Knowledge and Truth in Plato Stepping Past The Shadow of Socrates Catherine Rowett Full Chapter PDF
Knowledge and Truth in Plato Stepping Past The Shadow of Socrates Catherine Rowett Full Chapter PDF
Knowledge and Truth in Plato Stepping Past The Shadow of Socrates Catherine Rowett Full Chapter PDF
https://ebookmass.com/product/philosophy-and-autobiography-
reflections-on-truth-self-knowledge-and-knowledge-of-others-
christopher-hamilton/
https://ebookmass.com/product/models-of-leadership-in-plato-and-
beyond-dominic-scott/
https://ebookmass.com/product/insurgent-truth-chelsea-manning-
and-the-politics-of-outsider-truth-telling-lida-maxwell/
https://ebookmass.com/product/in-search-of-truth-sharon-wray-3/
In Search of Truth Sharon Wray
https://ebookmass.com/product/in-search-of-truth-sharon-wray-2/
https://ebookmass.com/product/in-search-of-truth-sharon-wray/
https://ebookmass.com/product/vow-of-the-shadow-king-bride-of-
the-shadow-king-book-2-sylvia-mercedes-3/
https://ebookmass.com/product/vow-of-the-shadow-king-bride-of-
the-shadow-king-book-2-sylvia-mercedes-2/
https://ebookmass.com/product/vow-of-the-shadow-king-bride-of-
the-shadow-king-book-2-sylvia-mercedes/
OUP CORRECTED PROOF – FINAL, 03/07/2018, SPi
Knowledge and
Truth in Plato
Stepping Past the Shadow
of Socrates
Catherine Rowett
1
OUP CORRECTED PROOF – FINAL, 03/07/2018, SPi
3
Great Clarendon Street, Oxford, OX2 6DP,
United Kingdom
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
© Catherine Rowett 2018
The moral rights of the author have been asserted
First Edition published in 2018
Impression: 1
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above
You must not circulate this work in any other form
and you must impose this same condition on any acquirer
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 2017962211
ISBN 978–0–19–969365–8
Printed and bound by
CPI Group (UK) Ltd, Croydon, CR0 4YY
OUP CORRECTED PROOF – FINAL, 03/07/2018, SPi
Preface
vi Preface
putting forward proposals that I consider to be serious contenders for solving or clari-
fying these questions about knowledge, conceptual competence, and truth; but the
proposals are also intended to be serious contenders for clarifying what Plato himself
was talking about, and I aim to improve our understanding of what he meant when he
referred to ‘knowledge’ (episteme), and when he talked about ‘truth’ (aletheia) and
‘being’ (ousia and to on).
Hence the main chapters of this book defend a novel reading of large parts of Plato’s
corpus. This will naturally be of interest to those who already believe that Plato’s views
are important. But I trust that it will be of interest also to some who may currently
believe that there is nothing of any interest in what Plato had to say about Forms, or
about truth, or about definitions and essences, or about science and what kind of thing
we can know. My reading is designed to undermine many of the objections to Platonism
that are typically bandied about, which are often directed against some caricature of
Plato’s epistemology and metaphysics. To show that such objections typically miss
Plato’s point, and thereby miss a range of promising answers to current topics of enquiry,
I have included substantial chapters that explore key texts from some of the most fam-
ous dialogues, reinterpreting them to show that Plato probably means something quite
different from what he has standardly been taken to mean. There is also more of this
work to come, on texts not included in this book, but here we make a start towards
developing these claims, defending them with close attention to some prominent texts.
My aim is to show that we should not try to reduce conceptual knowledge to some
other kind of knowledge, as though the list of basic kinds of knowledge were complete
at ‘know-how’, ‘acquaintance’, and ‘propositional’. Even less should we try to reduce one
or more of those three to one among them that is supposedly more fundamental. I shall
argue that there is a fundamental difference between knowledge of types and know-
ledge of tokens, and that all the kinds of knowledge that consist in seeing tokens as fall-
ing under types are parasitic on a more fundamental kind of knowledge—grasping the
type in question—which is not reducible to anything else, because all our knowledge of
facts, propositions, actions, and things presuppose a prior grasp of types. Whenever we
see some item as belonging to a class or deserving a certain description, we call upon a
repertoire of conceptual knowledge which—as I shall argue—cannot be reduced to any
finite set of propositions and is often not even expressible in propositional form; nor is it
reducible to any finite rule for how to continue a certain practice consistently (since the
practice may be creative, inconsistent, impromptu, and innovative).
Any agent capable of using language or classifying and reading the world draws upon
a repertoire of available descriptors (maybe linguistic, maybe practical/pragmatic)
whose meaning and relevance to the task at hand the agent grasps, sometimes in a
wholly inarticulate way. This repertoire enables conceptually equipped agents to see
things as instantiating types or kinds,1 and to extend the concepts indefinitely, to
1
Among conceptually equipped agents I include all animals that see the world in terms of kinds rele-
vant to their tasks and can engage with their environment on that basis.
OUP CORRECTED PROOF – FINAL, 03/07/2018, SPi
Preface vii
embrace and redescribe new examples in an intelligent and creative way, rather than
mechanically. Extensions and innovations are clearly not pre-programmed, since they
can be controversial, a subject of debate and uncertainty, and innovations that were
once unacceptable can become accepted.
Obviously we must possess a certain idea before we can innovate with it. But what is
that knowledge, such that someone who possesses it can recognize familiar cases,
handle unforeseen possibilities unfazed, apply the descriptor intelligently beyond its
normal application and in metaphorical or strangely reconfigured circumstances, and
reject it in other cases that are too alien? How is it that on the basis of our idea of justice,
which we may never have seen properly instantiated anywhere, we can discern justice
and injustice in situations never before envisaged, and discuss cases that might be
unclear, ruling that they are or are not examples of justice, or that they are marginal or
irrelevant to judgements of that kind? For an adequate account of this kind of know-
ledge, Plato’s model seems to me to be far superior to any propositional or definitional
or rule-based models. Plato, I shall suggest, models this knowledge as a quasi-visual
grasp of ‘justice itself ’. A key theme of this book is what Plato has to say about the
method by which we learn (or re-acquire) those ideas, and become philosophically
aware of them: namely, as I shall argue, that we acquire this knowledge by a kind of
analogical or pictorial reasoning, based on a very small sample—even just one example
which could in fact be a bad or negative example (as e.g. we may come to see what
justice is from encountering a peculiarly telling case of injustice). He correctly identi-
fies, I suggest, a kind of picturing process that enables a conceptually competent
thinker to grasp an abstract idea of great complexity and subtlety, by abstracting (or
ascending) from small and unrepresentative samples. The process is not induction, or
empirical generalization, since it can be done from one encounter alone, or from
untypical ones, or even from the total absence of any instance. It is also possible to gen-
erate an understanding (as we shall see) by constructing an imaginary instance, unlike
any known in real life.
To get hold of the idea of justice, we do not need to have encountered any sound or
exemplary cases of justice; nor do we get the idea by habit or practice, or by learning
from role models; but it is to our existing idea of justice that we turn to see what would
be the just thing to do in situations not specified in advance. This grasp of the concept
(our idea of what justice is) provides a more accurate and authoritative basis for judge-
ment than common practices that are considered just. The latter will always fall short
for various reasons, and our ability to pass judgement on those practices reveals that
we have a more authoritative grasp of the notion in ourselves: so we can criticize our-
selves and others for not doing in practice what we know in theory; and we can criticize
ourselves and others for being unable to explain or define something that we know and
can correctly deploy in practice. These discrepancies (a) between our intellectual grasp
of things like justice, honesty, etc., and our attempts to be just or honest in real life, or
(b) between our ability to act justly/honestly and our ability to explain what justice or
honesty is in words—these show that we find ourselves able to correct both our actions
OUP CORRECTED PROOF – FINAL, 03/07/2018, SPi
viii Preface
and our words by reference to something else—our knowledge of what justice and
honesty are—which, I argue, is epistemically more basic than the practices, and gener-
ally inadequately enacted in practice and inadequately expressed, or even inexpress-
ible, in words. On the basis of this fundamental knowledge of ‘what it is’ about the
relevant concepts, we can also recognize which instances are truly exemplary and
reject those that are not. Such corrective knowledge is the basis of all social, moral,
political, and artistic criticism and progress. Knowledge of facts is as nothing in com-
parison with this, when it comes to its importance for human life. It is this knowledge,
I suggest, that makes facts possible.
Thinking of this knowledge as quasi-pictorial helps us to understand how it can be
dense with meaning, yielding unlimited answers to an infinite range of unforeseen
questions. It is rather like consulting a map or drawing, in terms of the density of
its information, as contrasted with written directions that tell you just one or two
ways to go.
Preface ix
I begin each part with an ‘Introduction and Summary’, which sets out my basic claims
with respect to the dialogue under discussion, and evaluates their significance for the
overall thesis of the book. These are partly for hasty readers with no time for wading
through the exegetical material, or for non-specialist readers who are not concerned
with the detailed exegesis. Such readers can settle for reading the first chapter in each
Part of the book. But those chapters are also designed as orientation for the specialist
reader, setting the agenda and explaining the significance of each move, before
embarking on the longer chapters in which the exegetical proofs are done. The exeget-
ical chapters which comprise the rest of each Part of the book attempt to show how one
can read the text with those results, and engage in varying levels of nitty-gritty detail
with both the Greek text and with a selection of the work of other scholars, classic or
recent, who have offered rival views.
Those seven exegetical chapters (Chapters 4–5, 7–8, and 10–12) are internally sign-
posted with headings, which summarize the progress of the argument. This means that
by reading just the headings, one gets a kind of précis or list of premises and conclu-
sions. The headings are also designed to enable someone who is not reading the book
from cover to cover, but looking for its argument on some particular passage or its
solution to some tricky dilemma, to locate the argument and easily identify what my
view on the issue is. The index locorum can also assist with this. Wherever possible, my
proposed answer is stated in the heading.
The book concludes with a chapter that is neither exegetical nor evaluative, but more
an outline promise of things to come, since there are many further texts that would
be relevant.
Contents
Acknowledgements xix
xii Contents
Contents xiii
xiv Contents
II. In Which We Examine the Text More Closely, to Reconstruct the Hunt
for the City’s Justice in Republic 4, and We Discover No Definition 118
II.i That the hunting in Passage A is a hunt for the city’s justice, not for
justice as such; II.ii That the method and the finding are comparable to
finding the line on the diagram in the Meno, and to the things admired by
the lovers of sights and sounds; II.iii That this ought to be a bad answer to
the ‘what is justice?’ question, by Socrates’ own previous standards;
II.iv Ontological interlude: Forms and tropes
III. In Which We Consider How Socrates Is Able to Move Forward From
Identifying the Justice of a Particular City, to Grasping What Justice Is 123
III.i That Socrates uses the analogy of rubbing sticks together to explain
his method; III.ii That there are other passages that explain the method;
III.iii That the ambitions of the method are not to define justice
IV. In Which We Consider David Sachs’s Objection, That There Is a Fallacy of
Equating ‘Platonic Justice’ With ‘Vulgar Justice’ 128
IV.i That the shift to the inner disposition is already explained in Adeimantus’s
challenge in Book 2; IV.ii Three ways to read the passage about temple-raiding in
Republic 4, of which the third is neither fallacious nor reductionist
V. In Which We Consider the Accusation From Bernard Williams That the
Analogy Between Soul and State Will Not Support Socrates’ Desired
Conclusions 134
V.i That Williams is assuming (a) that the soul/state comparison is an
analogy and (b) that Plato is pursuing an essentialist agenda; V.ii That
Williams constructs a story of crisis, and attempted, but ineffective, solution,
which is of his own making; V.iii That a better interpretation is possible,
if we avoid attributing any essentialist or reductionist moves to Plato
VI. Conclusion 140
8. Platonic Method: The Philosopher’s Route to Knowledge
in Plato’s Republic142
I. In Which We Juxtapose Socrates’ Comments About Short Versus Long
Routes and About Outlines Versus Finished Drawings, to See Why
Socrates Is Employing a Method That Is Not the Best 142
I.i That Socrates identifies, and follows, a shortcut method of enquiry
that is not good enough for the Guardians, but is enough for now;
I.ii That Socrates never embarks upon the longer way; I.iii That Socrates
takes a route that is third best, in the Sun analogy; I.iv That the Republic
aims at no more detail than is required for the target question, and
that this is what is meant by contrasting outlines with finished works
II. In Which We Consider the Implications of 505a–506d, Where Socrates
Rejects Two Candidate Definitions of the Good, Indicating a General
Problem for the Definitional Project 148
III. In Which We Consider the Role of Icons in the Divided Line 150
III.i That the Line is set up to explain why shadows are epistemically
valuable, and informative, for philosophical enquiry; III.ii That the same
OUP CORRECTED PROOF – FINAL, 03/07/2018, SPi
Contents xv
xvi Contents
I.i That the search for a definition of episteme must fail, and this is partly
because Theaetetus is too young
II. In Which We Consider Whether Listing Examples Could Be a Good
Answer, and Discover Why Socrates’ Analysis of Clay Is an Unhelpful Model 185
II.i That mentioning examples is part of the iconic method, and that Socrates’
objections to the examples, and his recommended kind of analysis, are not
the usual ones, but are nonetheless misguided; II.ii That Theaetetus’s three
definitions of episteme are all designed to follow the model of a simple
compositional analysis, which is a faulty model and leads to failure;
II.iii That there is nothing wrong with using the term that is to be defined,
and that Socrates is confused on that score
III. In Which We Briefly Consider Theaetetus’s Proposal That Science Is
‘Perception’: Whether It Is a Good Suggestion and Why It Fails to Thrive 192
III.i That Theaetetus conceives his proposed definitions by the iconic method,
generalizing from his experience of geometry as a science, and that his first
proposal has some merits, since knowledge of concepts is a bit like perception;
III.ii That Socrates respects the standard constraints on what can count as
science, in developing his support package for EA
IV. Conclusion: That Geometry Invites the Thought That We Perceive
the Intelligible Types in the Diagram, and That No Other Scientific
Knowledge Is Presupposed 195
11. The Division Between Sense Perception and Non-Sensory Doxa in the
Interlude: Theaetetus 184a–187b 197
I. In Which We Clarify the Meaning of Doxa and Doxazein in the Rest
of the Dialogue 197
I.i That ‘believe’ or ‘judge’ are not good translations for D, and why not;
I.ii That when Socrates lists ‘being’ as (hypothetically) one of the features
accessible to D but not SP, he does not mean propositional form
II. In Which We Take Issue With Some Classic Interpretations of the
Interlude and Their More Recent Descendants 201
II.i That there are four ways of reading the reference to ‘being’, and that
the most popular reading takes it as marking a feature of propositions;
II.ii That many interpretations cobble together Reading C and Reading P,
sometimes with other interpretations as well, in trying to make sense
of Plato’s text
III. In Which We Embark on a Reading of the Interlude and Note That Socrates
Distinguishes Two Faculties Equipped to Detect Non-Propositional
Features, One With, and One Without, the Use of Bodily Organs 208
III.i That Socrates explains his distinction between SP and D by giving
lists of paradigm cases; III.ii That the argument does not assume that all
sensibles are special sensibles, accessible to only one sense; III.iii Whether
we see with our eyes or with our souls, and why stipulating some technical
terminology helps the argument here (but has nothing to do with correcting
OUP CORRECTED PROOF – FINAL, 03/07/2018, SPi
Contents xvii
xviii Contents
Bibliography 277
Index 287
Index Locorum 301
OUP CORRECTED PROOF – FINAL, 03/07/2018, SPi
Acknowledgements
I received generous funding to support this research from the Leverhulme Trust and
the Arts and Humanities Research Council, as well as two semesters of research
leave from the University of East Anglia. The University of Aberdeen Philosophy
Department provided a visiting fellowship, excellent facilities, and collegial conversa-
tion in autumn 2008, and Merton College Oxford provided accommodation and colle-
gial facilities in autumn 2011. I would like to thank Gail Fine in particular for inviting
me to Merton, and for helpful conversations both then and at other stages of my
research. In 2009 Oxford Brookes University provided a ten-day visiting fellowship,
with accommodation and opportunities to present two seminar papers. On three
occasions I have enjoyed the hospitality of the Fondation Hardt in Geneva, with its
fine Classical library and other resources provided by the Geneva University libraries,
and the welcome opportunities to discuss ideas with other Classical scholars from
around Europe.
I have benefited from advice from three readers who reviewed sample chapters in
2010, and two readers who read the complete manuscript in 2012. I am especially
grateful to Sophie-Grace Chappell for comments on the manuscript at that stage.
I have tried, not always successfully, to pay proper heed to the excellent advice from all
these readers, and the manuscript has gone through several major revisions as a result.
I hope that it is now better, not worse.
I developed ideas from Chapters 1 and 2, and some work on the Sophist (not
included here in the end), in conversation with colleagues in Aberdeen (including
Patricia Clarke), in Cambridge (the D Society), and at Oxford Brookes University.
Chapters 3 and 4 contain material presented at Trinity College Dublin, the Nordic
Wittgenstein Society in Uppsala, and the Ancient Philosophy Workshop in Oxford.
Chapters 5 and 6 contain ideas discussed at the Southern Association for Ancient
Philosophy in Oxford, the Ancient Philosophy reading group in Cambridge, Oxford
Brookes University, the University of East Anglia, and the Open University. Chapters
7, 8, and 9 include material that I have presented in Toronto and in Bordeaux (the
Celtic Conference). At a reading of the Theaetetus in Dublin in 2013, I learned much
from discussion with Brendan O’Byrne, David Horan, and John Dillon, and from
Vasilis Politis, who had kindly invited me. I am also grateful to generations of under-
graduates at Swansea and the University of East Anglia (including my Special Subject
class of 2013) with whom I tried out several ideas.
OUP CORRECTED PROOF – FINAL, 03/07/2018, SPi
“How can it be right to say that states will never end their troubles until philosophers
hold the sway in them, when we’ve agreed that the philosophers are useless to them?”
“Your question is one that needs to be answered by way of an icon,” I said.
“And you, of course, have never been known to talk through icons, I think!” said he.
“Bah! First you throw me an unfair topic, and then you mock the afflicted! But
here’s my icon . . . ”
Plato, Republic 487e (introducing the ship of state image)
OUP CORRECTED PROOF – FINAL, 03/06/2018, SPi
PA RT I
Knowledge, Truth, and Belief
OUP CORRECTED PROOF – FINAL, 03/06/2018, SPi
OUP CORRECTED PROOF – FINAL, 03/06/2018, SPi
1
Knowledge, Conceptual
Knowledge, and the Iconic Route
to Grasping an Idea
and what it is about, and what kind of truth it captures is not so evident. These seem
to be questions to which Plato was perpetually drawn. It seems likely that he never
answered them to his satisfaction, but that his explorations in some of his most famous
dialogues are among the best treatments of the issues known to this day. They raise
virtually all the puzzles that a good solution would need to address. Indeed, if I am
right, there are no recent treatments that do justice to the exploratory proposals that
Plato is investigating here, partly because his proposals have not often been recognized
as a contribution to this field, and when they have, they have sometimes been ham-
mered into alignment with the reductionist projects of the present day. Some of his
proposals, such as the theory of ‘Forms’, even though they are not presented in Plato’s
own voice, and are not presented uncritically as a problem-free account or a finished
answer, do at least take seriously the desiderata and criteria for success for explaining
this knowledge: namely that we need the content of conceptual knowledge to be some-
thing that is rich with quasi-pictorial density and not reducible to a formula, definition,
or rule of thumb.
suite of terms with the same implicit contrasts).1 In 1981, Lyons retracted his 1963
conclusions, and asserted that after all it is possible to align the triad of terms in Greek
quite successfully with analytic philosophy’s ‘troika’ of knowledge types (knowledge
that, knowledge how, and acquaintance).2 Myles Burnyeat had spent much of his early
career urging the importance of Lyons’s 1963 work, and particularly the mismatch
between episteme (paired with eidenai) and epistasthai (paired with techne), and
deploying the findings from Lyons’s Structural Semantics in his own studies of the
Theaetetus and other works. Recently Burnyeat has reiterated this continuing belief
in the 1963 work, and questioned whether Lyons should have retracted.3 It is true
that many of Burnyeat’s favourite observations survive the retraction, and I would
agree with Burnyeat that it is very unwise to treat analytic philosophy’s epistemic troika
as if it were a complete and definitive taxonomy of knowledge that is uniform across
cultures.4 Arguably my thesis in this book is that the troika omits the one that was
Plato’s most central case.
However, I think that there was something wrong with Lyons’s 1963 approach, which
has also infected much of Burnyeat’s work. It privileged a certain kind of contextual
semantics, and understood the semantics of a term to be fixed in all uses, but context-
ual in this sense: that the term’s meaning was determined by its relation to other terms
in the writer’s vocabulary (whose meaning was also fixed by that implicit contrast and
comparison). So the way to discover what a term means in Plato’s vocabulary was, on
that approach, a matter of discovering a consistent rule that identified which word is
invariably implying a contrast with which, and assigning a single constant meaning to
every word in every occurrence. What Lyons concedes (correctly) in his 1981 rethink
was that these words may have focal meanings (so that stereotypically, a certain term
tends to be used for a certain kind of knowing, even though others can also be used,
slightly outside their normal range). This is an improvement, though his attempt to align
the focal usages of the three terms with some modern focal senses of ‘know’ may be
seriously unhelpful, as we have observed. Burnyeat, by contrast, is still rather inclined
to the one-word one-meaning methodology, and to the attempt to use the supposedly
fixed structure of relations between these knowledge-terms to prescribe a meaning,
rather than looking to the immediate context in the text to see whether it is being used
in the way prescribed by the Lyons theory.
1
John Lyons, Structural Semantics: An analysis of part of the vocabulary of Plato (Oxford: Blackwell,
1963).
2
John Lyons, ‘Structural Semantics in retrospect’ in Thomas Edward Hope (ed.), Language, Meaning
and Style: Essays in memory of Stephen Ullmann (Leeds: Leeds University Press, 1981), 73–90.
3
Myles Burnyeat, ‘Episteme’ in Benjamin Morison and Katerina Ierodiakonou (eds.), Episteme, etc:
Essays in honour of Jonathan Barnes (Oxford: Oxford University Press, 2011), 3–29, 12. See further below,
note 50.
4
As e.g. in Ronald Polansky, Philosophy and Knowledge: a commentary on Plato’s Theaetetus (Lewisburg,
PA: Bucknell University Press, 1992), 14 who claims that Plato’s Theaetetus covers (and succeeds with) all
known brands of knowledge. (Incidentally it is also odd to claim that it succeeds, given that every part of it
fails.) See Part IV below.
OUP CORRECTED PROOF – FINAL, 03/06/2018, SPi
My view is that the structural semantics approach was wrong, and the cross-cultural
equations are very unwise. I also find that the meanings of the knowledge words are
largely determined by immediate context, and very much less by the built-in semantic
connotations of the chosen word. There are many contexts in which Plato freely substi-
tutes one knowledge word for another with gay abandon through a passage of text. We
shall find some such passages in later chapters.5 We shall also find places where Socrates
or his interlocutor assigns technical senses to words that had more flexible meanings
elsewhere,6 and many examples where Socrates stretches his terms, or even abuses their
conventional grammar, to raise new difficulties or make clearer distinctions where
ordinary language had not made any.7 To discover what Socrates means by a term, or
what work he is having it do in some place, one must read it carefully, in that context.
The fact that it means something different, or invokes a different contrast, in another
passage should never be invoked to undermine the most coherent reading of the target
passage. We can only discover whether there is a consistent meaning for any of Plato’s
terms by reading the text.
So I shall not take Lyons’s conclusions as gospel, neither his 1963 ones nor his 1981
ones; nor do I follow Burnyeat’s dependence on Lyons. It seems to me that Plato’s
language, and the Greek language in general, is as flexible as our own, and there are no
fixed constraints on how far one may stray from the focal meaning of a word, or on what
connotations or implicatures one may voluntarily cancel in any given context.
That said, I shall suggest that, very often, Plato uses the term episteme in a moderately
technical sense—perhaps a sense that he has newly invented, and for which he has
newly adapted an existing term. He uses it, I shall argue, specifically to denote a kind
of knowledge that is not part of our troika at all. Yet even if this is a novel usage, he
evidently thinks that this is what the term should mean: that exactly this knowledge
(the knowledge of Forms, of the what-it-is about our concepts) is what one needs to
count as wise, as having scientific understanding, as having the thing that qualifies as
episteme. He is setting episteme up as a standard, and prescribing what it is to reach
that standard. Some passages imply that the bar is high, since a lot of people seem to
fail to meet it, by turning out not to know what they thought they knew. Yet that is
not quite true in the end. It is only if you adopt the criterion that Socrates uses in the
‘Socratic’ bits of the dialogues for whether someone counts as knowing that those
people seem not to know. We shall find reason, repeatedly, to think that Plato the
author is challenging that criterion, and hence challenging the result: it turns out
that we know many things, and we have ways to discover yet more, once we discard
Socrates’ misguided demand that one who knows must be able to give a definition.
But more of that shortly.
5
See for example Chapter 3, note 21.
6
E.g. the redefinition of aisthesis and the introduction of a technical sense for doxazein at Theaetetus
186e1, 187a8 (see Chapter 11).
7
E.g. the novel grammar assigned to the term doxazein in the Theaetetus (see Chapter 11).
OUP CORRECTED PROOF – FINAL, 03/06/2018, SPi
8
See Myles Burnyeat, ‘Socrates and the Jury: paradoxes in Plato’s distinction between knowledge and
true belief ’, Proceedings of the Aristotelian Society Supplementary Volume, 54 (1980), 173–91, 186–8;
Myles Burnyeat, ‘Aristotle on understanding knowledge’ in Enrico Berti (ed.), Aristotle on Science: the
posterior analytics (Proceedings of the Eighth Symposium Aristotelicum; Padua: Ed. Antenore, 1981),
97–139, 133–6. See also J.M.E. Moravcsik, ‘Understanding and Knowledge in Plato’s Philosophy’, Neue
Hefte für Philosophie, 15/16 (1979), 53–69; Jon Moline, Plato’s Theory of Understanding (Madison, WI:
University of Wisconsin Press, 1982).
9
Burnyeat, ‘Episteme’, 8–12. See this chapter, Section I.ii.
OUP CORRECTED PROOF – FINAL, 03/06/2018, SPi
that p), whereupon adding understanding to the criteria does not seem to mark out
Plato’s idea as anything different from propositional knowledge, as currently under-
stood. Furthermore, if there is something further to know, besides the content of the
‘that . . . ’ clause—if we must also know or understand ‘why’, as Burnyeat implies—
should we take that additional knowledge ‘why’ to be knowledge of another fact—the
explanatory fact—or is it some other kind of knowledge from among the current troika
(some kind of know how or ‘knowledge-wh’, as the jargon has it)? If so, then we have not
escaped from matching Plato’s terms to contemporary analytic terms. Or is it some
additional kind of knowing that is not among those in the troika? In which case, we
want to know what it is and why it is sui generis and not reducible to one of those.
I dissent from the main thrust of Burnyeat’s proposal here, not so much because
I object to the term ‘understanding’ (since one could indeed use that term to mean
what I think Plato is talking about, though Burnyeat himself seems not to mean what
I mean).10 What I object to is (among other things) the inclusion of the factual knowledge
in the composite that Burnyeat has in mind, namely ‘knowing that p, plus understanding
why’. I disagree because I do not think that any kind of knowing facts or propositions is
ever included in episteme, as that term is used by Plato, whether or not you have some
further understanding of why they are true. Episteme just is not a knowledge of facts or
understanding of facts in the world. It is not knowledge that some particular state of
affairs obtains. Indeed, it does not relate to the particular at all.
However, I do take episteme to be a certain kind of understanding, namely conceptual
understanding—understanding a certain notion or idea, so as to be able to use it correctly
and explain it to someone else (as, for instance, in Wittgenstein’s famous example of
‘knowing what a game is’, and Socrates’ similar quest for knowing what a virtue is
or knowing what justice is).11 Clearly we can and do use the word ‘knowledge’ for this
(as Wittgenstein’s discussion, or rather the English translation of it, shows). We speak
of ‘knowing what a game is’, and of someone who doesn’t understand what games are as
‘not knowing’. It is a mistake to think that our term ‘knowledge’ means just knowledge
of facts, or ‘knowing that . . . ’, or that anything that deserves the name must be reducible
to something of that kind.12
Hence I remain sympathetic to some aspects of Burnyeat’s 2011 paper, in particular
his insistence that we should not come with some ready-made system of ‘kinds of
knowledge’ and see if Plato has got it right, but rather should invite analytic philosophers
10
The term ‘understanding’ is used as a translation for verstehen in Wittgenstein (e.g. Ludwig Wittgenstein,
Philosophische Untersuchungen, trans. G.E.M. Anscombe (Oxford: Basil Blackwell, 1953), §146) (henceforth PI).
The example is understanding how to continue a mathematical series, and belongs to a discussion of
rule-following as a way of thinking about conceptual knowledge. This is a text about the kind of knowledge
I have in mind (but it is not the only kind that can be called ‘understanding’). Arguably Wittgenstein was
inclined to reduce it to ‘knowing how’, a knowledge manifested in behaviour with no remainder.
11
See Chapter 4, and Catherine Rowett, ‘Plato, Wittgenstein and the definition of games’ in Luigi
Perissinotto and Begonia Ramon Camara (eds.), Wittgenstein and Plato (Basingstoke: Palgrave Macmillan,
2013), 196–219.
12
See further this chapter, Section IV.
OUP CORRECTED PROOF – FINAL, 03/06/2018, SPi
to be challenged by something that does not fit their categories. I also concur with his
sense that knowledge of the Forms is not correctly captured in terms of ‘knowledge by
acquaintance’, as traditionally conceived. I disagree with him mainly for the straitjacket
from semantics that he uses to constrain the philosophical sense. I prefer to explore
threads of topic-based enquiries across the Platonic texts and draw conclusions about
the semantics from the philosophical results. I also disagree with some of Burnyeat’s
conclusions about knowing ‘what it is’, because he seems to think that what is known
in, say, ‘knowing the good, what it is’ is the answer to an indirect question, conceived as
a proposition.13
13
Burnyeat, ‘Episteme’, 7.
14
See especially Chapters 5, 11, and 12. On the relation between beliefs and credence (for which Plato’s
term is pistis), see Chapter 2, Section I.ii.
15
See Chapters 5 and 12.
16
Timothy Williamson, Knowledge and its Limits (Oxford: Oxford University Press, 2000).
OUP CORRECTED PROOF – FINAL, 03/06/2018, SPi
need not be attempts at a successful compositional analysis of knowledge (as true belief
plus something). They may be showing that such an analysis would be unsuccessful.
This is the line taken by Lloyd Gerson in recent work.17
17
Especially Lloyd Gerson, Ancient Epistemology (Key Themes in Ancient Philosophy; Cambridge:
Cambridge University Press, 2009). Cf. Lloyd Gerson, Knowing Persons: a study in Plato (Oxford: Oxford
University Press, 2003).
18
By ‘mainstream’ I mean not including the tradition of Leo Strauss or the Tübingen School. On the
difference between my approach and that of the Straussians, see Section II.ii.
19
Timothy Chappell, Reading Plato’s Theaetetus (International Plato Studies; Sankt Augustin: Academia
Verlag, 2004), 16–21 gives a survey of this story as a choice of two models, unitarian or revisionist. He rightly
observes that the revisionist readings were motivated by charity, especially to address their antipathy to the
Forms. Naturally (though he does not explicitly say so), the unitarian readings, including his own, are also
motivated by charity.
OUP CORRECTED PROOF – FINAL, 03/06/2018, SPi
Plato’s, to explain how we grasp truths that are not mere human constructs. For such
readers, Plato becomes the hero in a project to reassert Platonism as a genuine philosoph-
ical position with explanatory power. Readers who take this view argue that Plato’s
texts should be read as sophisticated explorations either of the advantages of Platonism,
or of the problems generated by alternative views (such as pragmatism, and coherence
theories of truth). This tends to go with what we call a ‘unitarian’ reading of Plato—that
is, claiming that Plato adhered to the same model of knowledge throughout his career,
and always maintained that the Forms exist in reality, as the paradigmatic objects
of knowledge.
On such a view, there is no suggestion that Plato went on to reject that realist model
of knowledge in the Later dialogues (since that would be to suppose that he abandoned
a good theory in favour of a less good one). Rather, it is assumed that the Later dialogues
develop the same position on epistemology and metaphysics as the Middle-Period
dialogues, though offering support for it in new and subtle ways. F.M. Cornford’s Plato’s
Theory of Knowledge is the classic example of such a reading.20
This way of reading Plato’s Later dialogues went out of fashion for much of the second
half of the twentieth century, but there is evidence of a trend towards rehabilitating it,
going back at least to the 1970s, in, for instance, the work of Nicholas White, and more
recently, Lloyd Gerson and Timothy Chappell in the first decade of the twenty-first
century.21
When this reading of Plato’s later epistemology is applied to the Theaetetus, especially
to the first part of that dialogue, it is often called ‘Reading A’, following Myles Burnyeat.22
I shall call it the Metaphysical Reading, because its chief object is to secure a realist
metaphysics, with transcendent Forms as the objects of knowledge.
Republic, Plato came to see faults in his earlier theories and revised them radically: the
revisions are already evident in the total absence of Forms from the Theaetetus, but further
improvements are identified even after the Theaetetus, particularly in the Sophist.
This developmental way of reading Plato was (perhaps not accidentally) prompted,
or at least encouraged, by Gilbert Ryle’s influential work which dreamed up a story
about Plato’s ‘progress’, according to which Plato came to realize that all his famous
‘theories’ from the Middle Period were muddled. Instead, he became a sound Rylean
in the Later Period.23 Readers who took this developmental line could adhere to the
dogmas of early twentieth-century analytic philosophy and still hold Plato up as a
great thinker, because they found that Plato himself had been wise enough (in his
maturity) to abandon the realist structures that he had foolishly proposed in his youth,
and adopt the dogmas of contemporary analytic philosophy instead.
It is often said that Ryle’s work on Plato was not very influential.24 It is true that some
of its detailed speculations were treated with suspicion, but his way of explaining
Plato’s development has become deeply entrenched in all subsequent work. What may
be closer to the truth is that Ryle’s approach was already in circulation by way of his oral
teaching, long before the publication of the 1966 book, so the effect of his published
book was less than the very extensive effect of his approach. Developmental readings
of this Rylean kind were pretty ubiquitous throughout the middle and latter part of
the twentieth century (again being somewhat challenged by more recent work from
various quarters).
On this developmental story, Plato’s thinking in the Later dialogues, such as the
Theaetetus and Sophist, is supposed to be fully compatible with the idea that knowledge
takes propositions as its content, that truth is a property of propositions, that the world is
structured by language, and that there is no need for disembodied souls or non-linguistic
access to reality. Readers in this tradition find in the Theaetetus the idea that there can be
knowledge of sensible particulars, not just Forms; the idea that truth involves complex-
ity, in particular the predication of one thing of another, and always has propositional
form; that knowledge, if it is of the truth, will also have propositional form; that one
might try to define knowledge as a kind of true belief, rather than radically alien from
belief (as in Middle-Period Plato); that logical atomism is an attractive but ultimately
self-defeating account of how knowledge and meaning are grounded; and that there is no
need for pre-existent souls, or out-of-body experience, since the mind can be stocked
with materials during its everyday embodied experience among particulars.
This second way of defending Plato seems to have been the dominant tradition since
at least the 1970s,25 and can be found in the work of Myles Burnyeat,26 David Bostock,
23
Gilbert Ryle, Plato’s Progress (Cambridge: Cambridge University Press, 1966).
24
But see Myles Burnyeat, ‘Gilbert Ryle (1900–76)’ in Robert B. Todd (ed.), Dictionary of British
Classicists (3; Bristol: Thoemmes Continuum, 2004), 846–9.
25
It was clearly motivating some of Gwynneth Matthews, Plato’s Epistemology and Related Logical Problems
(London: Faber and Faber, 1972), 13–23, who struggles with it somewhat in passages where she recognizes that
Plato is not speaking of knowledge of singular propositions, but of kinds or the complex relations of kinds.
26
Burnyeat, Theaetetus and many preceding articles.
OUP CORRECTED PROOF – FINAL, 03/06/2018, SPi
David Sedley (with qualifications),27 and perhaps G.E.L. Owen.28 When applied to the
Theaetetus, this way of understanding Plato’s later epistemology is called ‘Reading B’ by
Burnyeat.29 I call it the Rylean Reading because it sees Plato as progressing away from
answers that appeal to metaphysics towards answers provided by the philosophy of
language, and because it finds in later Plato an embryonic awareness of what Ryle took
to be truths discovered by early analytic philosophy in the wake of Frege, Russell, and
Wittgenstein.
27
David Sedley, The Midwife of Platonism: text and subtext in Plato’s Theaetetus (Oxford: Clarendon
Press, 2004); David Bostock, Plato’s Theaetetus (Oxford: Clarendon Press, 1988). Sedley thinks that Plato
saw himself as continuing the same project throughout (Sedley, The Midwife of Platonism, 13–15).
However, he thinks that Plato endorsed the Socratic approach in his early work, and that in the Theaetetus
the text serves as midwife, to deliver a Platonic brainchild that Socrates himself could not conceive, which
allows the text to elicit from the reader an answer not explicitly offered by Socrates, which is Plato’s answer.
So although he finds continuity between the Socratic (pre-Forms) project and the Platonic one, Sedley is
closer to the Rylean reading on the issue of whether there is ‘progress’ towards an ‘improved’ understanding
of the propositional structure of thought and knowledge, and towards the ‘correct’ diagnosis of false belief
as propositional. Like Frede, Sedley considers that some puzzles are still inadequately addressed in the
Theaetetus, and that Plato finds a correct solution in the Sophist (Sedley, The Midwife of Platonism, 119).
28
G.E.L. Owen, ‘Plato on not-being’ in Gregory Vlastos (ed.), Plato I: metaphysics and epistemology
(Garden City, NY: Doubleday, 1970), 223–67 does not much discuss the Theaetetus, but seems to think it
still fairly benighted, though perhaps only about the relation between reference and meaning in sentences
(see Owen, ‘Plato on not-being’, 245).
29
See note 22. Burnyeat’s two options are not entirely satisfactory: see, for instance, some criticisms in
Lesley Brown, ‘Understanding the Theaetetus’, Oxford Studies in Ancient Philosophy, 11 (1994), 199–224,
209–13.
30
See particularly Fine’s articles collected in her volume of reprints, Gail Fine, Plato on Knowledge and
Forms (Oxford: Clarendon Press, 2003).
OUP CORRECTED PROOF – FINAL, 03/06/2018, SPi
31
His work on the Sophist is mainly contained in an early work, Michael Frede, Prädikation und
Existenzaussage (Hypomnemata, 18; Göttingen: Vandenhoeck und Ruprecht, 1967), together with a
more accessible article from later in his life, Michael Frede, ‘Plato’s Sophist on false statements’ in
Richard Kraut (ed.), The Cambridge Companion to Plato (Cambridge: Cambridge University Press, 1992),
397–424. Michael Frede, ‘Observations on perception in Plato’s later dialogues’, Essays in Ancient Philosophy
(Mineapolis, MN: University of Minnesota Press, 1987), 3–8 includes a very sketchy discussion of perception
and belief in the Theaetetus.
32
Frede, ‘Observations on Perception’.
OUP CORRECTED PROOF – FINAL, 03/06/2018, SPi
he still has Plato saying some unfashionable things about perception and metaphysics,
retaining Forms in the late period, and invoking a special ‘itself in itself ’ use of the verb
‘be’ for saying what the Form is in itself, and in other ways seeks to show how alien the
material is, even in the late dialogues, to our way of thinking and to the distinctions
between senses of ‘to be’ so beloved of analytic interpreters. This alienating ambition
aligns him with the Metaphysical party, who would maintain that we ought to learn
from Plato’s unfashionable and challenging ways of addressing his problems. Some
of Frede’s thoughts, and his alienating ambitions, are quite close to what I am arguing
for here.
John McDowell is also slightly hard to place in my scheme. This is partly because
his major work on the subject is a commentary on the Theaetetus which lacks any
introduction or extended essay on any of the issues, so his overall view is hard to extract
from his intermittent comments on problematic passages. His position is evidently
something like the Rylean one, in that he thinks that Plato makes progress, and that the
later works abandon certain ways of thinking that were characteristic of Middle dialogues.
Parts of his commentary imply that Plato was (by now) equating truth and being with
propositional structure, and he thinks that Theaetetus 184–6 delivers a sense-datum
theory for perception; also he thinks that the arguments and distinctions that Plato
musters in discussing whether knowledge might be defined as true judgement ought
to have committed Plato to a propositional or quasi-propositional structure for know-
ledge if Plato had been sufficiently clear about what he was doing.33 Yet at the same
time, he emphasizes at relevant points that Plato thinks of his topic as ‘knowledge of
things’,34 concluding that Plato was at least somewhat confused by this, and by failing
to realize that ‘to know something as the thing it is’ should be unpacked as an answer to
an indirect question, amounting to knowing what it is, and should therefore be couched
as knowledge that . . . ,35 McDowell reckons that Plato still needs to make further progress
(which he does in the Sophist), and that he is still some way from reaching clarity on
these issues in the Theaetetus.
So McDowell, like Frede, recognizes that Plato still does things that are alien to modern
sensibilities, even in his relatively mature works, but McDowell does not commend
these strange ways of thinking, but supposes that it is a stage on the way to less non-
sensical talk. By the end of the Sophist, Plato will be close to talking as we (post-Fregean
philosophers of language) would expect him to talk. Evidently then, McDowell, like Frede,
favours the Rylean Reading, but thinks that the progress to clarity is still incomplete
in the Theaetetus.
Aside from these mainstream interpretations, we should also mention Straussian
readings, which suggest that Plato’s dialogues have one message at the exoteric level,
33
John McDowell, Plato’s Theaetetus (Clarendon Plato; Oxford: Clarendon Press, 1973), particularly
118 and 192.
34
McDowell, Theaetetus, 115, 192. He also notes the direct object construction for doxazein, on which
see Chapter 11, Section I.
35
See Chapter 11, note 8.
OUP CORRECTED PROOF – FINAL, 03/06/2018, SPi
for a superficial reading of the text, and another hidden meaning for those in the
know. Unlike Sedley’s maieutic reading of the Theaetetus (which also uses a distinction
between surface meaning and philosophical message) and those Metaphysical readers
(e.g. Cornford) who think that Plato is nudging us to supply the Forms where they are
not mentioned, the Straussian tradition looks for political or social reasons, not liter-
ary or pedagogical ones, for why Plato must hide his meaning. It also looks for political
or social importance, not philosophical significance, in what is hidden. In what follows,
I shall often suggest that by reading some portion of text closely we can discover that
the discussion fails in what it overtly sets out to do, for reasons that the reader can see,
while the characters in the dialogue cannot see it, because those characters are missing
something important—something that Plato the author wants the reader to bring to
bear. In such cases there is a hidden message that is not stated in the text, but it is a
philosophical message, about how to escape an impasse that the characters have
encountered. When I suggest such an authorial tactic, the idea is more like Sedley’s
maieutic reading,36 and Cornford’s Metaphysical Reading of the Theaetetus,37 and quite
unlike the Straussian approach.
36
Sedley, The Midwife of Platonism. 37 Cornford, Plato’s Theory of Knowledge.
38
Not that one knows things on my view, but the faculties all take direct-object constructions. See further
Chapters 4 and 12.
39
Here I disagree with Sedley, The Midwife of Platonism, 119.
40
I cannot treat the Sophist in this volume, but see Chapter 13 for some thoughts.
41
I shall diagnose some peculiarities that are similar to what Frede found, especially his suggestions about
a special use of the verb ‘be’. See also Charles H. Kahn, ‘Some philosophical uses of “to be” in Plato’,
Phronesis, 26 (1981), 105–34 (reprinted in Charles H. Kahn, Essays on Being, Oxford: Oxford University
Press, 2009, 75–108); R.E. Allen, ‘Participation and predication in Plato’s Middle Dialogues’ in Gregory
OUP CORRECTED PROOF – FINAL, 03/06/2018, SPi
work, including the Theaetetus, without rewriting them to make knowledge or truth
propositional. However, I am not particularly concerned to make any of these works
(whether Middle or Late) defend the realist ontology that the Metaphysical Reading
found there. Rather, I share with Gail Fine the thought that the Platonist tradition
might not have the last word on what is going on, even in the Middle dialogues. The
continuity between the Middle and Late dialogues need not be a Platonism of exactly
the kind that has traditionally been attached to Plato’s name.
Still, I do not share Fine’s view of what we ought to find in Plato’s Middle Period. She
has a Rylean account of what knowledge should be, and suggests that we were wrong to
suppose that Plato ever had anything else. I, by contrast, want to suggest that we should
not have Plato end up there, with a Rylean view, even in the late period, and therefore, a
fortiori, not in the Middle or Early works either. In particular, I shall suggest that Plato
does indeed make a strong distinction between knowledge and opinion in both the
Middle and the Later dialogues, and he does not think, even in the Later dialogues,
that knowledge can be reduced to, or defined in terms of, belief, or opinion, or ‘seem-
ing’, plus something. I shall suggest that the traditional reading is right to link doxa
(seeming, discerning, opinion) with perception, and to think that Plato gives them both
the same status—that both fall short of knowledge because they are not accessing the
kind of content that can be known. In these respects I concur with the Metaphysical
Reading and dissent from much that Fine has to say.
On the other hand—and here I part company from the Metaphysical Reading—I do
not think that the distinction between knowledge and opinion entails a hard and fast
distinction between the sensible world and the intelligible world, or between Forms
and their instances, or between universals and particulars, or between being and
becoming. Certainly all these distinctions figure crucially in Plato’s thought, and inter-
sect in his discussions of both metaphysics and epistemology. But what falls on one
side of one of these divisions need not always fall on that side of the division, nor need
it always fall on that same side of the other divisions (if we imagine, for instance, that
we have laid these dualities out in a chart with two columns).
Here are two ways of thinking that things might be more complex than the traditional
‘two-world’ view of Plato’s distinction between Forms and Particulars. First we might
say that there are features (or attributes) and there are what the features belong to. An
individual case of justice can be regarded as a feature, or as a thing that has that feature.
In both cases the item is a particular that instantiates the generic form ‘justice’, but the
particular feature (a case of justice) is neither material, nor an object, although it is not
a Form.42 It is (or might be) abstract, but not generic. It is often unclear, when Plato
Vlastos (ed.), Plato: A collection of critical essays, Volume 1 (New York: Anchor Books, 1971), 167–83; Francisco
J. Gonzalez, ‘Propositions or objects? A critique of Gail Fine on knowledge and belief in Republic V’, Phronesis,
41 (1996), 245–75 (all questioning attempts to assimilate Plato’s uses of ‘is’ to the ‘is’ of predication).
42
It is what is known as a ‘trope’. See Peter Simons, ‘Particulars in particular clothing: three trope theor-
ies of substance’ in Stephen Laurence and Cynthia MacDonald (eds.), Contemporary Readings in the
Foundations of Metaphysics (Oxford: Blackwell, 1998), 364–84.
OUP CORRECTED PROOF – FINAL, 03/06/2018, SPi
ontological—but rather what kind of question one can know the answer to. Sometimes
Plato talks of knowing being, or ‘what it is’, which is the answer to a question about the
being of something. Being and truth and knowledge are closely connected in Plato’s
thought, as we shall see. The issue then is what is meant by knowing ‘what it is’, and
about what kind of thing (or under what circumstances) one can ask and answer this
question. One reason for trying to talk about Forms, among things that we might
know, is not that they are some special kind of permanent or real object, but that they
are the very same features as we use among our tools for predication and description of
other things, but when they are looked at in themselves, as the features that they are, not
as the features of something else, then we can say something about them, and indeed
know them, and not just use them to describe the incidental features of other things, as
we do in predication. Similarly, we can ask of some concept or generic idea not ‘what
things fall under it’, but ‘what it is in itself ’, which means something like what makes
a thing count as one of those, or what are the necessary and sufficient conditions for
falling into this class.
The question about ‘what it is’ seeks not the extension of the class but the criteria for
membership. It tempts us to an essentialist view, the search for definitions that give the
one key criterion that makes all just things count as just, or whatever. But although that
search for a single criterion turns out to be simple-minded, and never yields results,43
that does not mean that we never do know ‘what it is’ about an idea such as justice, or
that we should not ask what kind of knowing that is.44
II.iv That the contrast between episteme and doxa is the contrast
between the grasp of types or concepts and the recognition
of tokens or instances
In this book, I shall test the hypothesis that Plato thinks of knowing as answering the
‘what it is’ question about things of this kind—that is, knowing a concept or type, not as
an instance of another kind (e.g. that virtue is a kind of knowledge) but as what it is in
itself (what virtue is, what justice is). By contrast, seeming or perceiving are the terms
for seeing or recognizing a particular under some description, such as when we observe
that Theaetetus is ugly or that virtue is teachable, or that Socrates is a man or that virtue
is a kind of knowledge. (I use ‘seeming’ here as a handy term for non-sensory appearances
or doxa, where something seems to be F, i.e. dokei.) These, when expressed in language,
typically involve a kind of predication, forming a proposition to the effect that x is an F
(virtue is knowledge, for instance), but epistemically they typically consist in x seem-
ing F to A, or A perceiving x as F—or, as Socrates will more often say, A perceiving an F,
43
I simplify. It cannot yield results for the values that Socrates investigates.
44
For relevant reflections on knowing ‘what justice is’, as manifested in practice and not a definition,
see A.W. Price, Virtue and Reason in Plato and Aristotle (Oxford: Oxford University Press, 2011), 178 and
note 28.
OUP CORRECTED PROOF – FINAL, 03/06/2018, SPi
or perceiving the Fness of x.45 They involve applying a semantic idea or Form (such as
virtue) to an extensional particular (a particular virtue), even if that particular might
itself also be (on other occasions) itself a type (e.g. courage) that could have instances
or tokens of its own.46
This distinction between knowing Fness for what it is and seeing Fness as an instance
or example of some other kind or property G cuts across the divide between Forms and
particulars as traditionally conceived, since forming propositions about Forms or types
would then be a case of predication—and therefore a case of seeming, not knowing. It
might then seem that discovering, about a mere particular, what that thing is in itself,
would amount to knowing. But in fact it will not make sense to ask the ‘what is it’ ques-
tion about a particular, since knowing ‘what it is in itself ’ means knowing what makes
something count as one of those, and that is something that could only apply to something
that is generic, conceptual, or semantic—to a type or role, not to its token or occupant.
This helps us to see why Plato would want to say that the particular (qua particular)
cannot be known and is subject to becoming but not being.47
All of this runs completely counter to the pressure from some branches of analytic
philosophy to think that Plato was on his way to realizing (after some earlier confusion)
that ‘being’ and truth have something to do with predication, with structures of the
‘x is F’ sort, and hence that knowledge too should be propositional in structure. On the
contrary, I shall argue, Plato always assigns judgements about particular facts and tokens
to seeming, and even when true such judgements are no more than true seeming.
They fail to meet his criteria for being knowledge or science (episteme), not because
they are not truth-apt, but because the recognition that some x is F is the wrong kind of
content for episteme. In this book we shall look closely at some passages in both the
Meno and the Theaetetus that have been read as hints that predications of the ‘x is F’
kind might be serious candidates for being knowledge. I hope to show that Plato thinks
precisely the opposite.
Suppose I am right about what Plato is talking about, and he is right about the peculiar
nature of this knowledge. Should we then confine the word ‘know’ to this restricted range
of knowledge, as Plato confines his knowledge-terms to knowing what it is about a type
or form? Clearly not. On the contrary, knowledge will continue to mean what it ordinarily
means. Our vocabulary does not precisely fit Plato’s, and arguably Plato has tried to
impose an artificial restriction on how his Greek words can be used. He seems, on strict
days, to reserve episteme words for just one part of the knowledge spectrum. But his
usage can serve as a wake-up call to us. What Plato calls episteme is also something that
we would call ‘knowledge’, but arguably it does not occupy a sufficiently central place in
45
See Chapter 11, notes 12 and 52.
46
Again, see Price, Virtue and Reason in Plato and Aristotle, 178, 180–7.
47
For some discussion of this, see Chapters 4 and 5. The idea that the particular does not have being
of its own, but only becoming, is found more in the Timaeus and Sophist than in the dialogues we shall
discuss here.
OUP CORRECTED PROOF – FINAL, 03/06/2018, SPi
48
The topic of what it is to be a competent user of a concept tends to fall under philosophy of language,
meaning, or logic for us, rather than epistemology: e.g. Saul Kripke, Naming and Necessity (Oxford: Blackwell,
1980); Saul Kripke, Wittgenstein on Rules and Private Language (Oxford: Blackwell, 1982). See further
below (note 59, and Chapter 4).
49
E.g. in Protagoras 355d, 358b he discusses whether I can ‘know that something is bad’ and still do it
(I thank Michael Morris for raising these examples in debate). There is controversy over whether Socrates
claims to know ‘that he knows nothing’, but arguably he does not (see discussion in Gail Fine, ‘Does
Socrates claim to know that he knows nothing?’, Oxford Studies in Ancient Philosophy, 35 (2008), 49–88).
50
See Section I.ii.
Another random document with
no related content on Scribd:
made a visit to Hempstead, where she formed the acquaintance of
quite a number of young men that were in the army with us. On our
second day’s sojourn a Mr. Tom Clay, belonging to Company K of
our regiment, whose home was in Washington County, and who had
been intimately acquainted with Miss Josie during her stay at
Hempstead, proposed to me to call on Miss Josie, to which I agreed.
Alighting in front of their house, Miss Josie happened to be standing
in the door and recognizing us, rushed to the gate and invited us in.
Just then a little boy came along and asked me to give him a little
silk Confederate flag some young ladies had presented me with the
day before and I had sticking in the browband of my horse’s bridle.
Fearing the little boy would take the flag while we were in the house,
I suggested that I had better take this in with me. Miss Josie then
said, “That flag can’t come into our house.” Up to this time we were
unaware that they were Union people. My friend, bowing to her, said,
“We will certainly not go into a house where our flag is not welcome,”
and we declined to go in. By this time her mother had come to the
gate, when Miss Josie introduced us. Mrs. Newman having heard my
name, through the Cannon family, quite often, she insisted on our
coming in, when we told her Miss Josie’s objections. She chided her
for her discourtesy and told us to come in and bring the flag, when
my friend said to Miss Josie, “We will compromise the matter with
you and go in, if you will sing Dixie and Bonnie Blue Flag for us,”
which of course she had to agree to do and, while singing these
songs, I sat at the end of the piano with my little Confederate flag in
my hand and when she sang the chorus I would wave the flag.
After two days’ sojourn we moved on up towards Louisville, part
of our force dividing and occupying the town of Taylorsville on our
right; the balance of the command camping near Mount Washington
on the Bardstown Pike. Here General Forrest received an order from
the War Department to personally report to Richmond and turn the
command of the brigade over to Colonel Wharton. In about a week
the Federal forces advanced out of Louisville. They were said to be a
hundred thousand strong, while another force moved out of
Cincinnati, about sixty thousand strong, with a view of cutting us off
from retreat to Cumberland Gap.
CHAPTER XIV.