Chapter 5 - Do Forms Play the Role of Concepts in Late Plato?

Summary

Lesley Brown’s chapter ‘Do Forms Play the Role of Concepts in Late Plato?’ starts by noting a major issue of controversy concerning the Forms in the Middle Dialogues, namely whether Forms are explanatory properties whose role is to account for why things are the way they are and are therefore the objects of philosophical inquiry and knowledge, or whether Forms are concepts whose role is to explain everyday thinking and discourse. On the assumption that the former option best captures the role of Forms in Plato’s so-called Middle Dialogues, Brown addresses the question whether Plato’s later dialogues manifest a shift in emphasis such that the latter interpretation gains greater prominence. In her view, even though Plato’s later dialogues show increasing interest in matters of language and meaning, and hence may perhaps be taken to show a somewhat greater interest in the role Forms or Kinds play in our everyday thinking and discourse, nonetheless the prominence of the method of division in these works underscores that the Forms are primarily properties discoverable by philosophical inquiry, not everyday concepts or meanings.

1 Introduction: Rival Views About the Role of Forms

Did Plato’s treatment of forms in his later dialogues develop in a way we can characterise as him coming to regard and discuss forms as concepts?

To get a grip on this question, I start from a debate about the role of forms in Plato’s middle dialogues, specifically the Symposium, Republic and Phaedo. The debate I have in mind concerns two ways to understand ‘middle period’ forms.

1. (a) Forms are explanatory properties (hence universals) whose role is both ontological and epistemological. That is, they explain why things are the way they are, and are therefore the objects of philosophical inquiry and knowledge.

2. (b) Forms are (perhaps in addition to the above) entities whose role is to explain everyday thinking and discourse.

Those scholars who favour (b) such as J. L. Ackrill and D. Bostock understand forms in the middle period dialogues as, in effect, concepts, understood as the meanings of general terms. But their views have been criticised by critics such as G. Fine and D. Scott, who insist that Plato intends (a) and not (b). The debate turns in part on how to understand the Phaedo’s discussion of recollection. Proponents of (b) understand Socrates to be arguing, in Phaedo 72e–77a, that recollection explains our ability to apply the general term ‘equal’.¹ Against this reading Scott argues that recollection as described in Phaedo is the province of the philosopher, and that forms in Phaedo and elsewhere are invoked to explain philosophical inquiry and knowledge, not everyday thinking.² In arguing thus, Scott follows Fine in her book On Ideas. There Fine argued that the Phaedo’s theory of recollection does not concern concept acquisition, adding ‘More generally, it does not commit [Plato] to the view that we need to grasp forms in order to understand the meanings of terms. It does not even commit him to the view that forms must exist to confer meaning on general terms. Here as elsewhere Plato ignores meaning and linguistic understanding; his concern is to show how we move from belief to knowledge’.³

One way to characterise the debate about forms in the middle period dialogues is as between (a) Forms as explanatory properties and (b) Forms as concepts and/or as the meanings of words.⁴ However, some scholars regard forms as concepts even if they agree that (a) and not (b) is the correct understanding.⁵ So a better way is to contrast two theses about the role of forms: Role (A) Forms as the objects of philosophical inquiry and knowledge versus Role (B) Forms as playing a key role in everyday thinking and in the understanding of language. In any reading the middle dialogues assign Role A to forms: whatever else they are, they are the objects of philosophical inquiry and knowledge. The debate concerns whether in those dialogues Plato also discusses forms in ways suggesting he assigns them Role B as well.⁶

One preliminary point must be made. It is pretty much universally accepted that forms for Plato are mind-independent entities. Even those espousing the view that assigns forms Role B above – who discern a role for forms in everyday thought and discourse and think of them as concepts – accept that for Plato they are not themselves mental events or states, but the contents of our mental states in a way that also somehow allows them not to be dependent upon minds. Those who hold that the theory of recollection in Phaedo claims that we all innately have the concept of equality (the Role B theorists such as Ackrill) would agree that equality for Plato is both an innate concept and a feature of the real world. To the best of my knowledge there is no clear statement of the distinction between concept and property to be found in Plato, but we do find a tantalisingly brief examination of the question whether forms are thoughts.

In Part I of Parmenides 132b4–5, where Parmenides raises difficulties about forms with Socrates, one suggestion the young Socrates makes is that forms are noēmata, thoughts, existing only in souls.⁷ What is meant by the suggestion (soon to be refuted) that forms are noēmata? The term noēma, like ‘thought’ in English, can be used both for an episode of thinking and for the content of such a thinking. (At De anima 430a28 Aristotle seems to use it for the content of a thinking. It is apparently used for the act of thinking in the poem of Parmenides B7.2 DK.) But what does it mean here? Like Sedley I believe that we can infer from the sequel that here the suggestion is that forms are thoughts in the sense of thinkings.⁸ Parmenides swiftly refutes Socrates’ suggestion that forms are thoughts with an argument that the form must be what the thought is a thought of, that is, its content, not the thought itself. This follow-up shows that the original suggestion is best understood as the claim that forms are thoughts in the sense of mental episodes.⁹ In arguing that forms are the contents of thoughts, not the mental episodes of thinking, Parmenides here leaves open the question what kind of thing the content of a thought is, but Socrates’ next move, allowing that forms ‘are fixed in nature’ (hestanai en tei phusei) shows that he is no longer pursuing any possibility that they are ‘in souls’.

In this essay I proceed on the assumption that, for the middle period dialogues, (A) and not (B) gives the best account of the roles ascribed to forms there.¹⁰ And so my question is: do the later dialogues manifest a shift in emphasis, such that Role B gains more prominence? The Sophist is of particular importance here. Scholars such as Moravcsik and Ackrill argued that forms (also called kinds) in the Sophist should be understood as the meanings of the cognate terms (Moravcsik) or as concepts (Ackrill).¹¹ The view that ascribes to forms in Plato’s later dialogues a key role in everyday thinking and in understanding language relies heavily on the Middle Part of the Sophist (236–267) and in particular on the stretch which explores the five so-called Greatest Kinds (megista genē) and their communion with one another: Being, Sameness, Difference, Change and Stability. But before exploring the Middle Part of the Sophist – the likeliest source for finding forms serving Role B – I turn to another novelty of the later dialogues, the so-called method of division.

2 The Method of Division: An Old Enterprise in New Clothing?

The so-called method of division, described in some detail in Phaedrus but not much practised there, is prominent in the Outer Parts of Sophist and in its sequel the Statesman. In each of these dialogues the main speaker, the Stranger from Elea, conducts a search for a definition of its titular character, the sophist and the statesman. The surprising upshot in Sophist is that the Stranger offers no fewer than seven definitions, each of them said to be of the sophist (rather than of different types or sub-genres of sophist). One of these, the sixth, defines the sophist as an expert in refutation using terms deliberately reminiscent of Plato’s earlier descriptions of Socrates. I leave aside the vexed question of what moral Plato wants the reader to draw from this plethora of ‘definitions’ in order to focus on aspects of the method germane to this essay. First, here is a sample of the completed method, the definition of angling – offered as an early and easy example of the method – reached by progressive division of the genus technē (art or expertise).

Stranger: In regard to angling, then, you and I aren’t only in agreement about the name, now, but we’ve also achieved an adequate account of the very thing itself. Of art as a whole one half was acquisitive; of the acquisitive a part was coercive; of the coercive a part was hunting; of hunting, animal-hunting; of that a part was aquatic-hunting, of that the lower part was fishing; of fishing a part was striking, and a part of that was barbing. And of barbing the part with an upward yanking strike is designated angling (the name resembling the action) – which is the thing we were looking for.

(Soph. 221b2–c3)

In this sample definition by division of angling, the genus technē, art or expertise, is first divided into two, acquisitive and productive.¹² Following the first branch – the acquisitive – and after eight more divisions the Stranger arrives at a definition by division of angling, adding a pun on its name for good measure.¹³ The method assumes that at each of the divisions one or more sub-forms is discerned, until the sought-for form or kind is ‘captured’, that is, satisfactorily defined.

Here are some features – relevant to our inquiry – of the method of division, the method by which the speaker reaches definitions by successive stages of division, that is, by narrowing down a generic subject matter until the target kind is defined. The features show that Plato is still according his forms or kinds Role A, forms as the objects of philosophical inquiry and knowledge, even though he is pursuing a new method, and is showing a greater interest than hitherto in issues concerning the relation between words and forms or kinds.¹⁴

1. (a) The inquiry is prompted by a question about the practice among the people of Elea (hometown of the Sophist’s main speaker, the Stranger from Elea): do they regard the terms ‘sophist’ ‘statesman’ and ‘philosopher’ as indicating a single thing, or two, or three? (Soph. 217a). We know that these were, to use a modern phrase, highly contested concepts, and that a philosopher (such as Socrates) might appear in the eyes of some to be a sophist. Here Plato shows explicit awareness that it can’t simply be assumed that to three different names there correspond three different things, even though the position argued for is that there are in fact three corresponding kinds (not just two or one).¹⁵

2. (b) The inquiry is explicitly to discover the ti estin or ‘what is it?’ of the sophist. The discussants must ensure that they have in common not just the word/name ‘sophist’ but also the thing (ergon or pragma, Soph. 218c1–5). The ‘thing’ searched for is the kind (genos) sophist, and the inquiry succeeds when it has found a satisfactory logos or definition of the item in question. Here Plato makes explicit some presuppositions which have governed his inquiries in earlier dialogues.

3. (c) Though the sample definition of angler quoted above could be said to offer the essence of angling, there is no explicit emphasis, in the Sophist, on looking for the real essence, for what makes something what it is. At most, one of the seven definitions provided of the sophist can be intended to give the essence of sophistry; it is usually held that the final one is so intended.

4. (d) Not every general term denotes a kind or form. This is an important warning, delivered by the Stranger in Statesman, where he uses the examples of the terms ‘barbarian’ and ‘beast’, Plt. 262–3. This indicates that having an unambiguous meaning – ‘barbarian’ means non-Greek person, ‘beast’ means non-human animal – is not sufficient for being correlated with a genuine kind. This is a key text for clarifying that, in pursuing the method of division to define an item, Plato does not invariably postulate a form corresponding to the common meaning of a term but seeks to discern the correct division of reality into genuine kinds.

5. (e) Far from slavishly following everyday language, Plato allows the Stranger both to invent new labels for branches of a craft he claims to discern, and – on occasion – to leave a branch unnamed. Sometimes the Stranger allows that it is matter of some indifference which of two possible names is given to a certain division.¹⁶

6. (f) At each stage of a division a form or kind must be divided into two or more forms or kinds. If we are to take the ontological claims seriously and accept the suggestion that at each division two forms or kinds have been identified, then in the Sophist alone some hundreds of forms are named. The honorific attributes used of forms in the middle dialogues (‘themselves by themselves’, ‘unchanging’, and so on) are absent from stretches where division is being exemplified, leading some to suppose that the theory of transcendent forms has been replaced by a general theory of universals.¹⁷ The Stranger insists that, in its aim of achieving understanding (nous), the inquiry shouldn’t avoid investigating lowly kinds such as sponging or louse-hunting (Soph. 227a–b). Gone is the uncertainty evinced by the youthful Socrates in Parmenides (130b3–e4) when he doubted if there were forms for man, fire and water, and recoiled from postulating forms for hair, mud and dirt.

From these points it is clear that in the stretches pursuing divisions to reach definitions Plato is still invoking his forms or kinds in the search for knowledge, for reaching an account of what things are. We might expect Plato to pursue the divisions with the aim of producing a taxonomic map of a segment of reality (for instance, of all the technai or forms of expertise), but only a very few hints of such a map are offered. Instead, what is sought and produced is a definition of an endpoint of a given branch, be it angling, sophistry, statesmanship. Role A for forms/kinds is still prominent, and the parts of the two dialogues, Sophist and Statesman, where the Stranger practises definition by division offer no support for discerning a shift to an interest in Role B, forms as involved in everyday thinking and understanding language. What the pursuit of division employs is far from everyday thinking, and, as points d and e show, everyday language is not what guides the inquiry. True, a far wider range of forms is envisaged, and the inquiries impinge more on empirical matters than in the middle dialogues. But the primary role for forms is that they are the objects of scientific knowledge. As far as the method of division is concerned, forms or kinds are not invoked in the investigation of everyday thinking.

3 The Sophist ’s Middle Part: The Greatest Kinds and Their Communion with One Another

3.1 Forms as Concepts? Interpretations of the Sophist’s Middle Part

Those scholars who argued that Plato assigned what I have called Role B to forms – forms as playing a key role in everyday thinking and in the understanding of language – based their claims chiefly on the Middle Part of the Sophist, the part in which the Stranger breaks off his definitional inquiry into what the sophist is to pursue problems arising from labelling a sophist a dealer in appearances and falsehood, items which allegedly must be characterised in terms of not being. To solve the problems about not being and about being, the Stranger identifies five ‘Greatest Kinds’ – Being, Same, Different, Change, and Stability – and proceeds to investigate what he calls their communion with each other. It is on the basis of this stretch of dialogue that scholars discern their favoured approach.

J. L. Ackrill

I have gradually passed from talking about Forms to talking about concepts, and I have taken these to be, in effect, the meanings of general words. Correspondingly I have implied that the task assigned in Plato’s later dialogues to the dialectician or philosopher is the investigation and plotting of the relations among concepts, a task to be pursued through a patient study of language by noticing which combinations of words in sentences do, and which do not, make sense … .

(Ackrill 1971: 207)

J. M. E. Moravcsik

The importance of this argument (Sophist 255b11–c3) lies in the apparent identification of the Forms Being and the Same with the meanings of ‘is’ and ‘is the same’. This identification helps us to understand the nature of the Forms as Plato conceives them in the Sophist.

(Moravcsik 1962: 47)

G. E. L. Owen

Platonists who doubt that they are Spectators of Being must settle for the knowledge that they are investigators of the verb ‘to be’.

(Owen 1970, opening sentence of his celebrated article on Sophist)

As these quotations show, Plato was held to have developed an interest in the meanings of predicate expressions, and in particular of expressions such as ‘is’ and ‘is the same’, which these writers equated with the concepts being and same. In addition, all of them held that one outcome of Plato’s investigation of the combinations of Greatest Kinds was the identification of different meanings (or uses) of the verb ‘to be’ (einai), though they disagreed over which meanings Plato succeeded in distinguishing, and at what points in the text such a move could be found. It was commonly held that in the Sophist Plato had moved on from metaphysical speculation about transcendent forms to what Ackrill called a ‘patient study of language’ in order to plot relations among concepts.

Should we, as Moravcsik and Ackrill claim, equate the Greatest Kinds with the meanings of the cognate terms? I shall offer a somewhat different account, focussing on claims about the so-called Communion of Kinds. But these scholars are right to draw attention to the way in which arguments in this Middle Part of the Sophist develop from and attend to linguistic expressions, ways of speaking. Fine’s claim (quoted in my introduction) that Plato ‘ignores meaning and linguistic understanding’ – however true it may be for the middle dialogues – does not hold true of the Sophist, as we shall see in Sections 3.3 and 3.4.

3.2 The Greatest Kinds and Their Cousins in Other Dialogues

The Sophist’s list of the five Greatest Kinds has some similarities to what we find in other dialogues. In Section 3.4 I shall argue that what Plato does in Sophist is strikingly different, but first for some similarities.

1. 1. In Part 1 of Parmenides attention is first focussed on the pairs of opposites likeness and unlikeness, to which are added one and many, change and stability ‘and everything else of that sort’ (Parm. 129a–e). What that last phrase covers is unclear, but we note that in Part 2 the antinomies include exploration of being and not being, and same and different. So all five of the Sophist’s Greatest Kinds had made an earlier appearance in Parmenides.

2. 2. Theaetetus 184–6

   In the final argument against the definition of knowledge as perception, Plato introduces what he labels koina, common things. The list of so-called koina starts with being and not being, same and different, two and one, like and unlike: grasping any one of these about some object of perception is said to be done ‘by the mind on its own’, in contrast to grasping a sense object, which is done by the mind via one of the senses. The items are labelled ‘common’ on the grounds that they apply in common to objects of more than one sense – such as a colour and sound, which both are and are different from one another but the same as themselves. The overlap between these first-mentioned koina and the Sophist’s five Greatest Kinds is striking, but the resemblances cannot be pressed too far. The Theaetetus’ list of koina is soon extended to include odd and even (185c10–d3) and later the pairs fine and shameful and good and bad. These additions widen the list’s members beyond the realm of very general, topic-neutral concepts, such that the key feature seems to be the a priori nature of the grasp of any member of the list even when applied to empirical objects.¹⁸

3. 3. Timaeus

   Finally, Being, Same and Different have a starring role in Timaeus in connection with the composition of the world-soul, and its cognition. At Timaeus 35a ff. they are presented as fundamental ingredients of the world-soul, governing both how reality is structured and what sort of cognition is possible. This ontological or even cosmological role for Same and Different is a far cry from the role they play in the Sophist’s discussion of the Communion of Kinds. But insofar as these three also figure prominently in the content of what the world soul is said to know (37a2–c5), they seem to play a somewhat similar role to their counterparts in Sophist.¹⁹

These lists intersect with the list of five Greatest Kinds of the Sophist, and the discussions in the Parmenides and the Theaetetus have a certain amount in common with what we find in the Sophist. But what is unique to the Sophist is its designation of some kinds as ‘vowel kinds’ and its implications.

3.3 (Some) Greatest Kinds Are Like Vowels

Why does Plato label his chosen kinds ‘greatest’, and what is the import of comparing some of them to vowels?

First, the label megista: greatest, or very great. Three of the list, Being, Same and Different, are explicitly recognised to apply to everything, and this may be part of what lies behind the designation ‘greatest’. ²⁰ The remaining two, Change and Stability, have provoked debate, but a widespread assumption is that in Sophist Plato is treating them as mutually exclusive and exhaustive: that is to say, everything is either an item capable of change, or an item incapable of change. Nothing is characterised by both, but everything is characterised by either change or stability (Soph. 250c12–d4).²¹ When he comes to explore the relations between the five kinds, Plato makes it clear that Change and Stability have a different status from the other three, because Change shares in each of the other three, but not in Stability. Indeed, all kinds share in Being, Same and Different, and each of these shares in itself. Mary Louise Gill marks this distinction by labelling the pair Change and Stability categorial kinds, by contrast with what she labels the structural kinds, Being, Same and Different. What she calls ‘categorial kinds’ are ones that can be organised into genus/species trees, while the nature of a structural kind ‘is determined by its functional role in enabling categorial kinds to be what they are and/or to associate with or differ from one another’.²² The later dialogues manifest Plato’s especial interest in what Gill calls structural kinds; we can add to the Sophist’s trio the kinds likeness and unlikeness which feature in Parmenides and Theaetetus, as noted in the previous section. An alternative, perhaps more enlightening, designation for the trio Being, Same and Different is ‘topic-neutral’.

But the comparison of some kinds with vowels – found only in Sophist – takes Plato’s discussion in a new and important direction, I believe. At Sophist 253a–c Plato draws an analogy between the science of kinds and that of letters, grammatikē. This allows him to compare some kinds with vowels and (by implication) others with consonants. It will transpire that among the five Greatest Kinds three of them – Being, Different and Same – are like vowels, while the remaining two are like consonants. The salient feature, both of the actual vowels, and of the kinds likened to vowels, is that they enable combination among the elements in question. An expert in spelling knows that among letters some are vowels, which ‘go like a bond among all the letters’, such that without a vowel it’s impossible for any one letter to blend with another (Soph. 253a). Likewise, the expert – the philosopher, apparently – can show correctly which kinds are in tune with which, and which not. This expert knows (i) which kinds go through all, holding them together so they can combine and (ii) which are responsible for diaireseis (either divisions, or negations, 253b–c). Subsequent discussion will clarify that Being is the vowel-like form responsible for combination, and Different the vowel-like form responsible for division and/or negation. Just as a vowel is a letter itself with a role in enabling letters to combine into words, so a vowel kind can combine with other kinds (both vowel kinds and consonant kinds) in meaningful (or perhaps true) sentences.²³

Thus, as I shall argue, it is for their role in sentences that Plato figures some Greatest Kinds as akin to vowels. This will emerge in the sequel. At 254c5–d2 the Stranger promises to investigate what sort of power of communion the Greatest Kinds have with one another, with the goal of clarifying Being and not Being. He then makes good this promise in an important stretch investigating how the kind Change combines with each of the other four kinds selected (255e–256e). We will see in Section 3.5 how Plato illustrates the functioning of the vowel kinds when he comes to explore the way the greatest Kinds do and don’t combine with one another.

3.4 The Focus on Linguistic Forms of Expression

Before scrutinising the central passage exemplifying the communion of kinds, I note a striking feature of the Sophist’s Middle Part: the prominence of arguments that take their start from ‘what we say’, and in particular from what we say when we predicate something of a subject. Now Part 2 of Parmenides had already featured many such arguments. Especially relevant is a stretch forming part of the second deduction.

Now ‘different’ in particular is a name for something, isn’t it? Certainly. So when you utter it, whether once or many times, you don’t apply it to another thing, or name something other than that thing whose name it is. Necessarily. Whenever we say ‘the others are different from the one’ and ‘the one is different from the others’, although we use ‘different’ twice, we don’t apply it to another nature (phusis), but always to that nature whose name it is. Of course.

(Parm. 147e3–6)

Note that the argument illustrates its claim – that in the two locutions the same nature is named – with two predicative uses of ‘different’. We find the same move in the Sophist, when the Stranger raises problems about being. He asks: ‘do we understand what earlier theorists are saying when they utter “estin” (is) or “gegone” (has come to be) … ?’ (Soph. 243b3–4). ‘What should we understand by this “einai” (to be) of yours?’ (243e2). Confronting dualists who claim that just two things, say the hot and the cold, are, the Stranger asks: ‘what do you (dualists) want to indicate (sēmainein) whenever you utter “on” (being)?’ (244a5). The Stranger pointedly focuses on everyday locutions, asking what we mean or indicate by these occurrences of forms of the verb be. Evidently all these different uses of the verb (estin, on, einai) are held to indicate the same thing, what the Stranger will later call the Greatest Kind Being (on). A similar style of argument is found later when he is identifying his five Greatest Kinds, and adding Same and Different to Being, Change and Stability. He notes that ‘each of them (viz, of the trio Being, Change, Stability) is different from the other two and the same as itself’ (254d14–15), and this locution – to which he immediately draws attention – is enough to add the two further candidates for Greatest Kinds, Different and Same, to the discussion. So here the use of a meaningful expression such as ‘is the same as’ is enough – apparently – to add a further kind to those under discussion.²⁴

Further evidence of Plato’s attention to meaning and linguistic understanding in Sophist is found in the frequency with which the Stranger raises the question of ‘correct speaking’. The puzzles (aporiai) concerning not being are explicitly described as arising from the way we speak about not being. The following are all taken from the Stranger’s remarks.²⁵

1. 1. Don’t you see from the very words I’ve used (autois tois lechtheisin) that what is not reduces to aporia even the person who’s out to refute it? It’s like this: whenever someone tries to refute it he’s forced to contradict himself in what he says about it. (Soph. 238d5–8)

2. 2. … if one is to speak correctly (orthōs) one shouldn’t call it (not being) one or many or… (Soph. 239a8)

3. 3. … we shouldn’t look for correct speaking (orthologia) about not being in what I say’ (Soph. 239b3–4)

The theme continues when the Late-learners’ problem (Soph. 251a8ff.) is presented as one about what can and can’t be said: contrary to everyday practice the troublesome thinkers labelled the Late-learners ‘don’t allow you to call a man good’ (ouk eōntes agathon legein anthrōpon, 251b9–c1). Attention to bits of language – often ones generating a problem – is very much a hallmark of the Middle Part of the Sophist.

3.5 Attention to Ambiguity in the Communion of Kinds Passage?

The previous section demonstrated the attention to linguistic forms of expression in the Sophist. This scrutiny of forms of expression, of what we say, is a significant feature of the Sophist, one reflected by the claims about the dialogue quoted at the start of Section 3.1 above. We now turn to the stretch where the Stranger fulfils his promise to show how the kinds communicate with each other, in order to show ‘that not being really is not being’.²⁶ It is in this section that the special role of the vowel forms is displayed, and that, I suggest, turns out to be their role in complete sentences.

A major claim made by earlier scholars is that Plato draws attention to an important ambiguity in the verb ‘to be’, showing that ‘to be’ has two meanings or at least two uses. If this is correct, then it partly vindicates the verdicts of those scholars I quoted at the start of Section 3.1, such as Moravcsik who equates the kind Being with the meaning (or meanings) of ‘to be’, or Ackrill who interprets the text as ‘the investigation and plotting of the relations among concepts’ and who discerns Plato distinguishing different meanings of the verb ‘to be’.

Is it correct to find Plato alleging an ambiguity in the verb ‘to be’? He presents the apparent contradiction: ‘Change is the same and is not the same’ and explains away the contradiction with the remark that when we said it ‘we were not speaking in the same way’ ou … homoiōs eirēkamen (Soph. 256a11–12). So it is natural to infer that Plato is pointing to some ambiguity here. The context makes it clear that the second clause (‘Change is not the same’) is equivalent to the claim that Change is different from the kind, Sameness, hence it is not (the kind) Sameness, while the first (’Change is the same’) asserts that Change is (predicatively) the same (as itself, presumably). What ambiguity is Plato pointing out, when he makes the Stranger say ‘we were not speaking in the same way’? Many scholars have argued that Plato here distinguishes between two meanings of ‘is’ in the sentence ‘Change is and is not the same’. If correct, then it would follow that ‘is’ – which as we’ve seen designates the kind, Being – has at least two meanings.²⁷ This would be a key result of the exploration of the Greatest Kinds and their combinations, and would vindicate the claim that in investigating them Plato is (in part) investigating the meaning(s) of the cognate expressions.

But here’s a different reading of this important stretch. Yes, Plato is pointing out the two possible understandings of the whole sentence ‘Change is the same’ – whereby it is true if understood as a predication, but false if understood as identifying Change with Sameness – but no, he is not isolating a single word, such as ‘is’, as the locus of ambiguity. In favour of this rival interpretation is that the key passage signally fails to highlight ‘is’ when it claims ‘we were not speaking in the same way’. His explanation is that ‘when we called it the same, we speak thus because of its sharing in the same, but when we call it not the same, that’s because of its sharing in the different …’. On this line, Plato’s aim is to point to two different readings of the sentence ‘change is the same’ as found in the apparent contradiction ‘Change is the same and not the same’ (on one of which it is a predication, on another a statement of identity), but without making any claim about the ambiguity of the single word ‘is’.²⁸ Now such a solution has a special attraction for those such as David Sedley who hold the view that Plato was ‘ideologically opposed to equivocation’ and are reluctant to allow that a single term has more than one correct meaning.²⁹ Whether or not we accept that Plato was ideologically opposed to equivocation, it is best to avoid ascribing an explicit recognition of different meanings of ‘is’ to him on the strength of this passage, given that another interpretation is available that credits Plato with an equally insightful solution, but without signposting different meanings for the verb be.

If this is right, then we can sum up the purpose and achievements of the Greatest Kinds stretch of the Sophist not so much as an examination of the meanings of the terms used to designate the Greatest Kinds, (and hence not an examination of the concepts being, difference, sameness and so on) but as an exploration of what certain sentences say and how, to avoid an apparent contradiction, they can be given alternative readings. This answers the problem the Late-learners had with a sentence such as ‘the man is good’, which they forbade on the ground that it made two things (man and good) one. Or, to put it another way, on the grounds that it identified two things which are different. On this reading, the kind of investigation we find in the Sophist’s Communion of Kinds stretch is indeed something quite novel, and very different from the kind of search for definitions by division we find in the Outer Parts, and in the Statesman. It enables Plato to diagnose the source of certain problems, such as those of the Late-learners and also those purveyed in the Euthydemus by the eristic brothers.

My verdict, then, is that the scholars quoted above were right to discern in the Sophist’s Middle Part a new interest in language and in meaning. Plato here invokes what he calls the Greatest Kinds, and their combination with one another, to explain and analyse sentences in an enlightening fashion, sentences which, because they are susceptible of different readings, had given rise to difficulties and were the source of puzzlement, either genuine or of the kind to be exploited by logic-choppers such as the brothers in Euthydemus. So it is safe to say that in the Sophist’s Middle Part we find the forms/kinds in what I labelled Role B above. But should we conclude that, in this dialogue at least, forms or kinds are now conceived by Plato as concepts, as the meanings of the cognate terms? This may be a step too far. As we have seen in Section 2, in much of the dialogue Plato continues to accord to forms/kinds their role as the objects of knowledge and philosophical inquiry, Role A. They are aspects of reality, awaiting the investigation of the trained dialectician.

3.6 Forms of Negations in Sophist?

This essay’s aim is to see whether, by the later dialogues, Plato’s conception of forms has developed in the direction of regarding them as meanings of words, rather than or in addition to their role as aspects of reality, real properties accounting for how the world is structured and for our knowledge of it. A further pointer in this direction is the puzzling stretch of the Sophist’s Middle Part where the Stranger discusses expressions such as ‘not large’ and ‘not beautiful’, explicating how they should be understood before finally introducing the most problematic of all, ‘not being’. After all, it is hard to think of the expression ‘not beautiful’ as designating a real property, a way in which reality is structured. Whether Plato treats these negative expressions as indicating forms is controversial. The text clearly labels ‘not being’ as designating a form, and it seems to imply, but does not assert, that ‘not large’ and ‘not beautiful’ designate forms.³⁰ This stretch raises a myriad of problems. If Plato regards items such as not large as forms, this seems to conflict with a passage from Statesman mentioned above (Section 2, at point iv) where the Stranger uses the example of the term ‘barbarian’ to warn against inferring from a term to a corresponding form. ‘Barbarian’, he noted, indicates a part but not a form of the more general form or kind human being, for the reason that barbarians are too disparate. By such reasoning there should be no question of a form or kind corresponding to the expressions ‘not large’ or ‘not beautiful’.

These expressions are explained in two stages. First, we are not to think that ‘not F’ means the opposite of F; the Stranger notes how ‘not large’ needn’t indicate its opposite, small, but can indicate ‘same-sized’, which is only different from and not opposite to large (Soph. 257b3–c3). Thus the claim – on one understanding – is that ‘not F’ indicates an item different from F chosen from a range or set of incompatibles.³¹ The second stage of the account invokes the Greatest Kind Different, and claims that it has parts with cognate names, as knowledge has. Just as geometry is the part of knowledge which is knowledge of shapes, so the not beautiful is the part of the Different that is different from beautiful, or (in a further characterisation) ‘set against the beautiful’. We can certainly draw the moral that to explain the expression ‘not beautiful’ we have recourse to two undoubted kinds or forms, the Different and the Beautiful. Whether the expression also designates a form or kind, the not beautiful, is not entirely clear.

While it is left unclear whether Plato wants to claim the status of forms or kinds for these negations (as Aristotle alleged, Metaph. A.9 990b 13–14), he certainly shows an interest in explicating negative expressions. This confirms what we have already noted (especially in Section 3.4), an increasing interest in everyday language and understanding. If the passage is committed to forms of negations, then these forms cannot easily be said to be real properties suitable for philosophical investigation – as indeed the Stranger’s warning (in the later dialogue Statesman) about ‘barbarian’ and ‘beast’ indicates.

4 Drawing the Threads Together

We have seen (Section 2) that in the Method of Division Plato pursues the ‘What is it?’ question by a new method involving what he calls dividing into kinds/forms, to discover what reality corresponds to a given term, such as ‘angler’, ‘sophist’, or ‘statesman’. Here, even though there are many novelties in the treatment, and especially in the types of forms/kinds discerned, these forms or kinds play the ‘old’ role, Role A, of being the objects of knowledge and scientific investigation. The rather different investigations of the Middle Part focussed on ‘some of the greatest kinds’, and those inquiries are of a more logical character. In particular, the focus was on the ways in which the greatest kinds combine with one another, and the chief outcome was to help understand certain sentences, and how they can be given different readings (Section 3.5). The exploration takes on a more linguistic character (Section 3.4), and the results are important for the understanding of everyday thinking as well as more philosophical thinking. Thus, Role B for the kinds seems more to the fore.

But there is no sign that Plato recognised any such dichotomy in his employment of forms. He continues to use the same terminology (genos, eidos) for empirical and arguably trivial things such as the aquatic branch of hunting, and for the Greatest Kinds Being, Same and Different. Furthermore, there is evidence that he sees no major distinction in the roles accorded to forms (of the kind I have been exploring) in a famously difficult passage about dialectic in the Middle Part of the Sophist. There the Stranger, purporting to describe the philosopher’s pursuit, dialectic, appears to describe, in adjacent clauses, both exploring the communion of kinds and pursuing the divisions of kinds practised in the earlier stretch. Even more puzzling, he seems to say that to pursue the divisions is simply to discern how kinds can and can’t combine.³²

This brings us to one further issue: how are we to understand the occasional assertions that link the existence of forms with the power of dialegesthai (conversing or philosophising)?

5 Forms as Necessary to Ordinary Thinking, or Only Necessary to Philosophical Thinking?

In several places in the later dialogues we find remarks connecting forms to thinking and/or to dialectic. In her careful study of Aristotle’s On Ideas, Fine devoted a chapter to Aristotle’s statement of and objections to what is known as the Object of Thought argument, according to which (claims Aristotle) Platonists held that forms must exist to explain the possibility of thought.³³ After a discussion of the Platonic passages that may connect forms to the possibility of thought, Fine concludes that Plato is not committed to the Object of Thought argument. Important for our study is Fine’s observation that the Platonic passages suggestive of the Object of Thought argument are all from the late dialogues. This, she suggests, is in contrast to the middle dialogues, which simply take for granted that we have beliefs and understand the terms we use, and which posit forms only for metaphysical reasons (to explain shared natures) and to explain the possibility of knowledge.³⁴

When we scrutinise the handful of passages connecting forms to thinking we find in many cases a tantalising unclarity about just how the key terms should be understood. At Parmenides 135b–c, after raising a host of difficulties about forms, Parmenides cautions against drawing the wrong conclusions and dismissing the forms. For, he says, if one does not admit them one will have nothing on which to fix one’s thought (dianoia), and (by denying them) one will destroy the power of dialegesthai. The terminology used here allows of very different interpretations. Should we understand this as the claim that someone who denies forms renders impossible thinking of all kinds, and conversation – since the verb dialegesthai originally just means to have a discussion? Or as the claim that denying forms removes the possibility of high-level thinking and of dialectic, understood as the highest knowledge? In the light of its uses in Republic (511d and elsewhere) and Philebus (57e) it is more likely that Plato is here using the phrase ‘power of dialegesthai’ in the technical sense he carved out for it there.

In the Sophist’s Gigantomachia we find what may be a similar claim. The Stranger argues that nous is impossible without unchanging things (249b5–6) and that nous cannot exist without ‘that which is in the same respect in the same way and about the same things’ (249b12–c3). This last phrase seems to pick out forms, so here we have the claim that without forms there can be no nous. Again, we must choose between interpreting this as the claim that without forms no thinking at all is possible, or as the claim that no high-level intellection is possible. Recent work on Plato’s use of nous has established that for Plato nous regularly denotes a high-level cognitive achievement (and is not used for ordinary thinking, or for the faculty of thinking generally).³⁵ So this text, despite being from the Middle Part of the Sophist, does not permit the interpretation that any thinking requires the existence of forms.

One further Sophist text has encouraged the view that in the Sophist forms or kinds are the meanings of words, central to any thinking and speaking. If that were the correct interpretation, then we would have clear evidence that forms now play the role of concepts. The text in question is the famous claim at 259e5–6, where the Stranger says that ‘to detach each thing from everything is the most complete destruction of all logoi (statements?), because logos has come about for us thanks to the reciprocal interweaving of forms (sumplokē tōn eidōn)’. Much scholarly attention has been devoted to this claim, and we can only scratch the surface of the controversy here. The difficulty is reconciling this claim with the subsequent examples of statements discussed: the true ‘Theaetetus sits’ and the false ‘Theaetetus flies’. Attempts to understand Plato to be claiming that any statement involves its speaker interweaving two or more kinds have failed; Ackrill’s alternative, by which the interweaving of kinds is a sort of backdrop condition for the meaningfulness of predicate expressions in statements, also fails to fit the text. A recent account takes the claim about the combination of forms to refer, not to statements in general, but to those affirmative predicative sentences such as ‘Change is the same’ (my example) discussed earlier.³⁶ This deflationary explanation may well be the safest, especially since Plato’s subsequent treatment of true and false statements conspicuously fails to mention forms or kinds at all, as I will go on to discuss.

6 False Belief and False Statement

I end this exploration of the roles Plato accords to forms in later dialogues with some brief remarks about his treatments of falsity.

The later dialogues contain some intriguing discussions of false belief and false statement. Plato was aware of earlier puzzles purporting to make false statement and false belief impossible, but he was equally certain that such falsehoods were prevalent (see, for instance, Tht. 187e5–8). In several later dialogues the puzzles are paraded, and in the Sophist one major puzzle is finally solved. The texts in which we find Plato’s discussions of falsehood are among the most philosophically rich in the entire corpus, and none more so than the skillful parading of problems of false belief in Theaetetus. They underline how Plato has recognised the importance of being able to show that a false belief is still a belief, and a false statement is still a statement. We might say that he has become interested in accounting for thinking in general, and not merely in knowledge; and in statements in general, not merely those that are the vehicles for knowledge.

What is it that makes a person’s thought a thought about such-and-such? How can I be thinking about a thing but still think something incorrect about it? How can my statement say something but still be false? These important questions are discussed in Theaetetus, Sophist and Philebus. Here we have evidence that Plato’s philosophical interests have expanded to include important questions in the philosophy of mind and of language; or, to put it in the terms earlier in the essay, he has become interested in the philosophical problems arising from everyday thinking. But the intriguing stretches of text in which these discussions are conducted make no explicit references to forms at all. To take a key example, the wax-tablet model in the Theaetetus (191b–199d). That model is invoked to explain false beliefs as the mismatching of a perceived item with a known (i.e., remembered) item, as when I mistake a person I’m currently perceiving for another whom I know. In describing how memories get laid down, Socrates speaks of both perceptions and ennoiai (thinkings?), using a word – ennoia – which was to become an important technical term in later epistemology. Just what it refers to in the Wax Tablet model of false belief is disputed: perhaps the thought of an individual man, perhaps the concept of man.³⁷ What matters for our inquiry is that Plato doesn’t make his speakers call these thought items forms (even if they are correctly understood to be concepts).³⁸

I make the same observation about his discussion of true and false statements in Sophist. In modern terms, Plato indicates that the key to solving the problem of how a statement can be meaningful and yet false is to distinguish the referring part of a statement from the part that predicates something of the item referred to. The Stranger explains how, to make a statement, the speaker must name something, and, in order to say something about the thing named, the speaker must weave together with it something else (picked out by the verb, rhēma): he calls this an action or an inaction or a being (Soph. 262c3). The referent of the predicative part of the statement – ‘sits’ or ‘flies’, in his examples – does not get labelled by the terminology of forms or kinds (but by ‘action’, ‘inaction’, etc.), even though it would be quite natural to consider sitting or flying as kinds, given the range of kinds (including fishing, hunting and so forth) that were on parade earlier in the dialogue.

I am not insisting that Plato does not envisage the contents of our thoughts or statements as forms, or that (in his view) the form man plays no role in how I can come to have the thought that what I see is a man. I note only that in the intriguing discussions of ordinary thinking surveyed here, including false thinking, Plato does not use the language of forms, even in the later dialogues where his interests have expanded to include some philosophical questions about the nature of thought. This is true also of the stretch in Theaetetus (185a–e) where Socrates discusses what he calls the common items (Section 3.2). Their role initially is to be that which we think about perceived items, such as when we think that a colour and a sound are and are different from one another. It is highly natural to understand this stretch in terms of Plato discussing the mind’s role in applying concepts to various items. An influential treatment argues for understanding the koina as concepts, and for finding in the passage Plato’s solution to the problem of knowledge in the Theaetetus.³⁹ The overlap between the list of common items and that of the Greatest Kinds emboldens many critics to identify these common items with forms. But it may be significant that Plato avoids the language of forms at a point where he is making an important observation about what we may call everyday thinking, and the role of the mind on its own in making such everyday judgements.

7 Conclusions

The task was to consider concepts in late Plato. I approached this issue by asking whether in his later dialogues Plato allots to forms or kinds a role in everyday thinking and understanding language (Role B), where previously he has accorded to forms the role of being the objects of knowledge (in contrast to belief) and of scientific investigations (Role A). In Section 2 I showed how, in the pursuit of the method of division to answer the ‘what is it? question, forms evidently retain their role of being the object of knowledge and scientific inquiry in the later dialogues Sophist and Statesman. Even though Plato introduces a vast number of what he calls forms or kinds, in the course of dividing the very general kind technē, and even though they are not accorded the honorific designations familiar from the middle dialogues, they are still envisaged as real properties whose role is to explain what things are.

Section 3 surveyed the Sophist’s Middle Part for evidence of Role B, forms or kinds as the meanings of the terms, especially in the discussion of the five Greatest Kinds. It recognised (Section 3.4) that this stretch accords far greater prominence to language in discussing philosophical problems. In considering the treatment of the Greatest Kinds I cast doubt on one prominent claim, that Plato identifies different meanings or uses of the verb ‘be’, thereby removing one plank from the claim that the stretch is an exploration of Being, understood as the meaning(s) of ‘to be’. I suggested instead (Section 3.5) that the significance of identifying the so-called vowel forms was not so much to explore the meanings of ‘being’ ‘same’ and ‘different’ as to come up with an apparatus enabling him to disambiguate certain sentences. I argued that by displaying different readings of the relevant sentences (using paraphrases invoking being and difference) Plato showed how the sentences were not, despite appearances, contradictory. Section 3.6 noted another discussion which hinted at forms as the meanings of expressions, the discussion of items such as the not large and the not beautiful, explicitly framed in terms of the expressions ‘not large’, ‘not beautiful’ and so forth.

Section 4 surveyed some of Plato’s explicit treatments of dialectic, and noted how in them he apparently ran together the explorations characteristic of the method of division and – what to our eyes is rather different – the investigations of how the Greatest Kinds combine with one another. That is, his own programmatic remarks do not seem to recognise a major distinction in roles for forms. If his logical investigations in Sophist do manifest a greater interest in what we might label the exploration of concepts (as some stretches of the Middle Part suggest), there is no trumpeting of a new role for forms or kinds, other than the prominent announcement of an interest in the so-called Communion of Kinds, whose role I have discussed in Sections 3.4 and 3.5 above.

Finally, in Section 5 I offered a very quick survey of some of Plato’s penetrating investigations of issues in the philosophy of mind and language, including his discussions of false judgement and false statement. I noted that terminology for forms is absent from these discussions, even where some have thought that he must be invoking forms – for instance for what is predicated in a statement, whether true or false. A philosophical treatment of judging and of statement-making is likely to be a key locus for an interest in concepts. While we may say with some confidence that in the relevant passages of Theaetetus and Sophist Plato has evinced an interest in the role of concepts, it is notable that he does not explicitly invoke forms when doing so.