You are on page 1of 25

Review: Chaos and Control: Reading Plato's "Politicus

Author(s): Mary Margaret McCabe
Reviewed work(s):
Reading the Statesman, Proceedings of the III Symposium Platonicum by C. J. Rowe
Plato's Statesman: Selected Papers from the Third Symposium Platonicum by Peter
Nicholson ; C. J. Rowe
Plato: Statesman by Plato; C. J. Rowe
Source: Phronesis, Vol. 42, No. 1 (1997), pp. 94-117
Published by: BRILL
Stable URL:
Accessed: 24/09/2008 10:17

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the
scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that
promotes the discovery and use of these resources. For more information about JSTOR, please contact

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

Chaos and Control:ReadingPlato's Politicus'


On firstreading,the Politicus appearsa dismaldialogue(compared,for example,
to the immediacyof both the philosophyand the dramaof the Theaetetus).This
conversationbetweenthe Eleatic Strangerand the hopelesslycomplaisantYoung
Socratesseems unlikelyto captureour imagination;the lengthydiscussionof col-
lection and division may do little for our understanding
of dialectic;and even the
joke (at 266c, a pun on being a pig and coming last which is marginallymore
amusingin Greek)will leave us cold. It may be hardlysurprisingthat"thisweary
dialogue,"as GilbertRyle called it, has been left alone by scholars.
Howevera recentSymposiumPlatonicumhas revivedinterestin the Politicus;2
this generatedtwo volumes of papersgiven at the Symposiumand, more impor-
tantly,a new translationwith commentaryby ChristopherRowe.3The new OCT,
moreover,gives a freshly edited text.4This materialmakes it immediatelyclear
that the Politicus should not be dismissedout of hand- even althoughit stands
revealed as an extremelycomplex composition,both from the literaryand from
the philosophicalpoint of view.

I ChristopherRowe, Bob Sharples and Tad Brennanwere kind enough to read and
criticise a draft of these comments; I am very grateful to them.
2 Old habits die hard; I prefer Politicus (Plt.) to Statesnan, not least to avoid the

dangers of archaism (and the impossibility of capturing an extinct species) in the
English expression. In deference to Plato, however, I use the expression "the states-
man" (ratherthan "the politician")to describe the person with political understanding.
3 Reading the Statesman, Proceedings of the Ill Symposium Platonicum, ed. C.J.

Rowe, Sankt Augustin: Academia Verlag 1995. p. 421. DM 98. ISBN 3-88345-634-9
[hereafterRS]; Plato's Statesman: Selected Papers from the Third SymposiumPlato-
nicum, eds. Peter Nicholson and C.J. Rowe, Polis Volume 12, 1993. p. 220. ISSN
0412-257X [hereafterPSSP]; Plato: Statesman, with translationand commentaryby
C.J. Rowe, Aris and Phillips, Warminster, 1995. pp. vi + 248. ?35/$49.95 hb;
?14.95/$24.95 pb. ISBN 0-85668-612-3 hb; 0-85668-613-1 pb. [hereafter Rowe]. I
have eschewed a detailed summaryof each of the papers in the collections, not least
because the editors provide helpful introductions.
I In RS Nicoll comments on the manuscripttradition,and the OCT editor, Robinson,

? KoninklijkeBrill, Leiden, 1997 Phronesis XL/III

The rich disputes which arise in these three volumes are - inevitably - of
a rathermixed character.Rowe's commentaryis, as we should expect, complex
and extremelyscrupulous;it is also, in several respects,thoroughlyprovocative
(notablyin his heterodoxreadingof the myth and in his challengeto a straight-
forwardreadingof the political theory of the dialogue).The paperscollected in
the two volumes are, in some cases, new readingsof individualpassages, and in
other cases synoptic views of the dialogue as a whole (PSSP claims to be pri-
marily interestedin political matters;althoughthat brief is read with a generous
eye). That, of course, is what we mightexpect to issue from a huge International
Congress; and many of the individual papers are both valuable and exciting,
notablywherethey invite us to reconsiderour dustyold views aboutthe Politicus.
Huge InternationalCongressesproducea multiplicityof opinions and interpreta-
tions; they are not always, however,easy to digest - and the wealth of detail to
be found in these volumes is not always such as to producethe synopticview of
the dialogue we need for thoroughreappraisal.In consideringsome of the more
importantcontributionsof these volumes to our understandingof the Politicus, I
shall ask two questionsof a synopticsort:what is the Politicus about?And how
does the dialogue hang together?5
You mightthinkthatthe answerto the firstquestionis obvious- this dialogue,
the second in the trilogy which began with the Sophist and should end with the
Philosopher(an unwrittendialogue),6is aboutthe statesman(is the answerto the

discusses the changes made to the Burnet edition, in particulartwo changes which
affect the myth (at 269e4 and 271d4); see furtherbelow.
I There is a furtherquestion of just how this dialogue is to be aligned with the other
dialogues of the late period; and how it fits with the Republic and the Laws. Against
Owen's radical view of the Politicus (e.g. in 'The Place of the Timaeus in Plato's
Dialogues," in Logic, Science and Dialectic, 65-84) see here Kahn's largely unitarian
account of the place of the Plt. in the Platonic corpus as a whole, RS, and Gill's mod-
erate view of Plato's development, RS (discussed furtherbelow). On the place of the
dialogue in the rest of the corpus, Palumbo considers the relation between the Sophist
and Plt., RS, while several other contributorsdiscuss the relation between this and
other political dialogues. In PSSP there are several papers on the relation between this
dialogue and the dialogues about Socrates' execution: these papers too tend to be uni-
tarian(or else to ignore the possibility that in composing late dialogues about Socrates
Plato is engaging on something more complex and reflective than merely giving us
another chapter in the Socrates story).
6 The importanceor otherwise of the Philosopher, and the significance of the frame

narrativeto our understandingof the Plt. as a whole is little considered by the contrib-
utors to these volumes; even by Rowe. E.g. Arends, PSSP, takes it as simply obvious
that the philosopher is not discussed here. An exception is Ferber,RS, who considers
the question of the Philosopher in the light of the unwrittendoctrines and the views
of the Tubingen School; hence the cryptic reference to "the precise truthitself" 284d2.
I shall argue that Plt. is not so much a coded allusion to doctrines but ratheran invi-
tation to speculate on the activity of philosophising itself.

same questionfor the Sophistso obvious?).7The structureof the dialogue,then,
is dictatedby its subjectmatter- it advancestowardsthe final definitionat 31lb
which amplifies the conclusion of 305e: ". . . the one that controls all of these [sc.
subsidiarykinds of expertise],and the laws, and cares for every aspect of things
on the city, and weaves everythingtogetherin the most correctway - this. . . we
would. . . appropriately It does so, however,in a convoluted
call statesmanship."8
way, to say the least. It opens with a framediscussionof the interlocutorsand of
the topic of the dialogue (257a-258b)and the closing remarkmay returnto that
frame. There follows a laboriousdivision of the statesman(258b-268b), inter-
spersedwith a commentaryon their methodand its mistakes(260b; 261e-264b;
265a-b;266b-e; 267a; 267c-d; 268d-e).9Thenthe Eleatic Strangeroffersan elab-
orate cosmologicalmyth (268d-274e), again with a commentaryon just how the
mythrelatesto the purposein hand(268d-e;269c; 272d; 273e; 274b; 274e).10The
ES now insists thatthis makesclearjust wherethe firstdivisionwent wrong;this
develops into a complex accountof how we come to know (274e-278e)." The
seconddivisionensues (279a-283b),in which the ES promisesto revealthe states-
man by offering a model: weaving.'2 This culminates in a puzzle about method
onceagain(283b),andthena discussionof thenatureof "duemeasure"(so WpIov)13

I Dixsaut, RS, points out that even this account of the ostensible purpose of the
Plt. may need modification if the final speech of the dialogue is spoken by Young
Socrates and not - as Robinson/Rowe tentatively have it, by Old Socrates. If the last
speech is by OS, the dialogue is closed, by a remark surprisinglybereft of Socratic
irony. Dixsaut prefers YS, on the grounds that this makes the apparentnaivete of the
final remark obvious. For reasons that will appear below, I think the subtle Dixsaut
version is preferable,not least because it maintains the ratherindeterminatestatus of
the final definition.
8 305e2-6. I use Rowe's translationsthroughout.
9 Here see especially Chiesa, Cavini and de Pinotti in RS; also Fattal in PSSP.
10 Here see especially Brisson, Dillon and Ferrariin RS, with detailed comments in
Rowe; also Carone, Naddaf and Steiner in PSSP.
" This passage itself reflects on the first division, indirectly. The first division is
conducted in a rather high-handedway by the ES with YS cantering along behind.
The present section discusses coming to know: and invites us to reflect on two things:
how did the ES know what divisions to make in the first place? And has his making
them resulted in YS knowing anything at all? The characterisationof the interlocutors
is vital to the irony of this point - or else it is reflects the failure of the dialogue as
a whole. On this section see Kato, RS.
12 In the sequel, of course, the model turns into an element of the definition of the
statesman;some part of the work done here in the text is on the difference between
the use of an analogy and the use of an image. xapa6eibyga,extensively discussed at
277dff., is here turnedinto a term of art. Is a napLpa6&cyT a transcendentform? Surely
not: Kato RS is sensible on this, as is Rowe, Introduction.On this passage, of course,
see Owen, "Plato on the Undepictable,"Logic Science and Dialectic 138-147.
13 Lafrance'sessay on this passage, RS, rightly, I think, argues against the view that
here the ES outlines two sorts of measuringskill, roughly approximateto the two sorts

(283c-287a) - here at the centre of the dialogue, the ES once again turns his atten-
tion to method: the object of the exercise (285d, 287a) is to make the participants
better dialecticians. With this in mind he embarks on the third division,'4 which
allows the statesman to be separated off from the pretenders (287b-291c).5 Then
there is a long discussion of forms of government (29ld-303d).'6 Finally the dia-
logue closes with an account of the nature of the statesman's expertise (as the
overarching science) and his ability to weave together the disparate virtues in the
state (303d-31 1c).'7
One account of this sequence, then, is that it is designed simply to secure the
final definition; the remarks about method which pervade the dialogue are the
means to that end and nothing else. Now, however, the longueurs of some of
the divisions become almost intolerable; no explanation is given of how the point
of all this is that the ES and the YS may become better dialecticians (285d; 287a);
and the relation between this dialogue and the Republic becomes acutely prob-
lematic. In the Republic, after all, the king is the philosopher. In this dialogue,
apparently, the philosopher is to be the subject of a separate study.
But is that appearance a reality? Did Plato intend to write the Philosopher, but
never get round to it (maybe he was busy with the Laws)? Or does the Politicus
give us a view of the philosopher which supplants its separate discussion? After
all, the interlocutors spend a great deal of their time both discussing philosophical

of mathematics described at Philebus 56-7. The contrast he offers, however, between
measuringquantityand measuringquality tends to obscure the teleological component
of this passage, central as it is to the dialogue as a whole; and thus to miss is con-
nection with the closing pages of the Philebus (esp. 64dff.). Ferber suggests that if to
gxptov describes a judgement we make about concrete particulars,then it is not the
same in each case, and cannot be exact. This implies, he suggests, that there is a form
of the exact. I do not find this implication in the text (indeed, if this is an unwritten
doctrine, then I wouldn't, would I?); instead, I suggest that the account of good judge-
ment, of which the discussion of ro glrptov is a part, is central to the new episte-
mology of this dialogue.
14 Rowe objects to me in correspondencethat this analysis of three separate divi-

sions is disputable (on his view, we have the same division throughout,with each
stage modifying, but not rejecting, its predecessor). On his account the statesman is
still a herder at the end of the dialogue, albeit "of a special kind";on my account the
discussions of self-determinationand political structuresuggest that the statesman of
our era is not a herdsman(or a fatherfigure),but rathercloser to an equal, notablywhen
it comes to dialectical discussion. His knowledge may be super-ordinate;but he him-
self is not an autocrat,nor is his authoritydefined in terms of those whom he controls.
' Cf. here Brickhouse and Smith in PSSP.
16 See here Lane, Gill in RS, discussed below.
17 Dixsaut (RS) argues that here the function of the statesman is to be distinguished

from the function of the educator. Her paper is otherwise excellent; but this aspect of
her conclusions does not convince me; nor does it allow her to drive a wedge suc-
cessfully between the Politicus and the Republic. See also Bobonich's detailed analy-
sis of the last section of the dialogue, RS.

methodand doing it. The final definitionof the statesman,moreover,both repu-
diates standardpolitical notions (e.g. 292c-d suggests that the definitionof the
king does not dependon questionsof the size of the citizenry,nor on theircon-
sent;nor on the distributionof wealthamongthem,but only on knowledge)"8 and
fits well the Republic'saccountof the sole importanceof knowledgefor explain-
ing both virtueand the rightto politicaloffice. If the statesmanis the personwho
knows, he will be the philosopher(on this account,the Politicusfits the Republic,
but is at odds with the legislativeprogrammeof the Laws). But if the Philosopher
was never to have been written,since this dialogue has alreadysaid it all, then
the Politicus makes its point as it proceeds,and not just as the resultof the final
successful division.'9The discussionsof method, then, would be neitherinstru-
mentalnor tangentialto the task in hand;on the contrary,they are a partof the
accountof the philosopherwhich is given indirectlyherein.But in thatcase a dif-
ferentquestionarises:if the Politicusmerelyresketchesthe philosopher-king, why
shouldPlato botherto write it? If the Politicus comes too close to the Republic,
it looks redundant(and even duller than Ryle thought- what price the Eleatic
Strangerand Young Socrateswhen we could have Socrateshimself?).
I shall suggest two areasin which the Politicustransformsour (Republicstyle)
understanding of what it is to be a philosopher:in its teleology;and in its episte-
mology. As a consequence,I shall argue,the philosopheris now eminentlysuited
to being a king; althoughkings may no longer aspireto being philosophers.
The teleologyof the Politicusseems to beginwith the myth.A0 The mythappears
here early in the dialogue (as in the Phaedrus, not as in the Gorgias or the
Phaedoor the Republic); it is long (maybe overlong, as the ES pointsout at 277a),
complex and allusive (at least to the Timaeus2' and to the work of some pre-
Socratics).Y1 Whatexactly does it show, for the purposesof the discussionof the
statesman?Or - if its connectionwith the discussionof the statesmanshows us
only a small partof the point of the myth - what other purposeof the dialogue
does it fill?

But see below; and Clark, RS, Toney PSSP.

would make sense of the last speech being delivered by YS, who - ex
19 This
hypothesi - would not (yet) have seen this indirect purpose of the dialogue.
20 The discussionof the mythprovokessome importantwork in these volumes, notably
Brisson's fresh account of his re-reading of the myth, RS, and Rowe's similar inter-
pretationof the cosmic cycle; Dillon's paper, RS, on neo-Platonic readings of myth,
which asks just how myths should be interpreted;and Ferrari's suggestion, subtly
argued in RS, on the traditionalaccount of the cosmology of the myth to explain the
way that man is a technological animal.
21 Forwardor backward?The orthodoxy here is that the Timaeus is later than Plt.;
cf. e.g. Kahn, RS. Rowe has a more robust attitudeto chronology;and I remain stoutly
unconvincedthat these mattersare areas where we can know more than Plato intended
us to see. It is, of course, dangerousto suppose that, where we find the argumentsor
the theories of the Plt. unclear, the Timaeus is transparent.
At least to Heraclitus and Empedocles, cf. below.

At 268c the ES is dissatisfiedwith the division which suggeststhat the king is
the "sole herdsmanand rearerof the humanherd"becausethey have troublewith
the specification"sole":after all, there are many pretendersto this title, and it is
not obvious thatonly the king has a rightfulclaim. The myth is, in the firstplace,
designed to show just how we should separateoff the king from the pretenders.
This task is later completed in two phases - at 287b-291c, where the king is
defined negatively, as doing what the pretendersfail to do (he is avteicvra=ct-
Ico; - as I shall argue,"self-determining"); and then at 292b-294c, where the ES
first explains that the true king is so by virtueof knowledge;and then marksoff
the equitablerule by such a king fromrule by the laws. The mythis also, it seems,
to show how the king is not a shepherdafterall (275a).23 But then it is not imme-
diately clear why a detailed cosmology would be needed to explain either the
autonomyand the knowledgeof the king, or his differencefrom shepherds(the
Republicmanageswithout).So maybe the myth has some further,separatepur-
pose - for example, a discussionof divine teleology, or a seriousexplanationof
the structureof the cosmos.
The strangerrecalls threemythicalphenomena:the reversalof the sun and the
stars at the time of the quarrelbetween Atreus and Thyestes; the golden age
ruled by Kronos;and the race of those who were once born from the earth and
not by reproductionfromeach other.They are all, he says, explainedby the same
This universethe god himselfsometimesaccompanies, guidingit on its way and
helpingit move in a circle,while at othertimeshe lets it go, when its circuits
have completedthe measureof time allottedto it, and of its own accordit
revolvesbackwardsin the oppositedirection,beinga living creatureandhaving
had intelligenceassignedto it by the one who fittedit togetherin the beginning.
This reversalof the cosmos is the centralexplanationof the mechanismsof the
myth, and it is argued,in the first place, on metaphysicalprinciple.21 Something
with body cannot remain"permanentlyin the same state and condition,"nor be
"permanentlythe same";26the universehas body, so it cannot be fixed. On the
other hand, since it is the creationof god, it should be as close to uniformity
as possible; this is why "it has reverse rotationas its lot, which is the smallest

But see Rowe's objection to me, above, n. 14.
nOoq - Rowe has "state of affairs";the sequel allows this to be, not so much
that the same singular event caused all three phenomena at once, but ratherthat each
of them when it occurred, is to be explained by a single (type of) cause.
25 269d-e is remarkablylike some of the arguments of Metaphysics A.6; compare
the argumentat Phaedrus 245c-e for a similar style.
26 Here the ES uses language familiar from the theory of forms; I shall discuss below
whether we should just map this material onto middle period metaphysics. See here
Bravo's sensible account of how the ontology of the Pit. is related to the discussions
of limit and unlimited in the Philebus, and how both relate to collection and division.

possiblevariationof its movement,"269e3-4). Any variationin its circularmove-
ment shouldbe explained,the ES suggests,as simplyas possible:hence the alter-
nationbetween being moved by god and moving underits own power (simpler,
the ES says, than postulatingtwo gods to move it in differentdirections:this
methodologyof simplicityis vital to our understandingof what follows).
The reverserotationexplains a series of disparatephenomena:periodsor mo-
ments of destructivechaos (270c-d); the reversalof humangrowth, so that the
old become young and eventuallydisappear(270d-e);27and the fact that people
were once born from the earth,where now they reproducesexually (271a-c).3 It
also - somehow- explains the contrastbetween two eras. In the golden age of
Cronos,god ruled the universeand minordivinitiesits parts;then therewas no
pain or toil, no savageryor violence. Man and animal lived peacefullytogether,
pasturedby the gods on food that sprang spontaneouslyfrom the earth (271d-
272b); in this era, the gods were shepherdsof men. The era in which we now
live, however, is (said to be) the age of Zeus (and it is to describethis that the
myth is being told, 272d5ff.).For god "let go of the steeringoars and retiredto
the observationpost";29this caused a tremor,and chaos; and then the universe
"set itself in order, into the accustomedcourse that belongs to it, itself taking
charge of and masteringboth the things within it and itself, because it remem-
beredso far as it could the teachingof its craftsmanand father"(273a-b).In this
era, the universestartsout quite well, for it has good things from its creator;but
from itself it gets what is bad and unjust;and as the universeforgets more and
more the lesson it learnedfrom god, so the bad increasesuntil disasterlooms.

27 271a: there is an account here of how those who died in the chaos then also age

in reverse, very quickly dissolving into nothing. This tale may be pessimistic or opti-
mistic, depending on your view: pessimistic if the thought is that disappearinginto
nothing is insult upon the injury of dying by violence, optimistic if in this disappear-
ance all the evidence of violence disappears. The overall question to be asked, as I
shall insist, is whether this kind of growing backwardsand disappearingis held to be
a particularlyunpleasantsort of death (as Brisson and Rowe assume) or a denial of
death altogether.
28 At 271c2 there is a suggestion that some are exempt from this rebirthwhen god
takes them off "to another destiny." Does this refer to the escape of the philosopher
from the cycle of reincarnation?If it does, should it not be what happens in this era?
Both Brisson and Rowe make something of this: but there is no clear implicationhere
that being taken off to some other destiny is a reward.
29 It may make a difference what exactly is involved in retiring to the observation
post. One thoughtmight be that this is a point of vantage outside the ship; but Brisson,
RS 357 n. 29 suggests that it is a lookout on the ship: on his account even when the
universe is left to itself god has not abandoned the universe. Even so there is a
significant contrasthere between the control of god when he is at the oars and his rel-
ative helplessness when he is at the observation post; I shall suggest furtherthat this
theme of control and independenceis exactly the theme, not only of the myth, but of
the dialogue as a whole.

Thus over time its control deteriorates, until eventually god, anxious lest his crea-
tion should break into pieces,30 takes up his position again at the steering oars
(273e), and the whole cycle restarts.3' Once the cosmos has reversed into its pre-
sent course, then other things change too. Ageing runs from young to old, growth
from small to large, birth leads to death (273e). And now it is no longer pos-
sible for creatures to grow from the earth "under the agency of others' putting
it together" (274a3); instead human reproduction imitates the self-movement of
the cosmos.32 The crucial point is this (274b). Human beings, in this era, are less
fortunate than their predecessors because they lack both the care of the god and
the nurture he gave us. The beasts are now wild and competitive, and humans
defenceless,33 so that they were at first in great trouble. At this point they received
gifts from the gods - fire, technology and resources for agriculture, and the abil-
ity to teach and learn: these provided the foundations of human society, wherein
humans take care for themselves in exactly the same way as the cosmos does.
And to see that was the point of the myth; as well as to show how they went
wrong in supposing the king to be the shepherd of people (274dff.).34

3 Here the description of the difficulties in which the universe finds itself has some

affinity with the dangers of philosophy: it is in aporia and at risk of falling into the
"infinite/indefinitesea of unlikeness."
31 At 273e3 there is an echo of the metaphysical argumentof 269dff.; by virtue of
the cycle's recurrence,the universe is eternal;here, however, immortalityis conferred
on it by god - being immortal, ratherthan recurrent,is a divine prerogative,and can
only come from god. The dangers of inconsistency here are not, I think, to be pressed
too hard; but they do tell in favour of an interpretationof the myth which treats its
chronological features as allegorical, cf. Dillon RS, Carone, PSSP.
32 The expression here is a1Yxoiparopa Elvac.I shall later argue that this is directly

related to the later thought that the statesman has, where others lack, self-determina-
tion or autonomy. Indeed the Greek expression av&o1cpa&op expresses better the notion
of independencewhich I take to be at issue, than does the expression a tcvo'Vog,which
has ratherthe sense of self-legislating. CompareThuc. 4.63, 6.72, Laws 875b and Pit.
299c for ai6rocpa&tpand Thuc. 2.63, 5.33 for av6ovogto;;LSJ gives the meaning "of
one's own free will" for aurovopo; at Antigone 821; this, I think, is not quite the point.
In the Politicus, moreover, autonomy is defined not in terms of free will, but rather
in terms of independence from the mechanisms of causation. Finally, the notion of
being self-legislating is not really applicable to the statesman of this dialogue, who is
above fixed laws. But there is no English equivalent of avroicpa6cp (autocracy is
something quite different) so I prefer "self-determination."
33 Several commentatorspoint out the similaritybetween this account and the Great

Myth of the Protagoras; see esp. Narcy in RS. The purpose of such a deliberate echo
is less easy to discern. Perhaps the least we are meant to notice is the difference
between the gifts of the gods: for Plato (cf. also Philebus 16c) the gods give us gifts
which are practical or intellectual; for Protagorasthey give us social virtues. In the
sequel, I suggest below, Plato supposes that the practice of intellectual virtue has a
necessary social component.
34 This account of the origins of human intelligence, if the myth is taken literally,

The traditionalview of this myth is that there is a two phase cycle, where the
cosmos rotatesnow one way and now in reverse.In the first phase, the reign of
Cronos,thingscome from the earthand grow younger;in the reign of Zeus they
do the opposite,coming from the intercourseof two parents,and dying back into
the ground.Brisson, and after him Rowe, have objectedto this accountof the
cosmic cycle, mostly on the groundsthat this means that the presentcosmos, so
far from imitatingthe benevolentrule of god, actively opposes it; Brisson adds
that this makes a nonsense of the claim that the present era is also divinely
ordered."Moreover,as Rowe pointsout, it generatesa hopelesslypessimisticpic-
ture of the world now, when thingsjust wind down from bad to worse, and the
only hope of salvationis Doomsday;thereis not accountthen to be given of how
the world behaves in a way that fits its being a rationalcreature.Whatis more,
the traditionalview does not explain the chaos which apparentlybegins the sec-
ond era;36for in thatera the deteriorationseems to be gradualfrom a fairly good
beginning;how is that consistentwith its initiationby destructivechaos?37
So on the Brisson view, there are three periods, the age of Cronos,the uni-
verse left to itself and the age of Zeus. In the age of Cronosthe universeis ruled
by god and its partsby the minordivinities(so that in that era the divinitiesare
shepherdsof men); in the time when the universe is left to itself all divine
influencehas retreated;then in the age of Zeus god returnsto the helm, although

impliesthatthereare humans- in the age of Cronos- who may not be intelligent
(so muchfor a theoryof recollection;see here Katoand Pricein RS). Such a con-
clusionmay tell againsta literalreadingof the myth;or for the view thatthe happi-
ness of exercisingone's intellectis not merelyto be explainedin termsof beinggod's
product.Cf. Naddaf PSSP.
35 It is, as the ES says at 272b2, the era "whichthey say is in the time of Zeus";
and it is an era in which thereare divinegifts (274c5, "thegifts fromthe gods, of
whichwe haveancientreports"; cf. Philebus 16c5on collectionanddivisionas a gifts
fromthe gods). However,neitherof these claimsis unequivocal- they may merely
reportthatpeoplesay this is the age of Zeus,etc., ratherthancommittingthe ES to
this being (literally) true; cf. Erler RS 377, FerrariRS 394, n. 17.
36 The traditionalistmight say that it does: the chaos is explained by the fact of the
reversal, 270cd.
37 Both Brisson and Rowe offer problems with the detail of the myth on the tradi-

tional view, none of which seem to me decisive against a plain reading of the myth,
according to the myth's own canons of simplicity; e.g., Brisson argues that the march
of the ages and sexual reproduction must be separated - for reasons that remain
unclear to me, except that both Brisson and Rowe seem to think there are two differ-
ent ways of being earth-born(a good one, being sown in the earth, and a bad one, as
result of what Rowe calls the "traumaof reversal,"in the former of which there is no
sexual reproduction,but ageing goes in the same directionas in our era); Rowe claims
that "moving in the opposite direction"refers to two differentdirections at 269c and
270b (that "in the opposite direction"should be treated as a relative expression in a
dialogue which is good at relative terms, cf. 283, is hardly surprising,maybe).

the minor divinities are no longer closely involved in the life of man.38Rowe
agrees with Brisson'sgeneralprinciplethat the cosmos now shouldbe moving in
the same directionas the cosmos underCronos;so he supposesthat the periods
of movement in the reverse directionare brief catastropheswhich occur (as a
result of inertialmotionof the mechanismof the universe)just when god lets go
of the universe.Thus in an entirecycle, again thereare threephases:the universe
moves first undergod's guidance,then in the reversedirectionas a result of its
corporealnature,then back again in the same directionas the first era, although
this time as a result of the intelligentactivity of the world. Here the movement
is at first ordered,but later deteriorating;eventuallygod takes over again.
Now these constructionsof the motion of the cosmos need not suppose that
the cosmic myth is taken literally, as a sequencethat actually occurs in time.39
However both Brisson and Rowe seem to suppose that they are literal in some
way - these are descriptionsof the cosmos now, which allow us to understand
our own universein teleological terms. This modifiedliteralism,of course, may
not be the right way to read Platonic myths: they may offer more or less alle-
gorical accountsof man's condition;' althoughthey do requireus to supposethat
Plato somehow or othermeans his myths to representthe truth.Nonetheless,the
Brisson/Roweaccount does place some limitationon the (often uncontrollable)
business of interpretingallegory; and it does have some sort of continuitywith
the cosmologies we find elsewhere- notablyin the Timaeus.But it turnsa sim-
ple story (the traditional- one motion forwards,one back) into a complex one
(two motions forward,at least one back); so accordingto the canons invited by
the metaphysicalargumentwhich begins the myth, this complexityhad betterbe
justified. What is more, this account of the myth needs to make sense of its
explicit purpose,namely to modify the definitionof the statesmanas a shepherd
of men and by markinghim off from the pretenders.

38 How, now, does Brisson explains the gifts fromthe gods in this era, whichwere

one of the problems he found with the traditionalview? Brisson argues for this detail
by meansof Dies' readingof 271d, & vbv, cat icaxa t6'xouqTaucov roii,o which,he
argues,proposesan era wheregod is at the helmbut the minordivinitiesare absent:
andthe only erathatcouldbe is theeraof Zeus.Robinson,RS,p. 45, retainstheread-
ing of Campbell 8'a' caKraX T6o'ou; rav6ov ToVXroon the grounds that this fits the
traditionalview; he is followedby Rowe,who does not use Brisson'sdetail(thatin
the era of Zeus god is at the helm, but the divinitiesare absent),but supposesthat
this alludesto the previousera. None of these readingsseems entirelysatisfactory;
andeach demandsa morefundamental understanding of the teleology.
39 I have profitedherefromChristopher Rowe's objectionsto my firstview of his
40 See here Dillon's illuminatingaccountof neo-Platonist of the
myth,RS, or other,more generallyallegorical,views, e.g. CaronePSSP. Naddaf,
PSSP, offersthe followingcarelessargument:since Platodid not meanthe mythto
be taken literally, it follows that he never believed that there was a golden age, or a
periodicreversal,or earth-born

Both Rowe and Brissonrely on the thoughtthatthe presentuniversecannotbe
workingin the oppositedirectionfrom the universeorderedby god, becausethat
would make a nonsenseof any teleologicalclaims that may be made for this era
now. Teleology only lapses, on their view, in the interimperiodsof chaos, nei-
ther in the age of Cronos nor now. This, of course, is to make two separate
assumptions:first, that teleology (for Plato, or at all) can only be understoodin
termsof divine ordering;and second thatPlato supposesthatthe presentuniverse,
as a whole, is explicable as a reflectionof a time of perfectand divine order.4'
The second assumptionmight be thoughtwarranted- by a (traditionalist)read-
ing of the Timaeus.42 But is the first?
In the Phaedo, of course, Socratesrejectshis predecessors'accountsof "why
each thing becomes and why it perishesand why it is" (96a), on the groundsthat
while they tell of the stuffsand the causes of becomingand perishingand being,
they fail to explainhow everything"is as it is best for it to be placed"(99b), nor
how "the good and the necessarytie everythingtogether"(99c). Socratesis not
arguinghere thathis predecessors'mechanisticand materialistaccountsare false;
butjust thatthey are inadequate.The same story seems to be told in the Politicus
myth, which is markedlyreminiscentof both the words and the theoriesof the
Presocratics.43But after a mechanisticdescriptionof the cosmos," YS asks just
how the accountof the cosmic cycles fits with the worldhe is in now (271c). The
ES proceedsto describethe reign of Cronosin termsof the care thatgod and the

1' Carone, PSSP, argues that chaos, even intermittentchaos, could never prevail in
a Platonic universe;consequently,she suggests, the myth cannotbe understoodliterally.
42 Here chronology may matter:the consensus of these volumes is that the Timaeus

is late, and a seriously meant, Platonic account of cosmology. Owen's early dating of
the Timaeus is now given little credence (although this seems too easy: one thought
might be that the Timaeus may be later than Owen thought, but even so not so late
as to post-date this dialogue - how we place the Timaeus should depend on what we
think the Timaeus is doing); but there needs to be deeper considerationof just how
the arguments of the Timaeus might fit the purposes of the critical period (it is not
obvious that those purposes must be served by a straightforward,Platonic, cosmol-
ogy). In any case, the interpretationsof the Politicus and of the Timaeus may be
thought to supporteach other; in that case, there needs to be an independentaccount
of what is happening in this dialogue first.
43 Compare e.g. Heraclitus Fr. 31 and 270blO; or the strange claims about god
restoring immortalityto the cosmos at 273e3 with the view of HeraclitusFrs. 30, 36
that fire is etemal, everliving and everdying;or the theme of measure in Heraclitus30
echoed both at 269c6 and then again at 283cff.; likewise Anaximanderon the "ordi-
nance of time," Fr. 1, echoed at 273e3; or Empedocles Fr. 17 with the account of the
double cycle, throughoutthe myth; or Empedocles 62 on the earth born creatures.
Empedocles - as is widely acknowledged - is the main antecedentof the cosmology;
what is less widely acknowledged is how the myth thus constitutes an attack on its
4 Even the initial mention of the age of Cronos is silent on its being a golden age,

divinities have for their charges. In this period, the good care of the gods is
matched by the automatic appearance of the good things of the earth;45given the
abundance and the pleasantness46of the things produced by the earth, the gods'
task is to order everything well.
That was the state of affairs under Cronos, and we know what things are like
now. So, the ES (suddenly) asks, which life47 is the happier? When YS is unable
to answer, the ES's answer is famously surprising and opaque.48
... if, with so much leisure available to them, and so much possibility of their
being able to get together in conversation not only with human beings but also
with animals - if the nurslings of Cronos used all these advantages to do phi-
losophy . . . the judgement is easy, that those who lived then were far more for-
tunate than those who live now; but if they spent their time gorging themselves
with food and drink and exchanging stories with each other and with the ani-
mals of the sort that even now are told about them, this too, if I may reveal how
it seems to me, at least, is a matter that is easily judged. (272b-c)
The ES may mean that the men of the golden age would have been blissful doing
philosophy; even without it they are pretty fortunate compared to us. Or he may
be suggesting that nothing about the easy life of the golden age can make up for
not doing any philosophy.49The rhetoric of the passage tells in favour of the claim
that whoever does philosophy is better off than whoever does not.-I But then how

aS cx&o6aro; at 271dl, 27M1e,272a5, 274c2 is paralleled by a,&r&apc; used of the
divinities at 271d7. There is, I think, only one passage which suggests directly that
the produce of the earth is itself a productof god's benevolence, namely 273b7; other-
wise, there is a fairly general contrastbetween the raw materialsof life, and the order
imposed on it by everyone else.
46 There is, of course, already some sort of teleological claim in the description of
the raw material in the age of Cronos, and its difference from the difficult physical
circumstancesof the age of Zeus. However there is not, apart from the general claim
made in the passage in my previous note, any explicit claim that the materialis already
divinely ordered,before the gods do their pasturingof human beings.
47 It is, I take it, vital to the development of the point of the myth that this ques-

tion about the happy life is asked over and over again in Plato; and that it is posed,
not as a question about a period of time, or an era, but as a question about the lived
lives of individuals.
I Cf. e.g. Erler, RS, Carone PSSP, Steiner, PSSP.
49 Tad Brennan suggests a stoicising version to me: the good things of life, as they

are representedby the golden age, are indifferentto happiness.
so This interpretationis, I think, reinforcedby the language of judgement: if we are
supposed to judge of two which is the happier; the choice is easy, the ES suggests,
both in the case of the golden age philosophersand in the case of the golden age glut-
tons. If the second case is still fortunate,even though not as fortunateas the first, the
ease of the judgement would, surely, need qualification.It is, I take it, significant that
in this dialogue - as in the Philebus - the theme of making a judgement about lives
is matched by discussion of the importanceof judgement in a life.

exactly is that either relevantto or explainedby the myth of the cosmos? And
why is this questionaboutwhat life is best to live posed so abruptlyin the middle
of a myth to modify the definitionof the statesman?5'
If the mythof the cosmos gives us, as Brisson/Rowehave it, two phaseswhich
run in the same direction- the age of Cronos,where the runningof the cosmos
is smooth and pleasant;and the age of Zeus where it is more fraughtand dan-
gerous52- then life now is a poor imitationof life then. Then the choice of lives
is obvious:pick Cronosany time, and with any qualification.If, contrariwise,the
choice of lives tells us that philosophymakes all the differenceto which era we
shouldprefer,then theremay be somethingto be said for the life underZeus after
all. But in that case, is the mythexplainedby havingthe era of Zeus a poor imi-
tation of the era of Cronos?Or by having the age of Zeus run in reverseto its
golden predecessor?
Brisson/Rowe,I suggested,suppose that teleology is either the same thing as
divine creation,or an imitationof it. In the Phaedo, two otherteleologicalclaims
appeared:one an appealto intention("Whyis Socratesstaying in prison?""Be-
cause he decidedto.") and one an appealto order("Showme that the earthis in
the middle because it is best so")." Each of these is distinctfrom, if compatible
with,5'a theistic teleology (consider,for example, that ordinaryhumanpractical
reasoningis teleologicaleven if there is no god; or that naturalecology may be
in itself worthmaintaining,even if the presentbalanceitself evolved by virtueof
antecedentcauses,antecedentcauses which may not includethe divine).The order
of the cosmos underCronosis a good thing eitherbecausegod made it thatway,
or because it is well-ordered.But once god has let go of the cosmos, the order
startsto disintegrateand be threatened.If the secondera is one of decayingtele-
ology, what sort of life could we call happy?And undercircumstanceslike that,
whom shouldwe ask to rule us? If, on the Brisson/Roweaccount,the secondera

s' We should bear in mind that the life worth living is recalled at 300a in an aston-
ishingly independentremarkby YS, that the life in a state with an embargo on inquiry
would not be worth living.
52 Hard to see, of course, why that should be so on Brisson's account, when god is

still in charge even if the divinities are absent.
53 Compare here the discussions e.g. at Grg. 504ff. of the well-orderedsoul, which

is worth having just because it is well-ordered, not because god made it that way.
I4 Compatible,that is to say, under some circumstances.If, however, a theistic tele-
ology fully explains or determineseverything that happens, it may be hard to see how
an independent account of practical reasoning could be maintained (the Stoics, of
course, had troublewith this one). And if one cites the value of a naturalbalance, the
claim that it is created may be inimical to the very notion of balance itself (not least,
for example, if the balance is one which changes in a stable way over time). I leave
out of considerationhere a functionalistteleology (such as I take Aristotle's generally
to be) which looks to the proper functioning of each species independentlyof their
place within the whole system. There are, of course passages where Aristotle seems
more inclined to an ecological view; perhapsPhys. 11.8, and also Pol. 1256b15.

imitatesthe divine, why shouldtherebe any questionaboutindividuallives? And
why would there be any dispute about the natureof the king in such an era -
surely he should imitatehis divine predecessors,and be a shepherdof men?
Considerthree points of comparisonbetween the two eras: ageing, remember-
ing and reproduction.In the age of Cronos,on the traditionalview, the process
of ageing goes in reverse (270d-271c; 273e-274a);55so people emerge from the
earth old, and grow young.56So death no longer holds sway: the fear of dying,
when ageing is reversed,is done away with. The Phaedo, of course,frees us from
the fear of dying by philosophy;here the people of Cronoswill become younger
and younger,more and more naive and childlike.They may be going in the right
directionfor the facile assumptionsof modem ageism; but it is the wrong direc-
tion for the purposesof philosophising,which needs time and experience.57
On any view when the eras are reversed,there remainbut dim memoriesof
what has gone before (271b; 273b); moreover,if life is lived backwards,there is
no place for memory.58This might be a good thing if we are only interestedin
the pleasuresof the moment;but we need memory,as well as the otherprocesses
of reason,for structuringa life (cf. Philebus21c). Indeedthe Theaetetussuggests
that memoryis a conditionfor being who we are (e.g. 166aff.);on that account
the creaturesof the golden age are ephemera,rightlyinterestedonly in eating and
drinking,since no otheractivity is availableto them.
On any view, the myth contraststhe way people are born:from the earth in

ss Cf. n. 36 above; both Rowe and Brisson suggest that there are two phases of
earth-bornpeople, one of which ages from young to old, the other in reverse. Rowe
treats the emergence of grey-beards as a sign of decaying teleology (as in Hesiod).
But here the greyness disappears as the creaturesgrow younger. Likewise those who
die by violence in the age of Zeus come alive again and then vanish quickly - so that
there is nothing that remains of the murder.That seems to me to be the opposite of
decaying teleology.
"6 There is, I suspect, a telling parallel here with the odd argument of the Par-
menides that when we grow older we grow older than our younger selves. Rowe is
troubled by this, noting an apparentlack of fit between the idea that souls are seeds
at 272e3 and the birth of grey-beardsat 273el0. The interpretationof the grey-beards
as an Hesiodic nightmare,however, is itself to beg the question about the nature of
the teleological claims offered here; and Plato is, of course, quite able to turn or even
to reverse an old story to his own advantage.
S7 Unlike myths, of course: compare the ES' exhortation to the extreme youth of
YS at 268e with his complaint in the Sophist (243aff.) that ancient myths treat their
hearers as children - that is, they patronise them.
S8 Cf. Ferrarion this; compare 270e with the increasing forgetfulness of the uni-
verse in our era, 273b. Tad Brennan objects to me that the thought that memory is
incompatiblewith ageing backwardsdoes not take account of how we remember sto-
ries from the previous era (271a); but here it is people in our era who rememberthen,
not the people then who rememberus.

one era and from sexual intercoursein the other.59This reflectsthe ways of life
in the differenteras. When man lives on what natureproducesautomatically,he
has no need for social intercourseeither,60since he is safe from dangerand pro-
tected by the benevolentdaimones.In the exigent circumstancesof the age of
Zeus, by contrast,humanbeings need each other, both to reproduceand to stay
alive. It is only in the age of Zeus thatotherpeople matterat all. Of course,Plato
often uses sexual intercourseand reproductionas a metaphorfor philosophical
activity. Here, however,there is a direct and literal parallelbetweensexual and
social intercourseand philosophicaldiscussion;for each activity requiresother
people as partners6'(this is denied for the age of Cronos at 27le; and the gifts
of the gods to the men of this era include teaching and education, 274c).62
Philosophicaldiscussion,moreover,requireseach participantto be able to remem-
ber what has been said and to preserveconsistencyand for each to be able to
interactwith the other.For the backwardscreaturesof the golden age, this would
be impossible.63 So can we judge which life would be happier?
In the Theaetetus Socrates described the life of philosophy: "It is not possible,
Theodorus,for bad to perish;for there must always be somethingwhich is over
against good; nor is it possible for bad to take up residence among the gods;
instead it necessarily frequents mortal nature and this place here. So we should
try to flee this place and go there as quickly as possible. And flight is likening to
god as much as possible; and that likening is becoming just and holy with rea-
son" (176a). Likening to god is not the same thing as becoming the product of
god's good order. The two eras of the myth allow a contrast between the well-
run cosmos and the disordered one; and they allow a comparison between the
ordering activity of god in the age of Cronos and the activity of the philosopher

s9 Cf. n. 37 above. Brisson and Rowe must argue that there is no direct parallel
between the reverse process of ageing and the means of reproduction,because they
both suppose that ageing backwards,which implies being born grey-haired,is decay-
ing teleology. My suggestion is that this is only true when we think about worth in
terms of doing philosophy; in ordinaryterms there would be a great deal to be said
for ageing backwards.Experta dico.
60 The "need for talk" in the golden age is left as an open question at 272d5; but

the starting point of the myth is that in the golden age men had no needs at all.
61 The partnershipof philosophical discussion is parodied,I suggest, in the discus-
sion between the ES and YS; here the opening discussion of the dialogue makes some
suggestions about just what conditions are necessary for a good discussion: it ain't
enough to be called Socrates.
62 Cf. above n. 33. This myth leaves out what Protagorassuggested were the gifts
of the gods: ai&k and ticrq.I suggest that this is because this myth argues that tak-
ing others as partnersis a consequence of the circumstancesof the age of Zeus - that
is, they are explained by the myth, not by the deus ex machina device of gifts from
the gods.
63 It is hard to see, indeed, how they could lay claim to a life at all; they are like
the molluscs of the Philebus which feel pleasure but never rememberor plan.

in the age of Zeus. It is not for the philosopherto become a functioningpartof
the well-runcosmos, or to join the decay of the cosmos as it runs down; but to
imitate god. What sets god apart(as it sets Reason apartin the Timaeus)is his
controlover the work of necessity.What sets the philosopherapart,I suggest, is
his ability to orderhis world independentlyof the chaos of its mechanismas it
runs down. A large partof the Politicus is designedto explain this claim for the
philosopherin our era: firstly to explain his self-determination; and secondly to
explainjust how his orderingreflectsthe activities of god.
The ES suggeststhatthereare two mistakesin the early stages of the division:
the mistake of supposingthat the statesmanis a shepherd;and the mistake of
being unable to markhim off from the pretendersto his title. How is are these
mistakesrectifiedby the myth?
If the statesmanis a shepherd,he is wronglyassimilatedto god, and his charges
to a herd;' and if he is a shepherd,thereis no clear accountof just how he should
rule over a whole state (cf. e.g. 275a3; the "whole humancommunitytogether,"
276b). This mistakeis, it appears,systematic(and more than the errorof treat-
ing a mortal as a god). One thought,which comes easily to the modem mind,
mightbe thatthe importantdifferencelies in the consentof his subjects;65 by virtue
of the difficultiesof this era, the only way that the statesmancan go about his
business is by rulingover free men, who agree to his rule. However,the matter
of politicalconsent,or freedom,is, it seems, not partof the definitionof the states-
man (cf. e.g. 293aff.);' so why should it matterthat the statesman'ssubjectsare
not a herd? This may be an issue, not so much of political liberty, but of the
integrityand separatenessof the other partnersin the state, which makes them
suitable interlocutors,either for making political or philosophicalarrangements.
Sheep and cattle just feed and enjoy themselves- like the backwardscreatures
of Cronos, they never end up "inquiringfrom all kinds of creatureswhether
any one of them had some capacity of its own that enabled it to see better in
some way than the rest with respect to the gatheringof wisdom" (272c).67It is

' Again, this may reflect on whether the two eras are similar, or contrary;for my
money here the mistake is explained by their contrariety;cf. 274e11.
65 Clark, RS.
66 The title of Clark's suggestive paper, RS, suggests that he has misread this pas-
sage ("Herdsof free bipeds");but the content of the paper shows that he takes Plato's
point. Toney, PSSP, reads 293a, via Skemp's translation,as saying that it is indiffer-
ent to the statesman whether the citizens consent to his rule. Rowe's translation,on
the other hand, suggests that it is indifferent to the question of defining the states-
man whether we consider the consent of the citizens; the preceding speech of the ES,
293a2-4, suggests that his topic is the search for the statesman, not the statesman's
view of his citizens. Rowe seems to be right; the whole dialogue is concemed with
what contrasts are significant for division; so here, the matter of consent (as in the
case of doctors) is tangential to our understandingof their expertise.
67 The theme of inquiry will recur; see below and compare the inquiry which the

people of Cronos should do, 272c3, with the inquiry that is actually being done in the

the ability to cooperate- sexually, socially and intellectually- which marksoff
the humanbeings of this era from the herds of Cronos;it is this kind of parity
between the membersof the state which is implied by the need to form states
underthe difficultcircumstancesof the age of Zeus."gAs a consequencethe states-
man and the subjectsof the era of Zeus are like each other where god and the
nurslingsof Cronos are unlike. For the statesmanand his subjects share both
natureand education(275c).69
Despite the appearanceof 276e, then, the questionof politicallibertyand con-
sent is not, I suggest, a crucial one in the political theoryof the Politicus. But
there is a question of freedomhere - namely the freedomof the men of Zeus
from their own circumstances.Will the workingof the universetake them over,
or are they to assertthemselvesagainstit? This is not so mucha problemof free
will as one for teleology: againstthe (sometimesdecaying)constraintsof neces-
sity, can man pursuehis own gooef In the markingoff of the pretendersto states-
manshipfrom the true statesmanthe ES makes clear that it is the statesman's
pursuitof his own purposeswhich is vital to our understandingof who should
rule. First he returnsto the stage of the division which suggestedthat the states-
man is "self-directing."This led them to suppose that the statesmanis a herd-
rearer;and then let open the gates to the pretenders(275d). So how is the wedge
to be drivenbetweenaireixtaxTrt and herd-rearing?
The passage where the claims of the pretendersto be a statesmanare repudi-
ated startsunpromisinglyenough,with a discussionof weaving (279a-283).This
allows the ES to maketwo points,vital to rejectingthe pretenders:one aboutcau-
sation,the otheraboutexpertise.If we may distinguishbetweenthe end to which
some object is the means("thesesocks are to keep me warm"),we may also dis-
tinguishbetweenthe skills which providethe means and those which controlthe
end. Likewise,we may distinguishbetween skills which are deployedto achieve
the ends of others(the skills, for example,of the heraldor the diviner)",and those

dialogueitself, 275c6-7;comparethe ES' insistenceon discussionratherthanpaint-
ing with a broadbrush,277a-c.
61 At 271e8-272a3the absenceof both families and constitutions in the age of
Cronosseems to have two explanations: that peoplewere bornfrom the earth,and
thattheyremembered nothingof whatwentbefore.The firstmayexplainthe absence
so the secondmustbe the
of families,butis irrelevantto the absenceof constitutions,
accountof why therewereno constitutions. Comparethe descriptionof the gifts from
the gods in the age of Zeus - fire,technologyand agriculture; the gods do not give
the social or politicalvirtuesas Protagorasimaginedit. The ES replacespolitical
virtueswith "an indispensable requirement for teachingand education"(274c6):the
politicalvirtuesarereplacedby intellectualones.Onceagain,thisreflectson thebusi-
ness of philosophicaldiscussion:Zeus' peoplehave the necessarycapacitiesfor it;
Cronos'peopledo not.
69 Arendsmisleadingly,in my view, characterises politicalrelationsin termsof
power, PSSP.
Cf. hereBrickhouseandSmithin PSSPwho, reflectingon Vlastos,askjust how

which determine their own ends - it is only the person who determines his own
ends who may count - the argument suggests - as "self-directing." Such a per-
son should also be distinguished from the chimerical figure of the sophist, who
has no self to direct (291c). But the self-determination of the statesman is then
explained by virtue of three features of his expertise: firstly (294ff.) that his knowl-
edge is reflected in judgements which are accurate for the moment and for the
individual case; secondly, that his knowledge is systematic - this is what enables
him to bring the community together (306aff.); thirdly that his knowledge, because
it is complete, enables him to delegate: he is not himself engaged in political
activity, but instead on advising others what to do (310e).7'
What, then, is the political theory of the dialogue? How far is the Politicus
about politics at all? Suppose your doctor suddenly became malevolent, or your
pilot ill-willed; suppose that all the experts, in a sudden revolution, became evil
self-seekers.72We might try to defend ourselves against them by establishing rules
they must not break, or by setting ourselves up to determine what should be the
right treatment, which the best voyage. We might (in this parody of the emer-
gence from the state of nature) even establish democratic institutions; furthermore,
we would - to protect ourselves further - decree that no-one is to do any research
or experiments, on pain of the extreme penalty. This would - as YS is startled
into pointing out - make life not worth living. Worse still, perhaps," would be

the diviner could have a skill at all, where in the Apology they are castigated for their
ignorance. They conclude that Plt. allows that "divinationcan be the source of impor-
tant moral truths."It is, I think, a mistake to try to map isolated argumentsof the Plt.
directly on to the claims of earlier dialogues; Plato's interest in the issue of technai
is neither single nor simple - and at Plt. 279ff. he is worried about the autonomy of
the diviner, not his truths.
71 Cf. the claim in advance at 292b, that this expertise will be "concernedin a sense

with makingjudgements and controlling,"which anticipatesthe first and the thirdfea-
tures of statesmanship.The careful discussion of individualjudgement at 294ff. leads
in to the crisis of the constitutions, 300a.
72 298ff. Both Lane, RS, and Gill, RS, discuss this thought-experiment.Lane sup-

poses that it is designed to make us see just how dangerous is the inflexibility of the
rule of law; and how it may lead to the kind of repression described at 300a. Gill, by
contrast, sees the experiment as "bizarre,"if it supposes that there should be commu-
nities which forbid the practice of skills or science; but Athens in 399 may have been
one such.
73 Our understandingof this passage, 299eff., depends on whether we suppose that

in the thought-experimentthe rule of law is established on the basis of some exper-
tise; or on the basis of no good evidence whatever. Gill, for example, seems to take
300blff. to offer some mild recommendationof the constitutions, that they are based
on experience and tradition (cf. e.g. Skemp/Ostwald,"the fruit of long experience").
Rowe insists that at 300blff., the notion of Eitpais not approvingbut at best neutral,
at worst pejorative from a Platonic perspective ("on the basis of much experiment").
Rowe thus takes the second voyage pessimistically. Gill, by contrast, supposes that

the case where someone ignoresboth the expertsand the rule of law; then chaos
will follow.
The issue here seems to be, in part,the purposeof the thought-experiment. Is
it designedas a palliative,a (back-handed)justificationof law? In that case, per-
haps, the emphasislies in the way the experimentconstruesthe expertsas being
capableof malevolence,a returnto the old questionof how far an expertin jus-
tice might be the most competentat doing wrong; alternatively(and consonant
with the earlier claim that the questionof consent is immaterialto the central
definitionof the statesman)it is designedto show just how damaginga general
mistrustof expertise may be (hence the familiar rhetoricof dissatisfactionat
298a2ff.).'4Or is it designedto pointto just how radicalthe claims for the states-
man are in this dialogue?Is the Politicus endorsingconstitutionalism,or reject-
ing it?
Gill argues that the answer to this questionis to be found in the dialectical
strategyof the dialogue,291-303.75The ES firstunderminesYS's unthinkingcom-
mitmentto traditionalviews ("constitutionsare, roughlyspeaking,a good thing,
especiallywhen thereare tyrantsabout")by suggestingthat if the statesmandoes
have objectiveknowledge,then no constitutionhas a claim againsthim. Afterthe
thoughtexperiment,however, the ES suggests that there are indeed groundsfor
endorsingsome kind of constitutionalism.Whereknowledgeis absent,constitu-
tions are all we have; they are "copies"of the ideal.76
Lane77arguesthatthe politicaltheoryof the Politicusis radicallydifferentfrom
the Republic.Firstof all, the art of the statesmanis definedin termsof his grasp
of the xatpo', the rightmoment(his art is, that is to say, irretrievablyparticular
and immediate);for this reason he practisesjudgement;and the particularityof

there is something to be saved for the rule of law: it may be inflexible, but it is based
on some kind of principle:the second voyage is second-best.
If we accept the teleology of the central passage of the dialogue, knowledge
knows the right measure, and so producesjudgements which are directed at the best;
it is, of course, always a question whether Plato really ever shows that this implies
that the knower must have a good conscience.
7S Gill persuasively analyses the entire section as a piece of dialectic; and he argues

that the citizens of the fixed constitutionsallow there to be objective political knowl-
edge; but prevent its discovery in their institutions.This still - in my view - makes
these fixed constitutions very grim candidates for "second-best":especially in a dia-
logue which suggests that even the easy life of the golden age would be dismal com-
pared to the life of the philosopher.
76 Lane points to how differently"imitate"is used here from elsewhere in Plato: for

here it is a central feature of the imitation to remain fixed. So this sort of imitation,
at 300e, misses what is central to the knowledge it imitates, namely its sensitivity to
each situation; such an imitation is bound, therefore, to fail. Likewise, Rowe argues
that this fixity is itself inimical to the best life (the same point, perhaps,is made about
the written word at Phaedrus 276ff.).
1 In an excellent paper, RS.

his judgementsis not to be enshrinedin writtenlaw. But this kind of judgement
does not give rise to authoritarianism; instead,the task of the statesmanis to teach
the warringpartsof the state to work together.Here again, the Politicus and the
Republicpartcompany.The Republichas a staticview of the perfectstate,where
conflictingpartsare kept segregated.The Politicus, by contrast,sees the conflict
of differentvirtues (belligerenceand pacifism, for example) as a necessaryand
dynamicfeatureof the state, whose smoothworking(by means of the education
of the citizens) is the productof the statesman'sart. The true statesman,on this
view, is no autocrat:insteadhe is someone who enables the citizens to integrate
and govern themselves:78 "to do real politics is to do it yourself."'79
Both Lane and Gill supposethatthereis a final attemptin the dialogueto deal
with the exigenciesof politicallife, with the difficultiesof being in the statewhere
the philosopher-kingis absent.This may be where the Politicus improveson the
Republic:not so much by replacingthe philosopherking, as by allowing thereto
be stable states which are non-ideal.Rowe, on the otherhand, supposesthere to
be an immensegulf between even the best states (which stick to the established
laws) and the philosophicalstate (run by knowledge and good judgement).On
either account,there remainsan issue aboutjust how the political theory of the
dialogue- whetherit is construedas pragmaticor just pessimistic- fits with the
rest of the dialogue.If - as I have suggested- the bulk of the dialogueup to this
point has been concernedto discuss the life of the philosopher,why shouldPlato
now turn to the discussion of these second-bestconstitutions?If YS is right to
suggest that life in these fixed constitutionsis unliveable(even if, hyperbolically,
life in states with no law at all is worse), how can they be second-best?Con-
versely, how can the philosophersurvive in any state, if constitutionsare the
enemy of inquiry?Whatdoes any of this politicaltheoryhave to do with collec-
tion and division?80
If knowinghow to rule is like knowinghow to weave, then the politicalexpert
will know how to integratethe cloth. We might easily see how the skills of
rhetoricmight bring that about (by, for example,persuadingthe subjectsto con-
sent to be ruled);but that is not at issue here. So what sort of knowledgewould
unify the state?8'The dialogueinsists thatknowledgeis suppliedby (definedby?)

78 This fits the emphasis,in the centralpassages of the dialogue, on self-determination.
9 This is Lane's cash-value for the weaving metaphorin politics. Like Dixsaut and
Arends, she rightly insists on the way it brings unity to the state.
10 Kahn, for example, suggests that collection and division is the difference between
the Plt. and the Rep.: that the dialogues do not differ in their metaphysicalassumptions,
only in their account of philosophical method: ".... in the latest period Plato never
departedfrom his fundamentaldichotomy between being and becoming, between the
invariantreality of the Forms and the changing realm of the visible world ... But the
practice of dialectic looks quite different in Plato's later works, once the method of
hypothesis is replaced by the proceduresof Collection and Division." RS, p. 55.
81 The closing pages of the dialogue m-iight lead us to suppose that the statesman

the methodof collectionand division:which is, afterall, a methodof fully articu-
lating and makingcoherenta subjectmatter.This is given severalexpansionsin
the methodologicaldigressionsof the dialogue- first where haphazarddivisions
are distinguishedfrom naturalones (262bff.);and second wherethe ES comes to
discuss measure and the art of measurement.82 Here the preliminarycontrast
between"whatis measured"relativelyto somethingelse - large relativeto small,
fat relativeto thin - and what is "in due measure"which allows a value judge-
ment of what is measured83 generatestwo interconnectedaccountsof the art of
measurement:one evaluativelyneutral,the otherteleologicallyexplained.8" Then
these two arts may be seen as vitally connectedto the methodof collectionand
division,just because that methodworks, not only by dividingthings from each
otherrelatively,but by doing so accordingto the naturaljoints - accordingto due
measure- accordingto what is genuinelya partof some ?-8o; (compare263aff.
with 285aff.).But when this happens,the philosopheris able, not just to take the
synopticview of the field he discerns,but also to pick out individualpointsin it:
collectionanddivisionboth systematisesand individuates.The notionof measure-
ment, that is to say, is a featureof the propermethodof collectionand division;
which is itself a systematicexpertise,articulatingwhateveris its subjectmatter.
But in that case two furtherconsequencesbecome clear. Firstly,the methodof
collectionand division,the truemethodof philosophy,just by virtueof its evalua-
tive componentis itself a teleological science. Secondly, such an expertiseread-
ily lends itself to accuratejudgementin individualcases, becausethe very points
of division are determined- or judged - teleologically.So one who had such an
expertisewould be ideally suited to practisethe judgementsof equity which the
ES says are centralto any properaccountof the statesman'sknowledge(294aff.);
and those individualjudgementsareonly properlypractisedby someonewith such
expert knowledge. Only such a person, moreover,will be able to practisethe

needs to be a psychologist;but here again the issue is knowledge,not persuasion.
Dixsaut is sensible here.
82 This passage is, I think, underestimatedby Lafrance,RS.
" We should not too readily suppose that if this gives, against a sophistic rela-
tivism, an objective account of value, that the value so described is also transcendent.
Tordesillas, RS, well documents Plato's critique of the sophists here; he is too ready,
however, to see this in strong unitarianterms - perhapsas a consequence of reading
this material throughneo-Platonising spectacles, e.g. "Le kairos de ce passage corre-
spondraitau Bien de la Republiqueou, si l'on veut considererd'un point de vue stricte-
ment temporel, A l'exaiphnes du Parmenide," p. 107. Compare the more judicious
approachof Lane, RS.
94 Fattal, PSSP, seems ratherdisappointedat the importanceof teleological criteria
for division in Pit.; and his account of collection and division suggests that its objec-
tive is to pick out individual realities understood,I think, as transcendentforms. This
underestimates the importance of the introduction of weaving, which replaces the
piecemeal account of division of the first half of the dialogue with the sort of sys-
tematicity needed for an expertise.

architectonicscience which is demandedof the statesmanin the final page of the
dialogue.85For we should not make the mistakeof thinkingabout collection and
division as a kind of primitivefoundationalism,nor yet as a trivial process of
classification.If collection and division enables the philosopherto define some
particularitem, it does so by giving him systematicunderstanding of a whole sub-
ject. And that systematicunderstandingis provided,not because collection and
division derives the less primitivefrom the more so (in the way demandedby
foundationalism)but becauseit is offers a holistic accountof what knowledgeis.
Knowledge,that is to say, is like weaving, not like constructinga building,foun-
dationsup. In that case, only the knowerhas the kind of systematicgraspof what
matters(understanding which is also teleological)to rule a state by delegating,to
have authorityin the city by virtue of his overarchingscience. That person,we
may now see, is the same personas the one who can do collection and division.
So the statesmanis the philosopher.And for my moneythis claim brooksno com-
promise:only statesrunby the philosopher'sgood judgementwill producea polit-
ical life worth living.
But then if the political theory of the Politicus deploresany but the state run
by the true statesman,is the statesmanthe philosopher-king?How far does this
account - if I am right - merely echo, or polish up, or make practicable and sen-
sible the provisions of the Republic?11Where the Republic gives us the complex
education of the guardians, it could be said, the Politicus gives us a method of
dialectic87 that might even be practised (one day) by the Young Socrates - so the
Politicus is less idealised. The philosopher-kings of the Republic may seem to
lack concem with the minutiae of every day life in the city (their main interest
is in the grandiose schemes of the Forms), where the statesman needs to concern

85 The Republic,on many accounts, says that the philosopher should be the king
because only the philosopher grasps the first principles (the Forms), so that only the
philosopher can be reliably right when it comes to matters of state. The same dia-
logue, however, seems to suggest that the objects of the knowledge of the philosopher
must be clearly distinguished from those of belief, which is the sort of cognition we
can have of the here and now: but then if in the here and now we can only believe,
it is hard to see why the philosopher should be any better off here and now than the
rest of us. If we take even a modified unitarian view of the relation between the
Politicus and the Republic, this two worlds problem is still with us. But the Politicus,
far more than the Republic, is concemed with individual decisions - uncontrovertibly,
the here and now. This may be good reason to reject a unitarianaccount.
I Gill has some sensible things to say on the limitations of comparingon dialogue
with another, given the radically differentpurposes of the Republic, the Politicus and
the Laws. I suggest here, however, that we should do well to contrast the epistemol-
ogy of the Republic and that of the Politicus.
87 There is an old issue here: is Collection and Division reasonably described as a

method? Does it have heuristic value? (cf. the Meno's paradox type puzzle in Kato
RS about how we get an "anticipative grasp of the target";Fattal does not mention
this problem of interpretingCollection and Division.)

himself with individualcases and theirequity,88but here - it could be said - the
Politicusmerelyclarifiesand bringsdown to earththe Republic'sgrandview. And
the philosopherkings of the Republicmay well have put their knowledge into
practice by means of the architectonicdescribed in the closing pages of the
Politicus.So maybeit is truethatthe Republicand the Politicusare indeedbased
on the same- or a roughlycontinuous- metaphysicaldoctrine;the theoryof tran-
This claim may make sense of some of the passagesof the Politicus,and some
of the dialogue's apparentappealto technicalterminology.It does not, however,
fit the epistemologicalstructureof collectionand division (which is holisticrather
thanfoundationalist).It does not, moreover,explainwhat I have arguedto be the
two dominantthemesof the Politicus:the accountof teleology and the best life;
and the accountof the self-determination of the statesman.Nor, if the myth is a
two-cycle one, does it explain why the statesmanis to be found in a periodof
decayingcosmic teleology:it fails to show, therefore,how the purposeof the myth
to rectify the early mistakescould be fulfilled- for on the Republic'spictureof
benevolentdespotism,itself implied by the absoluteknowledgeprovidedby the
Forms,the king does indeedpresideover a golden age. If, contrariwise,the myth
is a three-cycleone, it providesus with no accountof why the statesmanshould
not be a shepherdafterall. So if the metaphysicsof the Politicusmerelyreplicates
the Republic,it fails to give us an integratedview of this strangedialogue- either
of its complex argumentor of its anomalousnarrativesetting.
There is a final difficulty.The myth of the Politicus seems to concem itself
with what makes life worthliving; and the attackon the constitutionsreflectson
whetherconstitutionsmay make life unliveable.In both cases, it is philosophical
inquirywhich makes life worth living. But is collection and division an account
of inquiry?Is this dialogue,dominatedas it is by the tediumof a division done
by an expertin the companyof an ignoramus,a recommendation of the philosoph-
ical life? Here again the overall structureof the dialogueprovidesan answer.We
should not confuse the Politicus' representationof collection and division from
its descriptionof it; nor should we underestimatethe significanceof Socrates'

88 Some, myself included, would argue that any account of the Republic must make

sense of just how the philosophers,who occupy themselves with Forms, can use their
knowledge to rule in the everyday world, which is, on many accounts, only susceptible
to belief.
89 Bravo suggests a returnto immanentforms such as are found in the Socratic dia-
logues. His account of the metaphysics of the Politicus makes it look very much like
(some interpretationsof) the Philebus; but he does not provide an integratedview of
the Politicus itself. Kato is sensibly uncommitted.Of course it is one thing to claim
that collection and division divides objective reality, or that the statesmanhas objec-
tive knowledge (Gill) and quite another to claim that knowledge is of transcendent
forms, or some other degree of reality; to say that something is objective does not
imply its transcendence,or its otherworldliness.

absencefromthe dialogue.Collectionand division is not a textbookclassification.
Insteadit is an epistemologybasedon two features:systematicunderstanding; and
individual good judgement. Those two features are exactly what the Socratic
methodof discussionpromotes;and they are exactly what the ES is tryingto instil
in YS.90That is why this dialogue is as much the Philosopher as it is the
Statesman.In the Republicthe goodness of the best life was explainedin terms
of the view of the forms9'and the influenceof the Formof the Good on the life
of the philosopher(here, more than anywhere,the happinessof the philosophical
life is explained in terms of the philosopherbeing someone who has come to
know, not someonewho is engagedon the difficultbusinessof doing philosophy).
In the Politicus, by contrast,the judgementof the lives in the myth suggeststhat
it is the activity of philosophicaldiscussion which explains happiness. In the
Republic,the self-determinationof the philosophercould be explainedby virtue
of the fact that he was free of the demandsof the physicalworld;and that free-
dom was a consequenceof his knowledge:reasonrules in the philosopher'ssoul,
and all chaos is eliminated.In the Politicus, self-deterninationis imitatinggod;
and imitatinggod is being able to achievecontroldespitethe chaos of the material
world out there;it is being able to see and carryout one's own purposesagainst
the backgroundof the stuffsof the cosmos, whetherthey be orderlyor otherwise.
In that case too, the good life of the philosopheris not explainedby virtueof his
knowledgeof some other,transcendentbeings who guaranteeeternity,purityand
truth;insteadthe good life is a matterof the personwho lives it; and that is best
done by doing philosophy.Doing philosophy,moreover,is no longerdescribedin
termsof emergingfrom the darknessto the light of the forms;insteadit is a mat-
ter of discussion and deliberation,good judgementand awarenessof due meas-
ure: philosophyis what philosophersdo, no longer what they see.

King's College London

90Onceagainthe questionof who speaksthe finalwordsof the dialoguemay here
be significant,especiallyif the finalwordsmiss the point.
91 See herethe salutaryremarksof Kato,RS, p. 163, n. 4.