BMCR 2021.12.02

Plato’s epistemology: being and seeming

, Plato's epistemology: being and seeming. Oxford; New York: Oxford University Press, 2021. Pp. 272. ISBN 9780198867401 £65.00.


The main purpose of this monograph is to show, through an analysis of Plato’s conceptions of epistêmê and doxa, that Plato’s epistemology is fundamentally different from the kind of investigation undertaken by contemporary analytic epistemologists. Moss breaks down this overarching contention into two sub-claims: first, Plato’s epistêmê and doxa should not respectively be assimilated to the current notions of knowledge and belief. Unlike their contemporary counterparts, Plato’s epistêmê and doxa are not topic-neutral cognitive states, but powers individuated and essentially defined by their proprietary objects, which, contrary to a common opinion, Moss regards as distinct: epistêmê is of Being and doxa is of Seeming – hence the book’s subtitle. So what sets these two modes of cognition apart are not purely cognitive prerogatives, but their exclusive application to non-overlapping objects. Second, Plato’s overall epistemological project differs conspicuously from those carried out by modern epistemologists. Notably, as Moss stresses in the conclusion, it is first and foremost driven by metaphysical and ethical concerns, its ultimate motivation being to contribute to the elaboration of an adequate answer to the question of how we should live.

Through both these claims, Moss pits herself against what she suggests is the standard scholarly view (more later), brought into prominence by the influential studies of Gail Fine[1], and based, among else, on the opinion that the objects of Plato’s epistêmê and doxa overlap. Moss regards this reading as anachronistic and insufficiently borne out by the texts, which in her view suggest instead that for Plato the crucial divide between cognitive states tracks the metaphysical divide between levels of reality. Moss thus provides a fresh defense of what has been the dominant interpretation of Plato’s epistemology since antiquity, fallen into disrepute among analytically trained historians of the last forty years mostly because ‘it violates [their] intuition that the same thing can be first believed and then known’.[2] This defense, conducted through a judicious assessment of the evidence, is, I believe, largely successful, and the book is a welcome event in Platonic studies. In any case, the ingenuity and exactitude of Moss’ argumentation and the commendable clarity of her writing will make this volume a rewarding read even for those who will remain unpersuaded by her conclusions. In what follows, I shall first provide a selective critical synopsis of the book and then voice two broader concerns regarding the scope and the method of Moss’ investigation.

The book is tidily structured around the opposition between epistêmê and doxa. The ten chapters constituting its core can be seen as naturally falling into four parts:

(I) The first two chapters set forth the basics of Moss’ account and relate it to the history of Platonic interpretation. In the preliminary status quaestionis of Chapter 1, Moss’ critical diagnosis of the main motivations for the ‘Overlap’ interpretation is followed by a survey of passages giving prima facie support to the ‘Distinct Objects’ reading[3] and by a bird’s-eye view of the history both of this reading and of objects-based epistemologies more generally.

The center of gravity of Moss’ account is the powers argument at R. V, scrutinized in Chapter 2 and judged our best window onto Plato’s epistemological doctrine. It notably shows, as Moss persuasively argues, that epistêmê and doxa, in so far as they are powers (dunameis), are [i] individuated and indeed [ii] defined by their proprietary objects, which [iii] they are uniquely suited to cognize because they resemble them (principle of like-by-like). To make her case Moss first outlines Plato’s notion of powers (54-61)[4]; then draws a salient comparison with Aristotle’s account of cognitive powers (esp. de An. B 4); and finally adduces evidence from dialogues other than the Republic. Granted the fundamental cogency of this reconstruction, some issues remain however unaddressed, e.g., if powers are individuated and defined by their objects, in what sense, if any, can ignorance (agnôsia), which relates to absolute not-being (R. 477a3-4: mê on mêdamêi), be a power at all, as the powers argument might suggest?

(II) Chapters 3 to 5 concern epistêmê and its object, Being. Moss ingeniously proposes to illuminate the notion of being by sidestepping the controversy on the senses of einai in Plato and focusing just on ontological superiority as the only notion in Plato’s conception of being relevant to the description of epistêmê. This notion, she asserts, ‘Plato takes for granted without explicitly theorizing’ (95), as a primitive concept ‘not needing explanation’ (101) and best captured by the adverb ‘really’. Thus, while Moss does not refrain from suggesting that Plato might have in mind the idea, inherited from the Presocratics, that ‘the real is the fundamental’, she is unapologetically reluctant ‘to attribute to Plato any very developed view’ (102). This hesitation is, I believe, unnecessary. The dialogues offer substantive insights into such fundamentality (some Moss discusses more in detail – e.g., stability and permanence; others less – e.g., being a cause) and a greater effort to integrate them would have been fruitful. No need of a full interpretation of Plato’s metaphysics (which Moss legitimately declines to defend), but the more Moss (rightly) insists on the relativity of cognitive powers to their correlates, the more informative her description of the latter’s ontological status should be. And this holds especially because Moss’ account proves, throughout Chapters 4 and 5, a helpful guide for the elucidation of the various features Plato assigns to epistêmê (explanatoriness, stability etc.).

(III) Chapters 6 to 8, perhaps the most successful chapters in the book, focus on doxa and its object, Seeming, i.e., the realm of the inconstant variety of perceptible appearances (also oft-termed ‘Becoming’). Moss contends that Plato’s doxa is best understood as a form of atheoretical dream-like thought applying exclusively to such appearances which, being immediately available, we tend to accept as real, without realizing that they are but defective copies of true Being.[5]

(IV) Chapters 9 and 10 propose to track ‘anticipations and echoes in the earlier and later dialogues’ (51) of the basic conception of epistêmê and doxa. Moss argues (ch. 9) that, despite the absence from the earlier dialogues of a ‘full-blown theory of Forms’ (208), the essentials of that conception are already discernable, but with a major exception, due to the absence of a robust notion of separation at this stage: Plato does not restrict epistêmê to Being and doxa to Seeming yet. Hence, the early Plato is a moderate overlapper (217). However, one might wonder whether the earlier dialogues really need to make room for such an exception. For [i] the Phaedo’s introduction of the theory of Forms as familiar (76d7-8, 100b1-3) and of the doctrine of anamnesis with a clear reference to the Meno’s slave episode (73a7-b2) suggests that the object of reminiscence is identical in both dialogues and that Plato saw (and wanted his readers to see) continuity where modern interpreters have spotted breaks. [ii] Euthphr. 6d-e already speaks of eidos/idea as paradeigma, and the Phaedo’s causal account of Forms (99d4-102a6) has parallels in the Hippias Major (combine 286d8, 289c3-d4, 296e8-9 and 300a9-b2)[6]; [iii] the claim at Euthphr. 5d1, La. 191e10-192a2, Men. 74a9 etc., that Forms are in or through the sensibles is no counter-evidence, because Plato stresses it even where separation takes the spotlight (e.g. Phd. 102b5-6, 103b8). [iv] ‘Plato is not under an obligation, every time he mentions Forms, to drag in a checklist taken from the Phaedo and affirm which of the features attributed to Forms there are true of Forms here’.[7] The case regarding Forms in the earlier dialogues has been made by many,[8] and these are only some of the reasons in its support. Moss’ admission of her limited engagement in ch. 9 with other readings (207) does not detract from the fact that she here reworks an account that rests on a premise which, however widespread, is questionable.

Later echoes of the basic conception of epistêmê and doxa are in fact explored only in the Theaetetus (ch. 10). On Moss’ engaging reading, which space limits prevent me from discussing here, the epistemology of this dialogue is compatible with that of Republic or Phaedo, but for reasons different from those traditionally adduced by Shorey, Cornford, Cherniss and others. Some new ground is probably broken here.

Two broader concerns now. First, while the focus on the epistêmêdoxa contrast is impeccably sustained throughout and gives the book a perspicuous structure, one would expect from a work on Plato’s epistemology a more comprehensive and systematic coverage of themes. Noteworthy omissions include: the doctrine of anamnesis,[9] the method of hypothesis and Plato’s views on dialectic more generally.[10] Relatedly, the insufficient attention to several later dialogues is perplexing. Surely Moss’ case would be strengthened by (and probably demands) some account of the verdict at Prm. 135c2 that the renunciation of Forms would destroy tou dialegesthai dunamis[11] altogether; or, more importantly, of the Sophist’s notion of the interweaving of Forms (the condition of logos for us (259e5-6)[12]), and the related description of the ‘science of dialectic’ (tês dialektikês epistêmês) at 253d-254b.

Second, the limited consideration of non-Anglophone literature[13] is unfortunate, especially since the scholarly view Moss convincingly combats has proved there much less fashionable and has already been variously criticized, often with arguments not so dissimilar (both in content and spirit)  to Moss’.[14] These reservations notwithstanding, Moss deserves the genuine gratitude of any serious Platonic scholar for a stimulating, rigorous (and to me largely convincing) volume,whose results will surely have a considerable impact on future research and command respectful attention.[15]


[1] Now collected in her two volumes  Plato on Knowledge and Forms (Oxford 2003) and Essays in Ancient Epistemology (Oxford 2021).

[2] J. Annas, An Introduction to Plato’s Republic, Oxford 1981, 194.

[3] R. 477e-478b, 508d, 509d, 534a, 507b-c, 529b-c; Ti. 27d-28a, 37b-c, 52a; Phd. 79a-d, 83d-84a; Phdr. 247c; Phlb. 58a, 59a-d.

[4] Surprisingly, the only two comprehensive treatments of Plato’s views on dunamis are entirely disregarded: cf. J. Souilhé, Étude sur le terme ΔΥΝΑΜΙΣ dans les dialogues de Platon, Paris 1919; and D. Lefebvre, DynamisSens et genèse de la notion aristotélicienne de puissance, Paris 2018, 183-344 (the latter perhaps too recent to be considered by Moss).

[5] Alleged evidence for doxa of Forms Moss sensibly explains away in §6b as passages ‘loose, awkward, or […] ambiguous’ (192).

[6] This should ring a bell for the interpretation of Men. 98a3-4.

[7] R. Barney, Names and Nature in Plato’s Cratylus. New York 2001, 151 n. 9.

[8] E.g., Shorey, Cherniss, Baltes, Kahn, Fronterotta, Gerson.

[9] With the Phaedo’s puzzling insistence (e.g. 66b-67a) that knowledge of Being is only possible for a disembodied soul.

[10] The very notion of dialectic is rarely mentioned (mostly in §7.6), though Plato famously calls it, among else, alêthestate gnôsis (Phlb. 58a4-5) and a dunamis (R. 511b4, 532d8, 533a8, 537d5; Prm. 135c2; Phlb. 57e7) whereby one can determine the being of each thing (e.g. R. 532a6-7, 533b2-3, 534b3-4).

[11] See previous note. Prm. 132b-c is also relevant.

[12] Sph. 257c7-d2 is also key to Moss’ point about the relativity of epistêmê.

[13] In a bibliography of 157 titles, we only have one in French (Lafrance) and four in German (Snell, Sprute, Zeller and Szaif) – with Lafrance, Snell and Zeller each quoted just once in passing.

[14] Cf. e.g. A. Graeser, ‘Platons Auffassung von Wissen und Meinung in ‘Politeia’ V’, Philosophisches Jahrbuch, 98 (1991), 365-388; C. Horn, ‘Platons episteme-doxa Unterscheidung und die Ideentheorie’, in O. Höffe (ed.), Platon:Politeia, Berlin 1997, 291-312; F. Ferrari, ‘Conoscenza e opinione: il filosofo e la città’, in M. Vegetti (ed.), Platone. La Repubblica, IV, Napoli 2000, 393-419 (and various other articles). Among older works, Moss would have found much food for thought also in L. Robin, Les rapports de l’être et de la connaissance d’après Platon, Paris 1957.

[15] Research for this review has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (Grant agreement No. 885273).