chapter 4

Are We Nearly There Yet? Eudorus on Aristotle’s

Categories
George Boys-Stones

Introduction
There is no doubting the important history of interaction that is there to be
traced between the Platonic tradition (in its widest construal) and the
Hellenistic schools; or, in particular, the importance that these schools had
as conversation partners with Platonists in the post-Hellenistic period. But in
this chapter I want to make the case for a degree of circumspection in our
approach to understanding how these encounters relate to the (undoubted)
success that Platonism came to enjoy. It is often supposed that the very identity
of Platonism was developed through the post-Hellenistic period as it learned
one way or the other from its rivals; indeed, as I shall show, this assumption is
effectively built into our historiography of the period. But I would like to
suggest that, on the contrary, its identity is already surprisingly well defined in
our earliest evidence, and that changes we might be able to discern as a result of
its subsequent engagement with rival schools may as often be relatively super-
ficial traces of a polemical move against them as a shift in theoretical commit-
ment. (Similarly, there is no good reason to think that the other schools,
including those with roots in the Hellenistic institutions, were any more open
to profound philosophical change than they had been before.)
To illustrate this point, I am going to take as a case-study Eudorus’ critical
account of Aristotle’s Categories. There are several reasons for choosing to
look at this. First and foremost, it is the clearest possible example of a case in
which a Platonist takes on board language from another school, but does so
within an explicitly critical frame – making the point that what it is to be
a Platonist ought to determine the sense to be given to material adopted from
other schools, rather than being determined by it. Secondly, it is an early case
of polemical engagement between Platonism and another school (and, if it is
right to think that Eudorus is the first ‘Platonist’ of the post-Hellenistic age,
then it is the very earliest there is) – which helps to make the point that the

67

Downloaded from https:/www.cambridge.org/core. Columbia University Libraries, on 30 Mar 2017 at 07:40:57, subject to the Cambridge
Core terms of use, available at https:/www.cambridge.org/core/terms. https://doi.org/10.1017/9781316694459.004
 68 george boys-stones
Platonism of the period under consideration in this volume has a robust
adversative identity from the very beginning.1 Finally, it is intended to
question the status of the claim that Stoicism is ‘the’ philosophy at the
nearside of the historical period covered by this volume (‘From
Stoicism. . .’),2 or that Stoicism had a significance above that of other schools
in the development of Platonist thought during the post-Hellenistic period.
The claim obviously fails if it is intended in blandly statistic terms;3 but
neither does it seem accurate to say that Platonists were especially or essen-
tially concerned about Stoicism above other schools (including
Aristotelianism, Epicureanism and the memory of the Sceptical Academy).
What they were concerned about above all was the failure in all of these
schools to see the aetiological importance of the forms; their success
(‘to Platonism’) was success in making the argument against them all.

Philosophy in Transition?
I have suggested that ‘Platonism’ has a fully-formed, adversative identity
when we first catch sight of it in our evidence, with Eudorus in the first
century BCE. But this is not how our histories have typically thought of it:
1
For Eudorus as the first ‘Platonist’, see, e.g., Dörrie 1944; Reale 1989: 313; Bonazzi 2002. It seems to
me bold to infer from the silence of our evidence that he does not have relevant forebears; but it is true
that we do not know of any earlier figure who was recognised as ‘one of their own’ by later self-
describing Platonists (as by Plutarch: cf. n. 22). (This is a criterion which Antiochus of Ascalon
certainly does not meet, by the way: he is mentioned by Platonists only to distance him from them:
see, e.g., Plutarch, Life of Cicero 4.1–2; Numenius fr. 28 des Places.)
2
Engberg-Pedersen 2010b: 1 (‘leading’), 4 (‘the reigning type of philosophy’), etc.
3
Epicureanism, for example, remained at least as popular as Stoicism, and for as long: see Castner 1988
(and cf. O’Meara 1999 for Neoplatonist awareness of Epicureanism); for broader statistical data on
the affiliation of philosophers through the post-Hellenistic period, Goulet 2013. Plutarch wrote more
works specifically against Epicureans (nine: Lamprias 80, 81, 82, 129, 133, 143, 155, 159, 178) than
specifically against the Stoics (seven: Lamprias 59, 77, 78, 79, 149, 152, 154). (Lamprias 148 deals with
Stoics and Epicureans.) And although we know of no specifically anti-Epicurean work by any other
Platonist of the time, we can only be sure of one other specifically anti-Stoic work: Taurus wrote
a work on the self-contradictions of the Stoics (Aulus Gellius, Attic Nights 12.5.5 = 17T Gioè). (His
On Bodies and Incorporeals, T3 Gioè, might also have had the Stoics in view, as most assume of ps.-
Galen, That Qualities are Incorporeal – although Todd (1977) has argued that this latter work is in fact
aimed at Epicureans.) Cf. also Hierax, On Justice, which attacked Stoic and Peripatetic views. There is
a wider spread of evidence for works against Aristotle (aimed, presumably, at contemporary
Aristotelians): in addition to the works on (or anyway containing criticism of) the Categories by
Eudorus, Lucius and Nicostratus, there is Atticus’ work ‘against those who undertake to reconstruct
Plato’s doctrines through Aristotle’, Eubulus’ Aristotle’s Objections to the Republic of Plato and Taurus’
On the Difference of Doctrines between Plato and Aristotle. (Plutarch wrote on Aristotle’s Topics,
Lamprias 56 and a Discussion on the Ten Categories, Lamprias 192; but both works are lost and we
cannot be sure what his intention with them was.) It was presumably not nostalgia that motivated
Marcus Aurelius to endow Chairs in Epicureanism and Aristotelianism in 176, alongside Chairs in
Stoicism and Platonism, but a reflection of the strength of all four movements even as late as that.

Downloaded from https:/www.cambridge.org/core. Columbia University Libraries, on 30 Mar 2017 at 07:40:57, subject to the Cambridge
Core terms of use, available at https:/www.cambridge.org/core/terms. https://doi.org/10.1017/9781316694459.004
 Are We Nearly There Yet? 69
for them, it gradually emerges through what they characterise as centuries of
‘transition’ leading from the end of the Hellenistic era to Plotinus in the
mid-third century CE – a sense of transition which resonates through the
title of this volume (‘from . . . to . . . ’). In one sense, of course, this is
a perfectly benign way of describing the period. It is agreed that the
philosophical landscape looks very different in 80 BCE and in 250 CE:
let ‘transition’ name the sum of relevant changes that took place between
these two dates. But such a description conceals a trap: phrased like this, it
is easy to see that, if we can meaningfully apply the term ‘transitional’ to the
post-Hellenistic era, we ought to be equally happy to apply it to any other
philosophical period which is book-ended by intellectual climates different
from one another. Indeed, given that this sort of book-ending is how
periodisation in our histories are established, it will turn out that any
commonly recognised philosophical ‘period’ is a ‘period of transition’. Yet we
do not typically talk, just for example, of the Classical period as one of
transition (between Pre-Socratic physics and the Hellenistic schools?).
The fact is that ‘transitional’ as a label does not, and cannot, function as
a neutral description. In marking the chronological boundaries of a given
period (from . . . to . . .), it privileges them over what happens in between,
and makes them the standards by which to judge it. Post-Hellenistic
philosophy offers us a good example of the negative effect that this
historiographical framing can have. The ‘transitional’ character of post-
Hellenistic philosophies has for a long time been cashed out in terms of
‘eclecticism’, the idea being that the various philosophers of the period
developed their positions by selecting and recombining elements of earlier
thought (in the first place, Plato and Aristotle, but also then the Hellenistic
schools).4 And although this has not always been meant as a criticism, it
tends to reductivism in any case: systems so described end up as no more
than the sum of those parts to which the term ‘eclectic’ draws our attention;
as if an ‘eclectic’ philosophy is understood when we understand the pre-
existing options from which it was concocted. For this reason, it seems to
me that the reaction against the term ‘eclecticism’ in more recent scholar-
ship allows for no real advance in our understanding of the philosophies to
which it was applied when it replaces it with terms such as ‘syncretism’ or
‘rapprochement’ or ‘absorption’ – which as far as this goes do exactly the
same job.5 To the extent that they describe a process of change which is not

4
See the illuminating account in Catana 2013.
5
‘Syncretism’: e.g. Merlan 1967: 64; Blumenthal 1972: 340; Lilla 1997: 147; Sedley 2003a: 22;
‘rapprochement’: Sterling 1993; Sedley 2005; ‘absorption’: Engberg-Pedersen 2010b: 4; Gritti 2011:

Downloaded from https:/www.cambridge.org/core. Columbia University Libraries, on 30 Mar 2017 at 07:40:57, subject to the Cambridge
Core terms of use, available at https:/www.cambridge.org/core/terms. https://doi.org/10.1017/9781316694459.004
 70 george boys-stones
rooted in philosophical purpose, they are all equally implicated in the
overarching narrative of ‘transition’: they are all ways of describing the
post-Hellenistic period as a bridge between our real points of interest.
In other words: if it is worth saying that post-Hellenistic philosophical
systems are ‘transitional’, it is so because we think that they are only
transitional, and have no intrinsic value to our histories.6 But how did
we come to think this?
The characterisation of post-Hellenistic philosophy as ‘transitional’ is
not, it ought to be emphasised, based on transparent historical data, nor on
self-description, nor even on the perspective of later antiquity. (Even when,
occasionally, later Platonists identify Plotinus as the beginning of a new age
of Platonic exegesis, they do not characterise what went before as
‘transitional’.)7 The idea, rather, can be seen to have its roots in well-
established historiographical prejudices of more recent times. Consider
these remarks from Giovanni Reale’s history of ancient philosophy:
Al medioplatonismo non mancarono uomini di ingegno, ma mancò il genio
creatore o ricreatore, e, appunto per questo, esso restò filosofia di transizione,
a metà del cammino che conduce da Platone a Plotino. (1989: 329, emphasis
mine)8
Thomas Szlezák, again, makes it clear what historians really hope to find,
even as he makes the case that there are other reasons for which the period
might attract our interest:
Doch ist nicht die Epoche als solche, nur weil kein ‘epochemachender’
Denker in ihr auftrat, deswegen schon selbst unwichtig. (2010: 392)9
The Great Man Theory of history is not often so clearly articulated as
here (but Reale and Szlezák have their own axe to grind as leading lights in
the ‘Tübingen-Milan’ view of Plato which makes him the Greatest Man of

212. Cf. terms like ‘generous’ / ‘inclusive’ / ‘fusion’ in Gill 2010: 30–1. (‘Eclecticism’ is still to be
found: e.g. Dillon 1988; Gill 2010: 41–2.)
6
‘Transition’, then, is sometimes linked to the idea of philosophical exile and homelessness (e.g.
Engberg-Pedersen 2010b: 2; I prefer ‘diaspora’, as e.g., Sedley 1997a: 112). The discredited hypothesis
of the ‘School of Gaius’ can be viewed in this light as a game attempt to write ‘Middle’ Platonism into
our histories by finding it institutional structures.
7
Cf., e.g., Proclus, Platonic Theology 1.6.16–21 with Opsomer 2007: 285.
8
‘Middle Platonism did not lack men of intelligence, but it did lack any creative, or re-creative,
genius – and for that very reason it remained a philosophy of transition, midway on the journey that
led from Plato to Plotinus.’ (‘Re-creative’ allows for someone like Plotinus who in Reale’s view is
brilliant but not, with respect to Plato, original.) Nicholas Denyer pointed out to me in conversation
that ‘midway on the journey’ might be a conscious or unconscious allusion to Dante.
9
‘The period is not unimportant in its own right, just because no “epoch-defining” thinkers appeared
during it.’

Downloaded from https:/www.cambridge.org/core. Columbia University Libraries, on 30 Mar 2017 at 07:40:57, subject to the Cambridge
Core terms of use, available at https:/www.cambridge.org/core/terms. https://doi.org/10.1017/9781316694459.004
 Are We Nearly There Yet? 71
all); but it has wide currency among ancient philosophers in an only
slightly modified form: as a normative interest in the institutional school.
I stress ‘institutional school’ because, as A. A. Long notes at the beginning
of his chapter in this volume, the English word ‘school’ has a range of
meanings, shading off at the far end in the innocently descriptive (‘school
of thought’). But the schools on which historians rely most heavily are
precisely those that offer frameworks within which we can identify – in
principle, if not always in practice – a designated authority or spokesman
for a given philosophy. (This is why ‘Great School Theory’ is really
a survival of Great Man Theory: for what we are generally talking about
when we talk about what such a ‘school’ thinks is what some authority
licensed by the school-structure thinks.) The philosophies of the post-
Hellenistic era are none of them ‘schools’ in this sense: even those with
their roots in the Hellenistic institutions no longer had officially designated
spokesmen or the means to identify and appoint them. With no self-
certifying genius (Reale’s ‘genio (ri)creatore’ or Szlezák’s ‘epochemachender
Denker’), but no ‘official’ authorities either, our histories have nowhere to
go: ‘any living philosophical movement, composed of independent minds
unfettered by an official establishment of Guardians of the faith, is’, we are
told, ‘going to be “eclectic”.’10
Great School Theory, then, excludes periods without schools from our
histories, just as Great Man Theory excludes periods without Great Men.
But one might feel inclined, not to accept the conclusion, but to question
the premise; to say that Great School Theory is false to those periods, and
bad history – just as Great Man Theory is false to the centuries in between
(say) Plato and Plotinus. In order to write histories that can encompass it
all, we need to acknowledge that philosophy is sometimes done, heroically,
by Great Men; is more often organised, institutionally, by schools; but may
also be carried on, collectively, by communities working without hierar-
chal structures.
Understanding exactly how philosophical communities of this sort
operate is a question for further research and reflection beyond the limits
of this chapter.11 The more restricted, but key, point on which I want to
focus for its remainder is the demonstration of the fact that, one way or

10
Dillon 1988: 125. Cf. Löhr 2010, arguing that Christianity failed as a philosophy (‘It remained an
unfinished project, a never completed fulfilled ambition’: 187) – because it failed to establish schools
on the Greek model. (The Antonine Chairs of Philosophy – cf. n. 3 – made no difference to this
picture: their holders had more social prestige than philosophical authority.)
11
A constructive starting point might be the notion of the ‘textual community’ recently explored in
this context by Baltzly (2014).

Downloaded from https:/www.cambridge.org/core. Columbia University Libraries, on 30 Mar 2017 at 07:40:57, subject to the Cambridge
Core terms of use, available at https:/www.cambridge.org/core/terms. https://doi.org/10.1017/9781316694459.004
 72 george boys-stones
another, philosophical identity was robustly and constructively maintained
in the post-Hellenistic period; that philosophical systems of the time can
be, and ought to be, treated as the proper study of the history of philoso-
phy, and not transitional phases to stops elsewhere. I am going to do this by
arguing that there are cases which conventional history has viewed as
evidence of dependence, concession and ‘transition’ which are in fact
exactly the opposite. At least sometimes we can see that the adoption of
elements from a rival movement is part of a strategy to reinforce the
difference between them.12 Eudorus’ treatment of Aristotle’s Categories is,
I want to argue, an especially clear example.

Eudorus on the Categories

The Categories of Aristotle is unique in the post-Hellenistic age for having
attracted sustained engagement, perhaps even lemmatic commentary,
from philosophers who at the same time placed themselves at a distance
from the tradition to which it belonged.13 Standard accounts of this interest
are altogether at a loss to explain it. It is not even as if the Categories is an
obvious site for debate about some particular issue, since one of the features
of the interest it attracted is disagreement over what the work was about in
the first place. Aristotelians and Stoics who discuss the work from the first
century BCE onwards consider it to be concerned with issues of language
(perhaps in the context of describing well-formed premises for logical
arguments); Platonists, as we shall see, take it to be a work of ontology.14

12
Whether it was ever anything else seems to me an open question, and one which can only be
answered within the framework of the more responsive historiographical model for which this
chapter is in part an argument.
13
By ‘it’ here I mean the text itself, not the ideas it contains. Part of the Platonist claim will be that the
ideas within the text were already implicit in Plato (see esp. Plutarch and the anonymous commen-
tator as cited in n. 14; also Alcinous, Did. 6.10, 159.43–4). There is no clear evidence for the existence
of dedicated Platonist commentaries, but it has been assumed that this was the context for the critical
work of Eudorus, Lucius and Nicostratus; and cf. Plutarch’s lost Discussion on the Ten
Categories (n. 3).
14
I discuss Eudorus frr. 15 and 17 M[azzarelli 1985] further later in this chapter; see also 18–22 M (which
concern matters relating to the internal consistency of the Categories). Among other Platonists, more
or less explicitly critical observations are preserved from: Lucius and Nicostratus (fragments in Gioè
2002); Plutarch (De proc. an. 1023D-E with n. 22; cf. his lost Discussion on the Ten Categories =
Lamprias 192); and the anonymous commentator on the Theaetetus (P.Berol. 9782, coll.
67.34–68.15). Cf. also Atticus fr. 2.136–8 des Places. Cognate material from the same period is to
be found in Philo (De dec. 30–1), Nicomachus (Intr. Arith. 1.1.3–4) and ps.-Archytas (On Universals
or Categories: 22.5 ff. Thesleff; also cited as On Universal Terms). In general see Moraux 1984; and
especially Griffin 2015.

Downloaded from https:/www.cambridge.org/core. Columbia University Libraries, on 30 Mar 2017 at 07:40:57, subject to the Cambridge
Core terms of use, available at https:/www.cambridge.org/core/terms. https://doi.org/10.1017/9781316694459.004
 Are We Nearly There Yet? 73
So far as this goes, we might seem to be in the presence of, at best,
a disagreement over Aristotle’s intentions – not in itself a philosophical
question, or one that Aristotelian commentators of the time need to feel as
a threat. (Nor is it obvious why Platonists should even have cared what the
Categories was about.) But what gives us some hope that there is
a philosophical issue at the root of it all is a further claim made, or implied,
by the early Platonist commentators on the Categories. They appropriate it
as a work of ontology, but they do this to criticise it as a rather second-rate
work of ontology – at the very least, as one in need of some emendation.
The fact that Platonists pick fights in this way suggests that there is some-
thing of philosophical value at stake. This is not only a disagreement about
Aristotle’s intentions, a ‘simple’ act of appropriation.
Consider the following report of a criticism moved already by Eudorus
(fr. 17 M = Simplicius, On the Categories 206.10–15 Kalbfleisch):
καὶ Εὔδωρος δὲ τῷ περὶ τῆς οὐσίας λόγῳ τὸν περὶ τῆς ποιότητος λόγον καὶ
μετὰ τοῦτον τὸν περὶ τοῦ ποσοῦ συνεζεῦχθαί φησιν· τὴν γὰρ οὐσίαν ἅμα
τῷ ποιῷ καὶ ποσῷ συνυφίστασθαι, μετὰ δὲ ταῦτα τὴν χρονικήν τε καὶ
τοπικὴν κατηγορίαν παραλαμβάνεσθαι· πᾶσαν γὰρ οὐσίαν ποῦ τε εἶναι
καὶ ποτέ, δηλονότι τὴν αἰσθητήν.

Eudorus too15 says that the account of quality and, after this, quantity is
joined onto the account of substance: for a substance is constituted along
with its quality and quantity. After this are taken the categories of time and
place: for every substance – he means perceptible substance, of course – is in
some place and time.

Eudorus’ disagreement with the normal Aristotelian (and, incidentally, the

Stoic) view that the Categories is a work concerned with language is evident
here: he takes the ‘categories’ to be terms of ontological analysis. What is
more, he linked the categories, at least those mentioned in this text, to
a certain kind of substance, namely perceptible substance.16 As far as that

15
I.e. in common with (ps.-)Archytas: see n. 18.
16
Griffin (2015: 88) expresses doubt about the provenance of the qualifying phrase δηλονότι τὴν
αἰσθητήν: ‘it is not obvious that the gloss is Eudorus’, or that Eudorus insisted that the Categories
addressed only sensible οὐσία. He may, for example, have allowed for an intelligible time and place
in which intelligible substance might subsist, as Iamblichus would later propose.’ But (allowing that
the phrase itself may well not be due to Eudorus) it is even less obvious why Simplicius would have
added the restriction unless it accurately represented what Eudorus was thinking: if he were
misrepresenting him (given that Simplicius has no motive to win our sympathies for Eudorus),
this is not an obviously damaging way to do it. (Griffin in general may be too keen to complicate our
evidence for Eudorus in order to square it with what we are told about Pythagorean interest in the
question: see also n. 19.)

Downloaded from https:/www.cambridge.org/core. Columbia University Libraries, on 30 Mar 2017 at 07:40:57, subject to the Cambridge
Core terms of use, available at https:/www.cambridge.org/core/terms. https://doi.org/10.1017/9781316694459.004
 74 george boys-stones
goes, he presumably thinks that the Aristotelians have simply misunder-
stood the intentions of Aristotle.
But there is also criticism of Aristotle himself in this passage. Most
strikingly, it asserts that the categories should be taken in an order different
from that in which Aristotle set them out. Aristotle lists them at Categories
1b25–7 as: substance, quantity, quality, relation, where, when, disposition,
possession, action and affection. Eudorus thinks that the first five, at
least, ought to be: substance, quality, quantity, time and place. In other
words, quantity and quality are transposed, as are where (‘place’) and when
(‘time’);17 and the latter two are promoted above the category of the
‘relative’, which is not mentioned here at all.18
But does any of this matter? How deep does the criticism go? We are
talking about a list after all – and it is not as if Aristotle himself claimed that
the order made any difference. Recent commentators on Eudorus have
tended to assume, therefore, that it does not in fact matter very much.
Eudorus at best is engaged in some tidying up here: shuffling things into an
order that answers to our normal analytical practices, but has no philoso-
phical import of its own.19
This tendency to trivialise Eudorus’ arguments evidently plays to the
assumption that Eudorus (like others of his time) is a ‘transitional’ thinker:
the less these criticisms have substance, the more it will seem obvious that
‘absorption’ or ‘syncretism’ is the point. Yet this seems to me precisely the
reductio ad absurdum of the ‘transitional’ model: for (to read it the other
17
There is of course a further ‘emendation’: in common with ps.-Archytas, Eudorus substitutes ‘time’
and ‘place’ for Aristotle’s ‘when’ and ‘where’. The philosophical motivation for this is not obvious,
although it might be intended to reinforce the ontological rather than the linguistic account of the
categories – considering them as designed to identify the concrete effects of certain causes, rather
than the sort of answer appropriate to particular questions.
18
Commentators, beginning with Simplicius himself, note that ps.-Archytas also adopts a non-
Aristotelian order of the Categories which has in common with Eudorus (1) that it begins substance-
quality-quantity; (2) that it adopts the names ‘time’ and ‘place’ for Aristotle’s ‘when’ and ‘where’;
and (3) that it reverses the order of this latter pair (to read: place-time). But it might be dangerous to
build too much on the back of these similarities (e.g. Theiler 1965: 205, with an inference about
relative date; Tarrant 2008: 592; Griffin 2015). The significance of the first is weakened by the fact
that Aristotle himself quite often uses the order substance-quality-quantity (e.g. An. post. 83a22;
Metaph. Ι, 1068a8–9; Λ, 1069b9–10; EN 1, 1096a24–6; EE 1, 1217b27–8); and there is at least one
divergence between Eudorus and ps.-Archytas which is as significant as the remaining coincidences:
the fact that Eudorus places time and place immediately after quantity. (Cf. on this Dillon 1977: 135.)
19
It is testimony to the grip of the conventional view of Platonic ‘eclecticism’ that commentators can
deny that it is polemical – as, e.g., Griffin 2015: 83, citing with approval Chiaradonna 2013: 47–50 –
even while listing the ways that it ‘criticizes’ (Griffin 2015: 84), ‘corrects’ (cf. 86) and ‘critiques’
(cf. 90) the ‘incomplete account’ (89; cf. ‘inadequate’ at 94) left by an Aristotle who was ‘sometimes
errant’ (97). Cf. also Tarrant 2008: 592 (arguing in relation to fr. 15 M that ‘Eudorus is somebody
with a liking for neat and orderly divisions’); Bonazzi 2013b: 180–1. A rare voice in contrast is Dörrie
1944: 300.

Downloaded from https:/www.cambridge.org/core. Columbia University Libraries, on 30 Mar 2017 at 07:40:57, subject to the Cambridge
Core terms of use, available at https:/www.cambridge.org/core/terms. https://doi.org/10.1017/9781316694459.004
 Are We Nearly There Yet? 75
way round) the more such an approach succeeds in demonstrating ‘syn-
cretism’, the less it allows there to be any philosophical purpose behind it at
all.20 At the limit of this argument, the ‘absorption’ or ‘syncretism’ comes
to exist entirely as an end in itself. The philosophy it describes is no longer
‘in transition’: it is nothing but transition.
What is more, there is some prima facie reason to suppose that there
is more going on here than such an account can encompass. For as soon
as we know that Eudorus is reading the Categories as an ontological
work (remembering that this must be a self-consciously adopted posi-
tion, since the Aristotelians themselves disagreed with it), it ought to
strike us as significant that he has removed the category of the ‘relative’
from the new group of five he has created. The reason that this ought
to strike us as significant is that the term ‘relative’ (πρός τι) also names
one of the two categories in Platonist thought. Plato was, since the days
of the Old Academy, commonly understood to have operated with
a two-category distinction between what exists in itself (the ‘per se’ /
καθ’ αὑτό) and what exists relative to other things (πρός τι).21 With this
in mind, it could start to look very much as if Eudorus’ list is the
beginning of a way of reforming Aristotle’s categories so that they
conform to this older Platonic distinction. The first five categories in
Eudorus’ list – exactly half of the total, with the ‘relative’ pointedly
excluded – might together be intended in some way to fall under the
Platonic per se.
This might seem problematic insofar as it is natural to assume that the
Platonic per se applies, or applies pre-eminently, to the realm of forms, and
we have been told that Eudorus is treating here of sensible substance.
Indeed, our other main testimony to Eudorus’ work on the Categories as
a whole (fr. 15 M = Simplicius, On the Categories 174.14–16 Kalbfleisch)
makes precisely the point that Aristotle does not treat of the per se, and

20
This is especially striking in Griffin, for example, who develops a clear account of the dialectical use
to which a Peripatetic of the time (Andronicus) could put the categories (2015: ch. 2), but has
Eudorus do no more than to ‘map’ them onto a pre-existing ontology (e.g. 75).
21
For Xenocrates, see fr. 12 Heinze; explicit attribution to Plato by Hermodorus ap. Simplicius, On the
Physics 248.2 Diels, and by Aristotle himself according to DL 3.108–9. The status of the ‘relative’
identified in these testimonia is contested: a relative might be, as variously in Aristotle himself,
a property which is relative to the substance whose property it is, or a property which is relative to
some other property (e.g. good vs. bad, or large vs. small in Simplicius’ report of Hermodorus). I am
assuming for Eudorus that ‘relative’ names a property which one (sensible) substance might have in
virtue of its relationship with another (sensible substance). (Compare his remarks on what makes an
action ‘appropriate’, reported at Stobaeus, Ecl. 2.7.2 [44.20–2 W.] = Eudorus fr. 1 M (part): ἃ μέν
ἐστι καθ’ ἑαυτά, ἃ δὲ κατὰ τὴν πρὸς τοὺς πλησίον σχέσιν.)

Downloaded from https:/www.cambridge.org/core. Columbia University Libraries, on 30 Mar 2017 at 07:40:57, subject to the Cambridge
Core terms of use, available at https:/www.cambridge.org/core/terms. https://doi.org/10.1017/9781316694459.004
 76 george boys-stones
might even be taken to imply that all of his categories identify entities
which from a strict Platonic point of view are ‘relative’:22
Αἰτιᾶται δὲ ὁ Εὔδωρος, διὰ τί ἀντιδιῃρημένου τοῦ καθ' αὑτὸ τῷ πρός τι
περὶ μὲν τοῦ πρός τι διείλεκται ὁ Ἀριστοτέλης, περὶ δὲ τοῦ καθ' αὑτὸ
οὐκέτι.

Eudorus raises a problem: why, when there is a division between the per se
and the relative, did Aristotle discuss the relative, but said nothing else about
the per se?
But the claim in 17 M does not have to be that the first five categories are
themselves examples of the per se, only that the per se has some special
relationship with them. For example, all of Eudorus’ first five categories
can be taken to identify things for which a Platonist would identify per se
causes: (perceptible) substance and those determinations (quality, quan-
tity, place and time) which serve to, as it were, concretise it. The remaining
five categories, those omitted in the report of Eudorus in 17 M, by contrast
all deal with relationships into which one perceptible substance can enter
with another, and for which there are no per se causes:23 disposition (is of
something on or around something else),24 possession (is of something
else), action (is on something else), affection (is by something else).
22
Griffin (2015: 90) suggests that the continuation of this passage, which proposes a ‘metaphysical’
reading of the non-substance categories, may itself be from Eudorus. But the formula with which it
is introduced (καὶ ῥητέον ὅτι . . .) is a standard way for the commentator to introduce his own
response to a problem just raised. In other words, the ‘metaphysical’ reading is part of Simplicius’
reply to Eudorus, rather than part of the report of Eudorus’ own views. For parallels in ‘Middle’
Platonism to the idea that everything in the sensible world falls under the ‘relative’, see especially the
anonymous commentator on the Theaetetus at col. 68.1–7. Cf. too Plutarch, Adv. Col. 1115E; and
especially De proc. an. 1023D–E. The latter passage is often taken to suggest that Plutarch applied the
categories to the intelligible world (e.g. Griffin 2015: 91–3), but in fact Plutarch 1023E can easily be
made to fit what I am suggesting for Eudorus in 15 M (allowing, as Griffin and others before him
have noted, that Plutarch may have been consciously following Eudorus, who is an explicit point of
reference for at least some parts of the Generation of the Soul). The context is that Plutarch is trying to
explain how the soul manages both intellection and opinion: the answer is that the soul can say (a)
whether something is the same (i.e. is per se; a form), or (b) whether it is other (sc. ‘relative’?) and then
(b1) it can say in relation to what it is other, and (b2) where and (b3) how . . . all of which, as Plutarch
now notes in passing, gives us an ‘outline’ (ὑπογραφή) of the ten categories. (NB When Plutarch
says that Plato will be clearer on the point ‘in what follows’, the point he has in mind is the soul’s
dual cognitive faculties and not, as often supposed, the ten categories which, as I say, are mentioned
in passing alongside the main point – ἅμα καὶ τῶν δέκα κατηγοριῶν ποιοῦμενος – and about which
Plutarch says nothing else at all.)
23
That is, given the sense of ‘relative’ I am ascribing to Eudorus (n. 21), and given the widespread
assumption among post-Hellenistic Platonists that there are no forms of relatives (e.g. Alc., Did. 9.2,
163.29–30).
24
In my construal, Eudorus would here have to be making a further, radical ‘correction’ of Aristotle’s
account of what makes ‘disposition’ (diathesis) a relative. Aristotle gives the examples of ‘lying’ and
‘sitting’, and says that these very things (lying, sitting) are relative to that which lies or sits (Cat. 6b6

Downloaded from https:/www.cambridge.org/core. Columbia University Libraries, on 30 Mar 2017 at 07:40:57, subject to the Cambridge
Core terms of use, available at https:/www.cambridge.org/core/terms. https://doi.org/10.1017/9781316694459.004
 Are We Nearly There Yet? 77

Aristotle Cat. 1b25–7 Eudorus fr. 17 M Tentative reconstruction of Eudorus’

scheme:
I. per se being, sc. intelligible
substance (15 M)
II. relative being (15 M):
(a) things with per se causes (17 M):
substance substance sensible substance
quantity quality quality
quality quantity quantity
relation time time
where place place
when
(b) relationships between sensible
substances (-):
... relation
disposition ... disposition
possession ... possession
action ... action
affection ... affection

Fig. 1: Eudorus’ ‘correction’ of Aristotle

So much, then, would explain Eudorus’ promotion of these five categories

in 17 M. But it could also explain the transpositions. For these five
categories, in Eudorus’ order, track the hierarchy of Platonic causes as
well. A substance (i.e. a sensible substance) would answer to some form-
paradigm, but its qualities too might be explained by reference to forms; its
quantity might be referred to specifically mathematical forms (which lie
below other forms, at least in the Line image of Resp. 6);25 and these in turn

and 12–14 with 11b10–11). In my reconstruction of Eudorus, ‘lying’ and ‘sitting’ identify a relation
between the thing lying / sitting and that on which it lies / sits. But we know that Eudorus did as
a matter of fact devote special consideration to the analysis of diathesis: frr. 18, 19 and 20 M all
concern it (and cf. 22 M, which deals with thesis). What is more, we know that he construed the very
similar term schesis in just the way I am suggesting (see again fr. 1 M, as quoted in n. 21). Finally,
there is an obvious philosophical advantage to the move I am ascribing to him: it heads off
a notorious problem that modern commentators too have seen with Aristotle’s treatment, namely
that it creates a slippery slope towards saying that everything that belongs to some substance – which
is to say everything falling under nine of the ten categories – is a ‘relative’ (cf. e.g. Ackrill 1963: 99).
25
Alternatively: the form-paradigm explains the palette of perceptible qualities that, along with
quantities, constitute sensible substances (cf. συνυφίστασθαι). (This would better fit my own –
admittedly controversial – view that Platonists before Porphyry did not believe in ‘immanent’ forms,
but discussed sensible substance in terms of what Plotinus calls ‘completive’ qualities,
συμπληρωτικά.)

Downloaded from https:/www.cambridge.org/core. Columbia University Libraries, on 30 Mar 2017 at 07:40:57, subject to the Cambridge
Core terms of use, available at https:/www.cambridge.org/core/terms. https://doi.org/10.1017/9781316694459.004
 78 george boys-stones
give concrete definition to time (the image in the natural world of the
eternity that all these things share: Timaeus 37d), and place (listed below it
perhaps for its closer implication with the receptacle; cf. Timaeus 52a–b).
If the Platonists’ critical engagement with the Categories were as super-
ficial and aimless as the recent literature on it suggests,26 one can only
imagine the baffled amusement it must have provoked among
Aristotelians: the Platonists would have been quibbling with the work for
trivial errors in doing what they knew it was not intended to do at all.
On my reconstruction of Eudorus, however, he has a clear philosophical
purpose in mind. He establishes that the Categories is a flawed attempt at
ontology by showing that one can see it to be what is in effect a degenerate
‘Platonist’ treatise on the subject (one that, for example, ignores intelligible
reality, mislabels some of the categories and shuffles them into an arbitrary
order). But once one sees that, the challenge to Aristotelians contemporary
with Eudorus becomes clear as well. The challenge lies in accounting for
their stance, not as a principled philosophical choice, but as the final
consequence of a witless drift away from the proper understanding of
Plato, a drift which began with Aristotle. The idea that the Categories is
not a work about language but a careless ontological one has no claim on
our philosophical interest in its own right; it is philosophically compelling
only and precisely as a critical explanation of how post-Hellenistic
Aristotelians came to lose sight of intelligible substance altogether (to the
extent indeed that they came to read the Categories itself as nothing more
than a study in words).27 As an act of polemical appropriation, Eudorus’
engagement with the Categories does not blur the division between
Platonist and Aristotelian, but presupposes and affirms it. For this to be
possible, there must be a strong and prior sense of purpose and philosophical
identity in Eudorus – a ‘formal’ identity which is not built up from its
‘material’ appropriations and influences but which, on the contrary, controls
them.

26
Cf. the observation in Chiaradonna 2009b: 108 that, despite the effort of bringing the Aristotelian
categories into their systems, Platonists in fact made little use of them. I take issue, then, with David
Sedley’s suggestion that Eudorus used the Aristotelian scheme ‘alongside’ the Academic, and might
just as well have jettisoned the latter (2002a: 349–50; cf. Bonazzi 2013b: 180–1). Chiaradonna 2009b:
107 talks of ‘intégration subordonnée’ (although he has a something purely formal in mind).
27
The materialist schools tended to claim that ‘forms’ were hypostasisations of linguistic items;
Platonists objected in their turn to the materialists’ reduction of ontology to language. For another
example of the latter move, compare Syrianus’ criticisms of Chrysippus, Archedemus ‘and most
Stoics’ for reducing the forms to linguistic conventions (in Metaph. 105.19–23 Kroll); I see a case of
the former in Seneca, Ep. 65 (Boys-Stones 2013).

Downloaded from https:/www.cambridge.org/core. Columbia University Libraries, on 30 Mar 2017 at 07:40:57, subject to the Cambridge
Core terms of use, available at https:/www.cambridge.org/core/terms. https://doi.org/10.1017/9781316694459.004
 Are We Nearly There Yet? 79
Conclusion
I wanted to argue for circumspection in our assessment of the interactions
between post-Hellenistic schools; in particular, that we should avoid the
temptation to use the evidence of these interactions as a ‘bottom-up’ means
of constructing our understanding of them. The point applies, indeed, to
any ancient school; but the temptation to do so is especially strong in the
post-Hellenistic age because the narrative of ‘transition’ encouraged by
Great School Theory has not left us adequate room for a robust, higher-
level description of its philosophical movements. At the (heavily popu-
lated) extreme, our histories have implied the attrition of Hellenistic school
identity on the one hand (‘from Stoicism’?), and the accumulation of
material into new systems on the other (‘to Platonism’?). The example
I have considered of Eudorus on the Categories shows that, on the contrary,
philosophical interactions continue to be determined by the (top-level)
identity of the schools involved, however in fact that was maintained;
and that this ought to operate as the frame for the assessments we
make – whether or not we conclude that the result of a particular interaction
is a substantial change of position.
To put this another way: post-Hellenistic philosophy can only properly
enter our histories when we see that the systems of the age and their
engagements are no different in principle from those of any other period.
That requires (but also allows) us to understand what they are before
thinking about what they do. From that perspective, the question of how
Platonism came to dominate philosophy by the third century CE is really
not such a difficult one. It comes down to the success of their position on
transcendent causes – a position they shared with the other philosophical
movements that thrived alongside them, including Pythagoreanism and
(especially) Christianity, and which they effectively argued against not only
the Stoics but Epicureans and Aristotelians as well.

Downloaded from https:/www.cambridge.org/core. Columbia University Libraries, on 30 Mar 2017 at 07:40:57, subject to the Cambridge
Core terms of use, available at https:/www.cambridge.org/core/terms. https://doi.org/10.1017/9781316694459.004