortality and secure his position against Cebes as well as
Simmias. So does this make Phaedo a Heracles or an Iolaus after all?
Socrates says that Ôit wonÕt make any differenceÕ. The point, perhaps, is
that it depends on how one reads SocratesÕ appropriation of (the histori-
cal) Phaedo. The arguments of the historical Phaedo came Ž rst; they are
also are logically prior, insofar as one Ž rst has to establish that the soul
is independent of the body before the question of its longevity is raised;
and Ž nally it might be supposed that the position they attack (psycholog-
ical epiphenomenalism) is the more seductive and the more dangerous of
the objections posed by Simmias and Cebes. For these reasons, then, one
might well think of him as the Heracles whose position is shored up in
SocratesÕ appropriation of him and the addition of the arguments against
Cebes. In another sense, however, Socrates is better seen as the Heracles
of piece: he draws on the arguments of the historical Phaedo (who to this
extent helps), but the labour is his labour: it is Socrates, not Phaedo, who
addresses the issue of immortality as such; and his is the deŽ nitive solu-
tion. 34 But whichever way one reads the situation, the arguments are not
affected. Philosophically speaking (and that, after all, is what is impor-
tant), it wonÕt make any difference.

According to my reconstruction, then, Phaedo of Elis thought that desires

and emotions were epiphenomena of physiological states; were to be
identiŽ ed with particular temperaments of the body. 35 To this extent,

34
The implication, in this case, would be that Plato has outdone Phaedo – an impli-
cation present already, perhaps, in the very fact of his appropriation of him. If G. W.
Most is right that the Phaedo (at least the end of the Phaedo) is intended to secure
PlatoÕs claim to be SocratesÕ legitimate heir, there might be a sharper polemical edge
to the suggestion (ÒA Cock for AsclepiusÓ, CQ 43 (1993), 96-111; cf. G. Tanner,
ÔXenophonÕs Socrates – Who were his informants?Õ, Prudentia 28 (1996), 35-47, esp.
42-3, arguing that there was rivalry between Plato on the one hand and Phaedo and
Xenophon on the other). But see note 36 below for an alternative interpretation of the
last words of Socrates, on which Most bases his argument.
35
There might even be a hint at this position in PhaedoÕs description of himself at
22 GEORGE BOYS-STONES

Phaedo could understand the force of psychological epiphenomalism. But

he argued that one must not be misled (perhaps by the economy of the
thought) into thinking that reason comes about in the same way. In fact
reason is a qualitatively different feature of human activity, and something
that is in no way deŽ ned by, or dependent on, or even (therefore?)
conŽ ned to the body – as emotions and desires seem to be. When the
activity of reason is correctly analysed, when one considers its capacity
to oppose the body and corporeal desires in directing action, one sees that
it must be something distinct and free.36 Apart from the particular argu-
ment against Simmias which I have suggested might have made its way
into the Phaedo from PhaedoÕs own works, PhaedoÕs psychology more
generally thus provides the ideal starting-point for PlatoÕs readers to think
about the immortality of the soul as the topic is developed in the Phaedo,
and the ideal standpoint from which Plato himself could defend the posi-
tion that reason at the very least must be immortal.
I should emphasise that I am not in all of this trying to claim that the
Phaedo at any point represents the views of Phaedo of Elis instead of
PlatoÕs views. Apart from the fact that the speciŽ c issue of the immortal-
ity of the soul is not fully confronted by PhaedoÕs arguments (Phaedo on
this score is only an Iolaus, or at best a Heracles in need of IolausÕ help),
I have been careful to leave open the possibility that we might, in the

the beginning of his narration. For he tells us that he was, on the day of SocratesÕ
execution, in the grip of a ÔmixtureÕ of emotions. The word for ÔmixtureÕ is krsiw
(59a5), the standard word in medical contexts for the ÔtemperamentÕ of the body – i.e.
the particular blend of corporeal elements or parts which underlies a given physio-
logical or pathological state (already in the Hippocratic corpus e.g. Nature of Man 4).
The very same word is used later on in the dialogue as a synonym for the ÔharmonyÕ
of corporeal elements which is said by the harmony-theorists to constitute the soul:
oämai ¦gvge . . . krsin eänai kaÜ rmonÛan aétÇn toætvn t¯n cux¯n ²mÇn (Phaedo
86b; cf. d2). PhaedoÕs ÔmixtureÕ of emotions might well have had a basis in his phys-
iological ÔharmonyÕ.
36
And perhaps it was Phaedo who originally spoke of the soulÕs ÔpuriŽ cationÕ
(k‹yarsiw) in much the terms used by Socrates passim in the Phaedo, and of its ÔcureÕ
in the terms implied by SocratesÕ dying wish to have a cock sacriŽ ced to Asclepius.
In any case, we have the evidence of Julian (as cited above) that Phaedo believed in
the curative and purifying power of philosopher (oéd¢n ŽnÛaton eänai t» filosofÛ&,
p‹ntaw d¢ ¤k p‹ntvn êpƒ aét°w kayaÛresyai bÛvn). This is not, by the way, to say
that either Phaedo or Plato saw death as the puriŽ cation or cure of the soul (which is
MostÕs objection to reading the last words of Socrates as a reference to his own ÔcureÕ:
Ô ÒA Cock for AsclepiusÓÕ, 100-4). The point, more generally, is about the freeing of
oneÕs rationality from service to the body – something for which death might itself
stand as a metaphor.
 PHAEDO OF ELIS AND PLATO ON THE SOUL 23

manner of a signiŽ cant number of commentators, read the psychology of

the Phaedo as an exploration of PlatoÕs own ÔstandardÕ psychological
model from a particular point of view. What I do hope to have done, how-
ever, is to have given that point of view a name and a context and, in
doing so, to show that the Phaedo can be used as further evidence for the
views of Phaedo of Elis himself.

University of Durham