BMCR 2004.11.04

Plato on Knowledge and Forms. Selected Essays

, Plato on knowledge and forms : selected essays. Oxford: Clarendon Press, 2003. ix, 447 pages ; 24 cm. ISBN 0199245584 $35.00 (pb).

Gail Fine’s collection brings together fourteen previously published essays which focus on two of the most central themes of Plato’s dialogues, namely knowledge and forms.1 Generally speaking, the first nine essays discuss problems of knowledge and focus on particular dialogues, such as the Meno, the Republic or the Theaetetus, while the final five essays discuss problems related to forms and proceed, in principle, from a cross-reading of some of the dialogues as well as from some critical remarks made by Aristotle on forms (see the list of essays at the end of the review). The essays are tied together by a comprehensive introductory essay.

To my mind, two of Fine’s claims are particularly significant: first, her view that Plato did not in fact subscribe to some of the Two-Worlds-Theories traditionally attributed to him — theories that strictly separate forms and sensible particulars. Forms, though separate in a certain way, are immanent in particulars as their properties. They are separate insofar as they have the capacity for independent existence. Second, and again against a traditional reading, Fine suggests that knowledge may cross the line: there is knowledge not only of forms (but also of particulars), and there is not only knowledge of forms (but also mere belief).

The essays are all well-argued. They provide the reader with a wealth of valuable interpretations and suggestions with regard to the two main topics and some connected topics. Even where one perhaps would not agree with some particular claim it is always rewarding to look very carefully at the arguments that Fine presents for it. Anyone interested in epistemology and metaphysics in Plato will be interested by this collection.

In what follows I shall first present a brief summary of the content. Secondly, I shall mention briefly some points in Fine’s interpretations which I regard as particularly attractive. Finally, I will discuss some issues which seem to me to be somewhat more problematic.

First, then, a brief survey of content (which is not intended to cover all chapters of the collection). The “Introduction” (ch. 1) very usefully presents and connects the major lines of thought in the essays. It responds to criticism which was made of some of Fine’s claims after their first publication. It also contains some revisions, or hints at such revisions. It closes with an explanation as to why Fine introduces Aristotle while discussing Platonic forms — which might indeed be a source of surprise to some, since Aristotle is not usually considered the most reliable witness to Plato’s thought. The introduction does not attempt to provide one unified theory — which would in any event be an unusual task to attempt in a collection — even though this collection is more integrated than other such collections by a single author. Nor is the introduction intended to replace a close reading of the essays it introduces.

In “Inquiry in the Meno” (ch. 2), Fine explores the main problem of the dialogue: how anyone can inquire into something (for instance, virtue) without knowing anything about it (p. 44, 50-51). In fact, it turns out, that inquiry is possible and does not require previous knowledge. Via elenchus, progress is made by getting rid of false beliefs while keeping true ones (cf. pp. 3-4, 54-61). Fine connects this with remarks on the “theory of recollection” and the role which it is traditionally thought to play in the solution of the problem, but she somewhat downplays this role (pp. 61-65).

In “Knowledge and Belief in Republic V” (ch. 3) and “Knowledge and Belief in Republic V-VII” (ch. 4) Fine pursues several lines, of which I shall mention only the more important ones. Firstly, she argues against a traditional assumption according to which there is a strict, and exclusive, link between forms and knowledge on one side and between sensibles and belief on the other side (pp. 67, 82). Secondly, she argues for a content analysis of knowledge. According to this analysis the “is” or “is not” to which, as their objects, knowledge and belief are supposedly related (as in ” x knows something that is”) is (mainly) to be understood as a veridical“is”, that is as an “is” that is to be supplemented by “true”. It is not to be understood in the traditional way, namely as an existential“is” (as in ” x knows something that is, i.e., that exists“). Thirdly, she attributes to Plato a coherentist view of epistemic justification and a holistic view of knowledge (pp. 99, 111). Fourthly, and principally, she takes issue with a traditional “Two-Worlds” interpretation of Plato’s (or a platonic) theory of forms (pp. 66-67, 85-86). According to this Two-Worlds theory there are two realms of being or beings, the realm of forms and the realm of sensibles, which are somehow, but strictly, separated from each other, though the sensibles depend on the forms, while the forms are said to be beings in the proper sense.

Chapters 6 to 10 concern the Theaetetus and its main question, what is knowledge? In chapters 6 to 8, Fine focuses on the first main part of the dialogue and its definition of knowledge, according to which knowledge is perception. The dialogue connects this definition with other theses, namely Protagoras’ “man is the measure of all things” and a Heraclitean flux thesis, according to which change is all that there is. These theses are said to amount to the same thing ( Tht. 160d6) and, while this is difficult to understand if taken literally, we have to interpret those theses in a way which at least establishes some informative relation between them, for instance, as Fine interprets them, a relation of mutual support.

In “Protagorean Relativisms” (ch. 6) Fine argues for the thesis that the view to be attributed to the Protagoras of the Theaetetus is not — as is usually assumed — some kind of relativism (p. 155), be it perceptual relativism or relativism about truth. According to perceptual relativism, there is no “white simpliciter, only white for you and white for me” (pp. 146-147). Relativism about truth — Fine’s principal target — amounts to saying that “‘true for A’ means the same as ‘believed by A'” and that “there are no absolute truths” (p. 140). According to Fine, this cannot be the Protagorean view, since relativism of neither kind is connected or connectible to the Heraclitean thesis.

Fine argues that the Protagorean view is rather some kind of infallibilism, according to which “all beliefs are true — true simpliciter, not [as in the case of relativism — F.B.] merely true for the one who holds them” (p. 155). Furthermore, according to infallibilism, “things are — really are — (and are only …) however they are believed to be”, and while “at least many objects and their properties are extra-mental, [..] the existence of all objects and properties depends on their being perceived or believed to be as they are” (p. 155 n. 58). As Fine herself would not deny, this is hard to swallow. Yet, Fine argues, only if interpreted in an infallibilist way is Protagoras’ view connectible to some, at least modest, Heraclitean view.

According to Fine, further confirmation for the appropriateness of an infallibilist reading comes from the refutation of Protagoras (see Tht. 169d-171d). This refutation, Fine argues (see “Plato’s refutation”, ch. 8), succeeds only on an infallibilist reading, not on a relativist reading: only the infallibilist could and would have to accept the premises from which a non-infallibilist conclusion is derived.

“Knowledge and Logos” (ch. 10) tackles the question of whether there is any indication of a positive answer to the main question of the Theaetetus, even though the dialogue seems to end with an aporia. According to Fine, there may be such a way out since not all options are dealt with. Thus, it still seems possible to presume some kind of “interrelation model of knowledge” in the dialogue (p. 228), according to which the basic elements on which knowledge is founded do themselves have an account which determines what they are. This account consists in their being interrelated to each other. An example of such an interrelation is provided by Theaetetus himself in the well-known letter-examples: in his attempt to justify his claim that elements do not have elements and, hence, no account, Theaetetus does precisely give some kind of account of such elements (here: of letters) by relating them to other (here: linguistic) elements (see Tht. 203b2-8, Fine pp. 237, 242-243, 250-251). Though, finally, this elementary account becomes circular — all elements are somehow related to each other —, we only have to draw the circle wide enough to obtain a substantial, informative account (for the coherentism and holism implied here, see already ch. 4). The Theaetetus implies no immediate, account-less knowledge of elements nor does the dialogue imply elements as foundations of knowledge that do not have some kind of account.

Chapters 11 to 15 turn to problems of our understanding of forms. This is closely connected to the discussion of knowledge since forms are the main object of knowledge — even if we are to assume a non-traditional cross-over that also allows knowledge of items other than forms. Aristotle thought that forms are to be conceived as separate from that of which they are forms. Yet Plato himself never says so, and, at most, it is true for some forms (pp. 254-255). In fact, according to Plato, “separation” means only that “forms can exist independently of sensible particulars”, and it is in this sense of “separate” that they are separate. Thus, they are separate even if they are instantiated by particulars (pp. 31-32, 263).

Thus understood, the fact that forms are separate goes well together with their being “immanent” in the things they are forms of, namely as “properties of them” (pp. 301, 303). In order to establish this view, we have to solve certain problems about self-predication (as in “largeness is itself large”). Fine suggests an understanding of self-predication according to which “the form of F can be F in virtue of making its possessors F; the form of large is large, not because of its size (not being extended, it has no size), but because it is the property that makes things large” (pp. 314-315).

Fine, furthermore, argues for a “relational theory of universals”, that is, of forms (since forms are said to be universals). “Relational” means that if x is F then there are two things involved, namely x and something other than x. This, however, does not imply that Plato would have accepted or would have had to accept bare particulars as property-bearers (see pp. 327-328, 341).

A difficult issue is the problem of how forms may have any influence in this world. How, if at all, do they figure in causation? Pace Aristotle, Plato did not ignore efficient or teleological aitiai (pp. 374, 389). Forms, according to Plato, though not themselves efficient causes (pp. 385-386), are relevant factors in causation: together with objects and time, they play a role in constituting events and, thus, are not irrelevant to change (pp. 357, 360, 390). This would be surprising only if, again, one adopted a Two-Worlds interpretation of Plato. In that case, it would be difficult indeed to find any causally operative connection whatsoever between the two worlds. There is, however, no good reason to attribute such a theory to Plato, and the remaining systematic problems are close to problems which we ourselves have to face when we talk about the causal relevance of, say, properties or instances of properties.

Finally, Plato’s forms fare no worse with regard to plausible substantiality criteria than do Aristotle’s candidates for substances (see pp. 424-425).

Having given a brief survey of content, I shall now single out some of the points I regard as attractive interpretations or suggestions. Given the richness of the collection, this cannot be a representative choice, but only a very subjective one.

I feel much sympathy for the criticism of the traditional Two-Worlds theory, according to which there is an ontological hiatus of a special kind between forms and sensible particulars (see chs. 3 and 4). In the first place, there is nothing interesting or even compelling about such a theory. Secondly, we are not forced to, or justified in, accepting it by the text of the Republic itself, if we read it very carefully as a dialogue and if we pay attention to what is actually claimed and by whom. Nor does the Republic justify the traditional view that forms and sensible particulars are strictly separate insofar as they are objects of knowledge and belief: as Fine notes, there certainly are things about which some have beliefs, and others knowledge, and, hence, a distinction between knowledge and belief does not imply a distinction between their respective objects. Thirdly, in the Theaetetus belief figures prominently in the account of knowledge (compare also Meno 98a), and yet the account is not criticised for this.

Furthermore, I regard Fine’s presentation of Aristotle’s relation to Plato in her discussion of forms as an important contribution. It is often assumed that Aristotle is a poor or unfair witness to Plato. There are, as Fine argues, at least some good philosophical reasons for Aristotle to understand Plato in the way he does, and, then, to criticize him. Fine does not claim that Aristotle chose the only possible or the most favourable interpretation. Nevertheless, he may be regarded as providing a systematically illuminating starting point (all of this is covered in, for instance, “Immanence”, ch. 12).

As for the problems of knowledge, I am sympathetic to the interrelation model of knowledge. Though this model is unlikely to convince everyone, nor even many, perhaps, it provides an answer, or part of an answer, that is at least as attractive as any other answer to the difficult problem of founding knowledge on items (elements) which in themselves were originally supposed not to be objects of knowledge.

On the other hand, there are some points which I find less convincing. I shall restrict myself to a few comments on the Meno and the Theaetetus.

I would like to begin with a more general point which concerns the question of how we are to read Plato. Fine is very sensitive to issues of arguments in the dialogues. Nevertheless, we could, and indeed I believe we must, go a little further, even though I fear that what I intend to say may be nothing new. Since dialogues usually take place between a “questioner” and an “answerer”, we have to be even more assiduous than Fine in considering what is actually claimed by whom. True, there are many cases in which Socrates is presented as giving his own view. Yet, more often he simply asks questions, and in these cases it is only the answerer who is committed to a claim that is construed from the question and his own answer to it. On some occasions in the dialogues, Socrates stresses the point that, regarding certain claims, he himself does not commit himself to a particular claim (see, for instance, Protagoras 330e7-331a1, Theaetetus 157c4-d3, 161a6-b8, also [Ps.?]-Plato Alc. 1, 112d10-e8; see also Aristotle Topics VIII 1). All this leaves little room for making claims such as “Plato says”, “Plato claims” etc. — expressions which Fine, and many others, do not hesitate to use regularly. When she says, for instance, that “Plato … has claimed that belief is a distinct capacity from knowledge” (p. 74) and when she refers for this to Rep. 477e6-7 (Burnet’s lineation), she is actually referring to a claim made by Glaucon who is speaking here on behalf of the doxazôn. There is no good reason to accept this claim as being Plato’s own (nor would Fine herself accept all the other claims made in this passage, that is, in Rep. 476-480, as Plato’s — a passage that is of crucial importance, and which is usually taken to be the basis, for the traditional Two-Worlds theory against which Fine herself argues here).2

As for the Meno, I think, Fine is quite correct in downplaying the role of the theory of recollection as the solution to the problem of inquiry. Indeed I would go much further than that. Given the situation of the dialogue, what exactly could possibly motivate us to accept this theory as Plato’s own theory? (What kind of story does Socrates use to introduce the theory? What is the point of introducing it? Is he being honest in 81a8 in calling it “true” and “beautiful”? Why, in offering the story, does Socrates suddenly give a Gorgian speech? Why does he give such a speech to Meno who had claimed to share Gorgias’ views (71d1-3).) The theory seems to be introduced in order to clarify, by way of analogy, certain features of the process of inquiry. Meno, however, does not understand what is going on and simply accepts the theory itself (while Socrates does not seem to accept it, see 86b6-7). He does not pay attention to the features of the theory that make it relevant as analogy.3

As for the Theaetetus, I wonder whether we are really better off with infallibilism as an interpretation of Protagoras’ “man-measure” thesis. Is infallibilism really connected to the Heraclitean view, for instance by a relation of mutual support? Fine claims that “if objects are, and are only, as they appear, then, given that objects constantly appear different, objects are constantly changing; and so infallibilism is committed to Heracleiteanism” (p. 157). Furthermore, she claims, that “Heracleiteanism supports narrow infallibilism” (narrow infallibilism being infallibilism about perceptual beliefs only; see p. 157). I am not sure, though, whether we may indeed assume either the commitment or the support. I am unable to see why a change of objects should be the interesting point in Protagoras’ ontology; in other words, I do not see why the case of changing objects should be the only case to be covered by Protagoreanism. Why, for instance, should the case be excluded from Protagoras’ consideration in which the very same object under the same aspect appears in a different way to different people at the same time (for instance, the case of some wind that appears warm to you and cold to me)? And why should the other case be excluded which presupposes different times and different appearances of that object yet no change (of properties) in the perceived object itself? The object could very well remain the same while it is perceived by people with capacities for perception that differ in the relevant aspect or capacities that change over time with regard to the relevant aspect. In these cases there is (so far) neither commitment to, nor support from, a Heraclitean view. (Fine seems to assume that the different appearances in question are to be understood as occurring at different times; see pp. 157, 179-180. The Theaetetus does mention such cases. But, again, they are not the only cases covered.)

In addition to this, if all beliefs are true simpliciter, as infallibilism claims, then this includes contradictory true beliefs, too (if, that is, we also take those cases just mentioned into account), such as my true belief that a is green and your true belief that a is red. This would imply that a is — really is — (and is only) red as well as that a is — really is — (and is only) green. What is worse is that, according to infallibilism, things are supposed to be as they are believed to be, so that they may, at the same time, have properties — for instance being (only) green in colour and being (only) red in colour -, which they could not possibly have at the same time in themselves. Neither is it clear how this could be made possible by any kind of Heraclitean theory.

So while Fine is correct to point out the importance of clarifying the link between the theses, it is not clear how this link is better provided by an infallibilist reading; nor is it clear how we could accept an infallibilist reading just by itself. (This is not to say that any relativist reading would be any more successful in connecting the theses. It is only to say that infallibilism does not do the job either: it is not the easy way out.)

In fact, change seems to play a quite different role in this part of the dialogue, namely as a constituent in perception. This is where Heraclitus comes in. According to the Heraclitean theory, all that there is comes to be through permanent change (cf. 152d2-e1, 156aff.), and this also holds true of perceptual content. This does not, however, mean that the object that is being perceived undergoes permanent change in such a way that, ordinarily speaking, it permanently changes its properties (for instance, ordinarily speaking, by changing its colours), but rather that its very property — for instance, being red — “exists” through being permanently brought about. In perceiving there is something permanently going on between the object and the perceiver. Change of appearance, however, does not imply change on the side of the object. This would seem to be a fairly good basis for a relativist view.

This collection deserves more detailed criticism than has been possible here, but I trust that this little has made evident, that even if one does not agree with everything in it, much can be learned from such a valuable contribution to our understanding of Plato on knowledge and forms.

LIST OF ESSAYS (with the location of their previous publication, as given by Fine, pp. viii-ix):

ch. 1: “Introduction”.

ch. 2: “Inquiry in the Meno“, in R. Kraut (ed.), Cambridge Companion to Plato (Cambridge, 1992), 200-226.

ch. 3: “Knowledge and Belief in Republic V”, Archiv für Geschichte der Philosophie 60 (1978) 121-139.

ch. 4: “Knowledge and Belief in Republic V-VII”, in S. Everson (ed.), Epistemology (Cambridge, 1990), 85-115.

ch. 5: “Plato on Naming”, Philosophical Quarterly 27 (1977) 289-301.

ch. 6: “Protagorean Relativisms”, in Boston Area Colloquium in Ancient Philosophy 10 (1994) 211-243.

ch. 7: “Conflicting Appearances: Theaetetus 153D-154B”, in C. Gill and M. McCabe (eds.), Form and Argument in Late Plato (Oxford, 1996), 105-133.

ch. 8: “Plato’s Refutation of Protagoras in the Theaetetus“, Apeiron 32 (1998) 201-234.

ch. 9: “False Belief in the Theaetetus“, Phronesis 24 (1979) 70-80.

ch. 10: “Knowledge and Logos in the Theaetetus“, Philosophical Review 88 (1979) 366-397.

ch. 11: “Separation”, Oxford Studies in Ancient Philosophy 2 (1984) 31-87.

ch. 12: “Immanence”, Oxford Studies in Ancient Philosophy 4 (1986) 71-97.

ch. 13: “Relational Entities”, Archiv für Geschichte der Philosophie 65 (1983) 224-249.

ch. 14: “Forms as Causes: Plato and Aristotle”, in A. Graeser (ed.), Mathematics and Metaphysics in Aristotle / Mathematik und Metaphysik bei Aristoteles (Bern: Haupt, 1987), 69-112.

ch. 15: “Plato and Aristotle on Form and Substance”, Proceedings of the Cambridge Philological Society 209 (1983) 23-47.


1. I would like to thank Damian Quinn and Sheila Regan for corrections to my English.

2. For an interpretation of Rep. 476-480, which considers these problems, see Th. Ebert, Meinung und Wissen in der Philosophie Platons. Untersuchungen zum ‘Charmides’, ‘Menon’ und ‘Staat’ (Berlin, New York, 1974), 105-130.

3. For these questions and the corresponding interpretation see, again, Ebert, Meinung und Wissen, 83-104.