Negation and Not-Being in Wittgenstein's Tractatus

and Plato's Sophist

by Robert B. Pippin, San Diego

The origins of our contemporary fascination with language are, of course, quite com­
plex and go to the very heart of that persistent twentieth-century attempt to see
philosophy as a "critique of language". But, in investigating those origins, it does no one
an injustice to insist upon the importance of Ludwig Wittgenstein and especially his little
book, the Tractatus Logico-Philosophicus', in bringing the issue to the prominence it en­
joys today. Central to that book is a doctrine that has come to be called the "picture
theory of meaning", Wittgenstein's attempt to answer some continually raised paradoxes
about the relation between language and reality. This account, although rejected later by
Wittgenstein himself, proved markedly influential when it first appeared, and still stands
as one of the most straight-forward and elegantly simple descriptions of how language
works. It also sec�ms to return to a much older historical tradition (older especially in ap­
pearing to be pre··Kantian) in its insistence on a "correspondence" theory of truth and the
iconic nature of language. In fact, it seems to return quite explicitly to Plato's account of
language as an eidolon in a dialogue like the Sophist. In a certain sense, one could claim
that the central problem of dialogues like the Theaetetus and the Sophist was Wittgen­
stein's major concern in his early work. Indeed, Anscombe has claimed just that and
more.

Plato's Theaetetus 189 A: 'In judging one judges something; in judging something, one judges
something real; so in judging something unreal, one judges nothing; but judging nothing, one is not
judging at all.' Wittgenstein returned to the problem presented by this argument again and again
throughout his lif,e2•
Further, in the opinion of some commentators, the Eleatic Stranger and Wittgenstein
not only begin with very similar problems, they seem to arrive at very similar solutions.
The picture theory's representational model of language's relation to the world, the on­
tology taken by some to be supported by the picture theory (Wittgenstein's infamous
"simples"), the doctrines of logical space and the "form" of objects, and perhaps more
than any other issue, Wittgenstein's "derivative" explanation of negation (the claim that
any not-X depends on X for its intension and the claim that it has no negative extension,

I Unless otherwise indicated, the Pears-McGuiness text of the Tractatus Logico-Philosophicus is

used (London. Routledge & Kegan Paul. 1961).
l G.E.M. Anscombe, An Introduction to Wittgenstein's Tractatus (London, Hutchinson Univ.
Library, 1971), p.13.

0022-8877/79/0702-0005$2.00
Copyright by Walter de Gruyter & Co.
180 Robert B. Pippin

that there are no negative facts), all count as evidence for Platonic shadows stretching
across the Tractatus. This seems especially true when we consider that Wittgenstein re­
garded as a major consequence of the picture theory its ability to account for meaningful,
false propositions, that it could explain how "Thought can be of what is not the case".
Plato's discussion of images is clearly and directly concerned with much the same prob­
lem in "capturing" the elusive sophist.
In the following, I will consider two such comparable issues - the general theory of lan­
guage involved in both accounts, and their specific solution to the problem of negation
and false propositions. What I hope to accomplish by this contrast is to illuminate two
very different kinds of analyses appropriate to the topic of "not-being", differences one
could roughly characterize as "semantic" versus "ontological". Further, this difference
in orientation and in emphasis will involve differences within each mode; specifically it
will involve a "picture" versus an "image" theory of language, and atomistic versus non­
atomistic ontologies3_
Our first task then is a provisional summary of Wittgenstein's picture theory.

Unfortunately, though, there is no simple way to summarize the details of Wingen­

stein's theory of language as Bild. The bare list of propositions which prc�sents the theory
is embedded in some spectacular ambiguities which have not gone unnoticed by Witt­
genstein's commentators. But, at least we can notice where pictures first enter the text. At
2.1. Wittgenstein announces, "Wir machen uns Bilder der Tatsachen". Up until this
point the task of the Tractatus had been explicitly ontological; a series of propositions
had described the "world" as composed of "facts" and those facts as immediate combina­
tions of "objects", which themselves never occur except as combined in a "Sachverhalt".
As is often pointed out, Wittgenstein thus arrives at two ontologically primitive notions
in this initial account, Sachverhalt and Gegenstand, neither really prior to the other".
But even here, before we reach any point where we could begin to undc!rstand how lan­
guage could relate or refer to the "world", there are some initial problems in this brief ac­
count of world itself, as "to-be-pictured". The first, and one of the most famous am-

3 I take it as established by now that "positivistic" interpretations of the Tra ctatu5 are mistaken,
and that a discussion of its (albeit limited) "ontology" does not sound as strange as it once would
have. Cf. Anscombe's critique of the positivistic inte rpretatio n, An Introduction, p.25ff., and
p.lSOf£., as well as Anselm Miiller, Ontologie in Wittgenstein 's 'Tra ctatus' (Bonn, Bouvier &
Co., 1967).
4 "Atomism" would seem an appropriate, even if a bit am biguous, word to describe this position. A
full discussion of in what sense it is atomistic would take us into the complex issues of "particu­
lars", "bare particulan", "simples", etc. I can only refer here to the excellent discussion of such is­
sues by Sellan, Copi and Bernstein in E5SaY5 on Wittgenstein'5 Tractatus, ed. by Copi and Beard
(New York, The Macn,ilIan Co., (966). We can also note that, despite their differences, all work
to preserve the independence. in some sense, of objects and Sachverhalte from one anomer.
 Negat ion and Not-Being 181

biguities that must be resolved is the relation between a "Tatsache" and a "Sachverhalt".
As is well known, the original Ogden English editionS translated these terms as, respec­
tively, "fact" and "atomic fact". Such is the translation Wittgenstein preferred and the
one adhered to by Russell in his Introduction. Under this translation, a Sachverhalt
would be just a special kind of Tatsache, one not reducible to any further fact. In that
sense, all atomic facts would be facts, but not all facts would be atomic facts. There are
tbough, a number of other interpretations of this relation, many based on the Pears-Mc­
Guinness translation of Sachverhalt as "state of affairs". The dispute (and dispute it is,
with Anscombe, Black, Hartnack, Stenius, Maslow, Favrholdt and Schwyzer all having
different interpretations), as far as I can see, really centers on the meaning of 2: "Was der
Fall ist, ist das Bestehen von Sachverhalten." A resolution of the dilemma posed by such a
statement is crucial before we can understand what "facts" Wittgenstein is talking about
when he tells us we picture facts to ourselves('. For, 2 either means that a Tatsache is the
actuality (or actualization) of a Sachverhalt, which latter would then be a "possibility",
so that a "fact" is;an actualized or existing (Bestehen) "state of affairs"; or, 2 could mean
that a fact is made up of "atomic facts", as a "molecular fact" would be. This latter in­
terpretation has some weighty evidence on its side. In the first place, it is much truer to the
translations Wittgenstein himself preferred; in the second place, when Russell asked him
this question explicity, Wittgenstein responded

'What is the difference between Tatsache and Sachverhalt?' Sachverhalt is what correspon d s to an
Elementarsatz. if it is true. Tatsache is what corresponds to the logical product of elementary propos­
itions w hen this pmduct is true'.

And, in the third place, this interpretation would coincide with 2.034: "Die Struktur
derTatsache besteht aus den Strukturen der Sachverhalte." However, it has one serious,
notto say fatal, difficulty. According to the unambiguous insistence of the Tractatus, no
two "atomic facti;" can ever be said to "combine" in any way, least of all in a way that al­
lows that combination to achieve the separate status of a "fact". (2.061: "Die Sachverhalte
sind von einander unabhangig.") So it can never be correct to speak of a Tatsache as a
"group", in whatever sense, of Sachverhalte; but something like that must be true for
there to be a distinction between the two. It may be that there is no final way to explain
successfully the relation of propositions like 2 and 2 .034, with 2.061. Wittgenstein simply
uses the word "Tatsache" in tOO many different ways for one to define precisely its rela­
lion to "Sachverhalt". But, at the very least, one can say that Tatsache is a derivative no­
tion, derived from what Wittgenstein means by the more fundamental Sachverhalt; that
Wittgenstein sometimes uses Tatsache in the same general sense as Sachverhalt (d. 1.1
with 2.04) and sometimes uses it to cover a wider "logical space" than just a Sachverhalt

5 Ludwig Wittgenstein, Tractatus Logico-Philosophicus, transl. by C. K. Og d e n (London, Rout­

ledge & Kegan Paul, 1922).
, For a det;pled i nves ti gation of the numerous possible interpre tations of this problem, see David
Favrholdt, An interpretation and Critique o!Wittgenstein's Tractatus (Copenhagen, Muskaard,
1967), p. 38 ff.
1 Ludwig Wi ttgenste in. Notebooks. 1914-1916 (New York, Harper & Row, 1961), p. 129.
182 Robert B. Pippin

(cf. 2.034: "Die Struktur der Tatsache (i. e. a singular fact) bestehtaus den Strukturen der
Sachverhalte" (i. e. plural states of affairs)8.
We are assisted, though, in trying to understand what Wittgenstein thinks language
pictures by his later remarks on the difference between a Satz and an Elementarsatz,
found most explicitly at 4.21 ff. and especially in the contrast between 4.21 and 4.4:

4.21: The simplest kind of proposition, an elementary proposition ( der Elementarsatz) asserts tbe
existence of a state of affairs (Sachverhak).
4.4: A propo sition is an expression of agreement and disagreement with t!1Jth-possibilities of
elementary propositions.

When these two statements are coupled with 4.0312, "My fundament:!1 idea is that log­
ical constants are not representatives (nicht vertreten); that there can be no representa­
tives of the logic of facts", then some conclusions may finally be drawn :!bout what Witt­
genstein means when he says we picture facts to ourselves. Since he seems to use the word
Sachverhalt as a "kind" of Tatsache, and since "propositions" are truth functions (not
pictures) of "elememary propositions", and since the truth function:!l connectives of
these propositions represent nothing (as, indeed, they could not, since all Sachverhalte
are independent), then, from all of this, when Wittgenstein says, "Wirrnachen uns Silder
der Tatsachen", he must mean a picture of a special kind of Tatsache, al Sachverhalt, an
"atomic fact" consisting only of objects in immediate combination9• The proposition that
is that picture must be an Elementarsatz, all other Satze being truth functions of these
elementary ones, which truth functions "represent" nothing in the W'orl d 1o•
Finally, then, to complete this initial account of the essential features of the picture
theory, we need to consider what Wittgenstein thinks pictures, as propositions, have "in
common" with their facts so that they can picture them. His well known answer to this
question is given, basically, in propositions 2.16-2.22 and especially at 2.18:

What any picture, of whatever form, must have in common with reality (Wit·klichkeit), in order
to be able to depict it-correctly or incorrectly-in any way at all, is logical fonn, i.e. the form ofre­
ality (die Form deT Wirklichkeit).

And even more important is 2.202: "Das Sild stellt eine mogliche Sachlage im logischen
Raume." Pictures picture by having their logical form in common with a "place" in what
Wittgenstein calls "logical space". defined as the nexus of possibilities determined by tbe
inherent possibilities of combination (the "form") of objects. A proposition thus has

• Cf. H. R. G. Schwy:z.er's explanation of this relation, an explanation which contrasts with this and
Wittgenstein's just quoted formulation in the Notebooks: Wittgenstein's Picture Theory of LIn·
guage, in: Essays, ed. Copi and Beard, p.276.
9 This result is supported by Copi in his Objects, Propenies and Relations in the 'Tractatus', Emql,
ed. Copi and B eard, p.170.
10 Wittgenstein goes on here to describe how "names" in the proposition "stand for" ("vertreten"at
2.131 and 3.22) objects in the fact and thus how the "meaning" of a name is just its bearer. But it is
clear that names do no "picture". Only propositions picture and only they picture since a picture
is necessarily a picturing that so and so. Every picture must present its pictures as like /hi!,
whereas a name only "stands in" for its bearer. Cf. Copi's criticism ofDaitz, Objects, PropmiLl
and Relations, p.167ff.
 Negation and Not-Being 183

meaning only if it represents, or "pictures", one of these possible points in the realm of
logical space; it is" further, true if that point happens to exist, false if that point does not
exist I 1 Thus we have what is intended as the solution to the first of many traditional
•

philosophic problems about the relation between language and world - meaning and
tnIth. We are able to understand the meaning of a proposition without knowing whether
or not it is true in roughly the same way we can understand what a picture is about with­

out yet knowing whether it is actually a good representation.

But before dealling with this theory's treatment of false propositions, we should note
here one odd feature of this account, a characteristic that will eventually be important in
our attempt to distinguish Wittgensteinian "picture" from Platonic "image". We need to
consider the observation first made by Columbo, the Italian translator of the Tractatus.
He claimed, in effect, that Wittgenstein's strong insistence on this "isomorphism" of
"logical fonn" between language and reallity was wildly overstated; that, from what
Wittgenstein says in the Tractatus
...it was difficult to see why a described fact should not be regarded as itself a description of the
proposition that would nonnally be said to describe it, rather than the other way around 12.

Anscombe, in her commentary, admits this is indeed a "strong point", but claims to
counter it by simply quoting2.1513 and 2.1514 in which Wittgenstein insists that the pic­
turing relation itself is contained in the picture; serving, for Anscombe, as a sufficient
characteristic to distinguish picture and fact strongly enough to preserve our ability to say
a proposition is a picrure of a fact and not the other way around. But this is not sufficient.
If th e proposition can picture because of the picturing relation it shows and does not
state, then the fact pictured must "contain" the ability to be pictured, or, again, the "pic­
turing relation" itself. Wittgenstein says as much at 3.001: "A state of affairs is thinkable:
what this means is that we can picture it to ourselves." A Sachverhalt must contain the
picturing relation Anscombe thinks sufficiently preserves the tenuous distinction in the
Tractatus between language and the "real"t3. Wittgenstein's language throughout the
Tractatus always insists on the identity of logical fonn between picture and pictured 14,
so much so that even the necessity for "verification" of propositions in the "world" (by'

II
It is necessary here to distinguish twO notions which Wittgenstein uses in consistently different
ways, but which he never fonnally distinguishes. "Welt" in the Tractatus clearly includes only
those SachverJJalte that actually "are the case", while "Wirklichkeit" denotes the existence and
non-existence (Nichtbestehen), or the "possibility of existence" of Sachverhalte, as at 2.06 and
2.04. Thus a picture is meaningful as a picture of "reality", it is true as a picture of the "world".
We shall have some difficulty later with this "non-existence". Cf. also James Bogen, Wittgen�
stein's Philosophy of L4nguage (London, Routledge & Kegan Paul, 1972), p.24.
12 M reported by Anscombe, An Introduction, p.67ff.
13 Anotherconscquence of this feature of the picture theory that makes it hard to appreciate the am­
biguity of soplustry in a Tractarian world is the sense-detenninacy thesis. What we mean, in
elementary propositions must always be clear; we must know what it would mean for the propos­
ition to be true or false in order to say that the proposition is meaningful. The sophist's claim to
imitate the whole docs not seem a likely candidate for such detenninacy. Cf. Bogen, Wittgen­
stein's Philosophy of Language, p.43.
14 Contrast Schw)'zer's denial of this, Wittgenstein's Picture Theory, p.272ff.
184 Robert B. Pippin

far the most stupendous ambiguity in the book) seems muted by the fact that world, as de­
scribed in the Tractatus, is, only in accordance with the logical form demanded by pro­
positions. Rather than verify propositions by the world, we seem rather to describe the
world as a result of the character of propositions, as again we seem confused about
"what" is original and "what" picture IS •

At any rate, having seen at least provisionally what it is that Wingenstein thinks lan­
guage pictures and how he thinks it accomplishes this, we turn now to the most difficult
problem raised by that analysis - the "meaning" of negative and false propositiGns. In the
first place, it is clear that these must be separate problems. Simply put, a negative proposi­
tion (-p) can be true or false, in obvious contrast to a false proposition. Wittgenstein's
position on the former of these two problems is well known. To sum up, to say "_p" for
Wittgenstein is not to "machen ein Bild" at all, but is to perform a truth functional opera­
tion on the picture "p". Both in tlie sense of so-called "normal" negation ("The river is
not long".) and in the sense of "radical" negation ("There are no unicorns.") negative
propositions have meaning for Wittgenstein only because of, or as derivative from the
meaning of the pictured proposition it negates. In .. -p", it is the "place" in logical space
designated by "p" that determines the "Sinn" of the truth functional statement "-p". The
famous and strikingly odd "picture" in the Notebooks "shows" all of this immediately.

(-p) (P)

... The proposition, the picture, the model are - in the negative sense -like a solid body restrict­
ing the freedom of movement of others; in the positive sense, like the space bounded by solid sub­
stance, in which there is room for a body. This image (Vorstellung) is very clear and must lead to the
solution I ••

Or, as in the Tractatus itself,

4.0621: But it is important that the signs 'p' and '-p' can say the same thing. For it shows that no­
thing in reality corresponds (entspricht; to the sign '-'.
4.0641 : The negating proposition detennines a logical place with the help of the logical place of the
negated proposition. For it describes it as lying outside the latter's logical pla<:e.

By far the most important consequence of this treatment of negation is expressed in

..... nothing in reality corresponds to the sign' -'." Wittgenstein has effectively excluded
the question of negation from any ontological arena. Just as with the famous distinctions
supposedly discovered in the Sophist. to say not-X (where X is here a whole proposition,
not a part), is not to assert any contrary of being ("the non-X") but is just to "use the ne­
gation sign correctly", to follow Owen's characterization of the Stranger's strategyl7.

.. For difficulties associated with this world-language conflation, as well as a number of other exp­
licitly ontological difficulties, see Muller, Ontologie, p. 231 ff.
1(. L. Wittgenstein, Notebooks, p.30.
11 G. E. L. Owen, Plato on Not-Being, in: Plato: A Collection of Critical Essay.!, ed. by G. VlastoS
(New York, Doubleday, t 971), p. 231 ff.
 Negation and Not-Being 185

The cryptic ontology of the first few pages of the Tractatus asserts only a positivistic uni­
verse of separate, unaffectable Sacbverbalte, composed of Gegenstiinde in combina­
tions so "immediate" as to deny also any ontological reality to "relations".
But it should also:> be emphasized that this does not mean that negativity, or, tradition­
ally, "not-being" has a secondary or derivative "reality" in Wittgenstein in something
like the way dynamis achieves an ontological status in Aristotle. Rather the sign "-" re­
presents no reality at all; it merely instructs us to perform a certain operation on com­
pletely positive pla.ces in logical space. All of this also follows quite consistently from the
atomistic nature of these Sacbverhalte. To be able to say, as suggested above, that any
one "place" in logical space is determined only by, in some determinate way negating·
others (or to say ",·ith Spinoza, "Determinatio negatio est. "), would be to claim that the
elements of this "world", rather than being primordially independent, are critically inter­
connected in order to be said to have a "place" at all. Such a conclusion would obviously
be unacceptable for Wittgensteinl8•
Negation is thus merely an operator and negative propositions are not pictures at all,
there being no "negative" Sacb'rlerhalte to picture. But there is another issue that always
hovers near this solution to the negation question and one which is often confused with it
-the more famou!;, traditionally "platonic" problem of how false propositions (here, let
us say false Elemtmtarsiitze, or false pictures) have meaning. If elementary propositions
are always pictures, to put the whole problem simply, what is "it" that a false elementary
proposition pictures?
Initially, it is tempting t o see this problem as merely a version of the problem of nega­
tion. One could reasonably claim that so-called "false" propositions were merely a
species of negativt� ones, in the sense that the meaning of asserting "p" and having it tum
out false would lx, just the meaning of "_p" being true. Likewise "-p" as false only means
that "p" is true. 1bere is, though, an immediate, obvious error in this analysis, at least in
terms of the Tractatus. That is, a proposition does not cease being elementary merely by
being false, and, even more critically, it does not cease being a picture. Because an elemen­
tary proposition is false, its meaning does not cease to be explained by the picture theory
and begin being explained as a "truth function" of a picture. As false, it is still a picture,
its meaning is stilll dependent on what it pictures and cannot be explained by a wholly dif­
ferent sense of "meaning". To move from "p is false" to "-p" is to move from a picture
theory of meaning to what Favrholdt has called the "thesis of extensionality", to a truth
functional realm where a meaning is derived from pictures but is not itself a picture.
Favrholdt even goes so far as to suggest that there should have been two explicit symbols
in the Tractatus, one for negation and one for truth and falsity ("p" and "-p" on the one
band, and "p t '" and "p � " on the other), designations we shall follow belowl9•

18
The role of "negativity" for Hegel, even in such fonnulations as [he law of identity, should also b e
contrasted with Wittgenstein. CE. Wissenschaft der Logik (Hamburg, Felix Meiner, \969), v. II,
pp.26-32 with the Tractatus at 4.242 and 4.243.
19 Favrholdt, An Interpretation, p,4S.
186 Roben B. Pippin

This point is crucial for understanding why the sophist will pose diffi(:ulties for us later.
The whole point of much of the Tractatus' account of language is to explain how we can
know a proposition's meaning without knowing whether it is true or false . Likewise, in
Plato, it is not only the possibility of knowingly "being in error" that must be explained
(the possibility that, when p is asserted, we can claim not-p and avoid the wrath of Father
Parmenides), but also the possibility of "seeming to be true", or of being meaningful
prior to being either true or false. The Sophist does not only taunt us with a supposed ina­
bility to Sf!:)' not-p, but also with our inability to explain how p could make sense in the
first place, if not-po Put most directly, it is easy but dangerous to assume that the Sophist
is the "answer" to the Theaetetus' problem of error. But, unfortunately for our attempt
to capture the Sophist or explain falsity in Wittgenstein, the sophist does not elliptically
say not-p; his eikon would not be deceptive if it claimed to be of what is "other than p",
leaving us only the task of explaining -po In that case, we would need to .know that p were

false before we could explain what the sophist said.

It is, of course, true that part of what is involved in capturing the sophist is in showing
that we can meaningfully say that he "says what is not"; that our response of not-p to his
p is sensible without positing any contrary of being. But also involved is explaining the
negative status of what he says, how he makes an image of something, and how that
something also is not. The sophist does not image not-X; he images Y ;15 X. Our attempt
to unmask him is thus committed to two tasks: (1) Understanding how not-X could have
a meaningful status, and (2) understanding the mode of being ofY, or more precisely, Y's
referent, especially since it seems to be X. This problem of the mode of being of the image,
or here in Wittgenstein, of false pictures, is lost if it is treated as some elliptically negative
claim. In short, a false picture is still a picture and we must still ask what it pictures,
especially as contrasted with a true one20• In order to be finally clear on what status a false
proposition can have within the whole explanation of the picture theory, consider the fol­
lowing analogy, very loosely borrowed from what Wittgenstein says at 4.063. Let us

suppose that the whole of logical space could be represented by a syst/:m of coordinates,
within which we may locate "points".

4.

3. •

2. Then, consider the following propositions:

<a) There is a b lack dot at (:l,3).

1. (b) There is not a black dot at (1,2).
1. 2. 3. 4. (c) There is a black dot at (1,2).

20
We should take note here of the various well-known attempts out of this problem. Stenius, Witt­
genstein's Tractatus: A Critical Exposition of its Main Lines of Thought (Oxford, Blackwell,
t 960) would have only true propositions "really" picture (p. 96) and Anscombe would havc pic.
tures themselves b e neither true nor false. Both of these interpretations seem to me unacceptably
in conflict with 2.21.
 Negation and Not-Being 187

According to the analysis thus far, assuming all except (b) to be elementary proposi­
tions, (a) and (b) would seem to be explained. (a) pictures the existence of a state of affairs,
it pictures a possible point in logical space which also happens to be the case. (b) makes use
of a possi ble picture of a black dot at (1,2) and excludes it, and, by doing so, happens to be
true. But already, within that negation, a problem has arisen. Namely, what does (c), or
what (b) negates, picture? It is certainly not a picture of what is the case. As a false pic­
ture, it seems to picture what "is" only as a possibility, and what also happens not to be.
Butthen it picture:! what is and is not. -p may not picture a negative fact, but p � seems to.
lothe language ohhe S ophist, what is "other than what-is" (the world) seems to be. And
of course, Wittgcnstein explicitly wanted to avoid any Meinongian subsistings or existing
possibilities etc., yet his relegating all questions of otherness to the operational status of
negation would seem to be leading in that direction. (Or, of course, in a direction far re­
moved from the piicture theory, towards the rule-following, speech-act locution of Witt­
genstein II.)
Thus we arrive at three strange Wittgensteinian ontological planes: (1) language, or pic­
tures of (2) logical space, or points of possibility which may or may not happen to be the
case in (3) the world, the acutal existence of states of affairs.
But if pictures are, initially at least, pictures of "possibilities", then some revisions will
have to be made in Wittgenstein's earlier stated ontologf'. For, nowhere in the Trac­
tatllS is that realm of logical space as a realm "other-than-the-world" explicitly discussed

in ontological terms. Its status is indeed highly questionable. It is hard to see how it could
have any reality independent of the "world" (as it must have, if false propositions are to
have sense) and yet still "be" . For, the Sachverhalte occupying points in logical space are
supposedly made up of Gegenstande, which are explicitly said to constitute the sub­
stance of the "world" or "what is" and do not seem capable of having any status as what

merely seems to be. This means that all we can say thus far is that "logical space", or the
system of Sachvf'rhalte-possibilities must be ontologically distinct from the "world" or
what only "is" thl� case. This then means that a false proposition pictures only a possibili­
ty, but still "something" that has sense, and is as a possibility, but yet as actual is not. (As
apicture then, "it is and is not.") But, with this as a consequence, is this really any differ­
ent from saying that it pictures what is not (thus reintroducing negativity as having some
"onto l o gical" status)? This problem is compounded when we recall Columbo's objection

and the fact that a. picture must share exactly the same logical form with what it pictures.
In that case what "logical form" would a false proposition have or "share", especially in
contrast with the logical form a true proposition shares? The difference must be between
the logical form of that which seems to be but is not, and the logical form of that which
is. But, of course, to re-raise the traditional problem, how can that which is not have a
"logical form"?

" Clearly, revisions can also be made in the explanation of pictures, interpreting them more as pic­
turing acts, rather than as facts necessarily connected with other facts. Forthe moment, however,
here and in the Sophist, I am more concerned with the ontological issues at stake in Wittgenstein's
eady phase than in the revisions made later, revisions which, by moving much closer to conven­
tionalism, abandon the "Platonism" I am interested in pursuing.
188 Robert B. Pippin

We have thus arrived at a situation where the whole concept of "picture" seems to me
to have become questionable, precisely in the sense in which it is quite clumsy to speak of
a picture of a possibility, or of that which is not but may "seem" to be, especially since,
except in very special circumstances, we do not mean to picture possible points in logical
space. To speak loosely, there cannot be a picture of a "hole" in logical space, a place
where there may be, but is not a "point". And even further, beyond this problem of false
propositions, it must be the case that all propositions are pictures of possibilities, or pos·
sible situations in logical space and, then, that some of them happen a�,o to pictUre what
holds in the world as true. As pictures, they depict only possibilities, of what is and is not
at the same time. But a "picture" does not, in any sense, call attention to this "distance"
from what is; quite the contrary, Wittgenstein's claim is an "identity oflogical form" be­
tween picture and pictured.
To sum up, there are three elements in Wittgenstein's account which contribute to such
a problematic status for false pictures. First, Wittgenstein's emphasis> on mathematical
isomorphism between picture and fact ignores the crucial element of intentionality neces­
sarily involved in the picturing relation. We Can surely use language f()r ends other than
picturing (as in sophistry, for example), and this should convince us that the speaker's
activity, his "making" is central to the sense of the proposition. Now , Bernstein, Copi,
Black, and especially Schwyzer2 have tried to discover some of the la,ter Wittgenstein's
act locution in the Tractatus, but, for the most part, I think Bogen is right here23.ltwas
only after Wittgenstein discovered the role of intentionality that he saw that the picture
theory had to be abandoned. What is meant by this discovery of the role of intentionality
is shown clearly in the Blue Book.

An obvious and correct answer to the question, 'What makes a portrait of so-and-so?' is that it is
the intention".

The Tractatus had claimed a necessary relation between picture and fact, a relation we
found difficult to assess in false propositions.

A proposition presents a situation to us and so it must be essentially connected with that situation
(4.03).

To put this move in the Sophist's language, a picture is made, it is the result of an act.

"p" need not share a logical form with anything (like a possible point in logical space);
there need be no "shadow" between the proposition and the world.

21
Even Schwyzer's heavy insistence on interpreting pictures as acts, as pict�lrings or presentings.
admits that we must not only present a state of affairs (assert that such and such is the case), but
we must also show what we are presenting, reintroducing some of the traditional picture locu­
tion he had excluded. Finally, Schwyzer's admitted difficulties with 4.012 and 4.014, as well as
the conflict of his interpretation with 2.16,2.17, 4.023, and the Notebooks (p. 129) weaken his
overall attempt considerably. Boge.n's evidence for a change in Wittgensteiltl's thinking afrertbe
Tractatus seems to me more supportable in the text.
23 Bogen, Wittgenstein's Philosophy of Language, pp.98 and 104 ff., and hi� summary on p.159.
24 Ludwig Wittgenstein, The Blue and Brown Books (New York, Harper and Row, 1958), p.32ff.
 Negation and Not-Being 189

Now, as is well known, this shift to an act locution involves other serious alterations.
As Bogen puts it, we replace
the question 'how can an assertion be internally connected with a possibly non-existent fact?' with
the question 'how does a speaker use such and such a sign to say so and SO'25?

But before we travel down the road opened by this move, we will want to consider, via
the Eleatic Stranger, whether admitting the role of the speaker's activity and intentions
cannot yet preserve the "connections with reality" language implicit in the description of

Uimages".
Secondly, Witlgenstein's discussion of negation as a linguistic operator ignores all the
"material" questions such an issue might generate. The referent of a false proposition
must be other than what-is-the-case in a far more determinate sense than "not the same
as". And that otherness cannot be explained as if it were an instance of negating true
propositions. A negative proposition has sense by virtue of the meaning of the proposi­
tion it negates. But a false proposition has its own sense, it is meaningful prior to being
understood as tht� negation of what is, and the status of that otherness still requires an ex­
planation.
Finally, we nf,ed to distinguish propositions that straight forwardly "picture what
isn't", from deceptive propositions, so constituted as to picture not ju�t what can be, but
what "seems" to be. Or, we need to explain the Sophist'S activity by including an account
of the conditions for his "deception".
I believe all three of these qualifications are taken into account by the Eleatic Stranger
(ES) and that his discussion argues for the "reality of otherness' in a way quite different
from Wittgenstein's, a way that may prove helpful in eventually defining the difference
between "dialectical" and "dianoetic" logics.

II

The issue of philosophy and sophistry, the issue of the dialogue Sophist, is especially
relevant to this last point. The Tractatus, at least as much as any other book in this cen­
tury, was extremely persuasive in insisting that philosophy was only a "Sprachkritik", a
methodology for purifying sentences of their occasionally bogus appearance of meaning
and reducing them to the bare structure of elementary propositions and logical connec­
tives. Indeed, it could be said that the ultimate claim of the Tractatus is that, in the over­
whelming majority of cases, what passed for "philosophy" through the centuries, was
only "sophistry''', only the deceptive "appearance" of meaning. But, although there is a
dubious theory of false propositions in the Tractatus, there is no explanation of the pos­
sibility of this strangely deceptive "seeming", a seeming which is the life blood of the
sophist's activity. There is no way, speaking strictly within the Tractatus, of explaining,
either by reference to "intentions" or to the simple structure of language itself, the fact

2S Bogen, WittgEnstein's Philosophy of Language, p.123.

190 Robert B. Pippin

that that which is not appears as ifit were. Allpropositions, as pictures, appear as mere
po�sibilities, all of which need to be referred to their verification instances, whatever these
"originals" might be. Any proposition that cannot be so verified is "non-sense". This
elimination of kinds of sentences which only appear to be about those verification in­
stances but are not (and thus whose status is "non-verifiable") eliminates not only the
Tractatus'ability to define "sophistry" at all, but as is well known and much argued,its
ability to define its own "philosophic"enterprise by propositions themselves regarded as
nonsense since not "pictures" of anything. In short, there must be some intermediary
state between saying what simply is and saying what simply isn't, how,ever difficult this
latter is to defend just by itself; a state "partaking" of both and requiring an account of its
peculiar ontological conditions. But, if it is true that sophistry, the per,ennial accusation
against philosophy. can only be defined with reference to "image"and "not-being"how
is it that Plato's Eleatic Stranger arrives at a way of "hunting" the sophist different from
the reduction of language to logical form?
We obviously will be unable in the following to deal with a number of the well known
issues involved in that famous hunt, or to approach an overall interpretation of the
dialogue. Doing so would involve such complex issues as explaining the identity of the
ES, understanding Socrates' silence, placing the dialogue in its projected trilogy,etc. For
our purposes, we shall be concerned mostly with how the issue of notoobeing arises, and
how it is initially resolved (2321:r-24 1d and 252e-259c).
For reasons not immediately clear, the Eleatic stranger of the Sophil,t appears to have
considerably less reverence for Parmenides than did Socrates2(.
, "Whoever" the stranger
is, he is willing to discuss the sophist and false opinion explicitly in terms of images and
thus to confrontthe problem of not-being head on, risking even the accusation of being a
"parricide" (patraloias). By doing so, the stranger makes explicit the philosophic conse­
quences that arise when language is conceived as "image" instead of picture.
After six sometimes comically involved diaeretic attempts to capture the sophist, the
difficulties in the stranger's much heralded method seem to lead him 1:0 decide that the
sophist can only be pinned down by considering in detail (and non··diaeretically) the
mode of being of the image. The transition, though, from the diaeresis section of the
dialogue to the much more famous discussion of images and non-being is not accomp·
lished immediately. It is preceeded by a strange "return" to an earlier diaeresis, the fifth,
where the sophist had been defined as a "disputer" (antilogikon). The stranger now adds
that he is also a "teacher" (didaskalon) of this artand shows Theaetetus how he claims,as
such a teacher, to be able to teach "all things", both of them then wondering at this re·

markable claim. At this point (233b ff.) the stranger argues forcefully that,obviously, tbe
sophist must only appear to be the wisest of men, thus concluding that his rnLlm'II.L1J is
only a [)ol;,uO'ttxij. This description of the sophist elicits an exciu:d response from
Theaetetus: "nUVTwtaOt !Ltv oilv"he rejoices, deciding that this claim is the "most cer·
tain"of all the descriptions thus far. As far as Theaetetus is concerned, we have succeeded
in being so correct thus far without mentioning the extremely troublesome problemof

2(. Plato. Theaetetus. 183 e2 ft.

 Negation and Not-Being 191

"image". Thus far, no questions about not-being have arisen. The sophist as a deceptive
teacher of disputing is a definition that leaves us safe within the genus of doxa, within es­
tablished common opinion. But, as in both the Theaetetus and the Tractatus, the central
problem of how deception occurs still remains. The stranger seems to realize this lack by
responding to Theaetetus ' superlative (orthotata) by asking for a still "clearer model"
�IITadeigma) of this supposedly clearest of statements.
It is with this different "paradigm" that the dialogue takes its most drastic tum, a
change in direction that defines most precisely the difference between Wittgenstein's
Bild and the stranger's eileon, and which is ultimately responsible for the most decisive
difference in the two accounts of language - the status or "place" of not-being. It is from
md7ff. that the dialogue explicitly begins to tum on the twin issues of "image" and
"non-being". Nothing had as yet been suggested to explain how the sophist could de­
ceive youths with his "doxastic" knowledge, nothing especially about the nature of
logos. That defect is now remedied by the substitution of two infinitives, "making"
�iein ) and "doing" (dran) fUr those used thus far to define the sophist's activity,
"speaking" (legein) and "contradicting" (antilegein). The stranger does not mention, of
course, that hardly any sophist (except perhaps for Hippias) literally claimed to make all

things; the stranger himself says that the sophist only claims to be able to teach disputing
about all things (233 b ff. ); if is the stranger who now asks that we consider this as "real­
ly" a claim for making. It is finally because the stranger is implying that "speaking" is
"making" that the: products ofthe sophist'sworkcan ultimately be called " spoken images"
(eidola legomena) (234 c 6). We should also note here, with reference to legein, that no­
thing the strangeI' has said leads us to believe that only the sophist "makes" what he says.
The explicit characteristic of sophistry that permeates the following discussion is the
sophist's claim to StJ IUtclVta or :n:avta, to , . everything". That is the "joke" (paidia, a pun
on the earlier used paideia) which leads us still closer to the sophist, and it is that claim
which is so clearly false as to make a division of false from true images necessary. "Our"
speeches as well :lre "spoken images"; the uniqueness of the sophist'S images lies in their
deceptive claims for totality, not their eikastic nature.
Parenthetically, we should remember here that the stranger has not yet at all made clear
how this discussion relates to the dialogue's earlier and, there, apparently definitive men­
tion of "making" - at the very beginning of all the diaereses (219 b). According to the be­
ginning of this discussion (232 b) we have returned to the sophist as a "disputer", in the
ruth diaeresis, one which, of course, in the grand structure of all the first diaereses, is exp­
licitly divided off from "poetics" by being "under" the genus "capture" which is under
"acquiring" (ktetilee), in the very early divisions. Much, much later in the dialogue
(265 a-b) this confusion is explained directly. We are told there that "mimetics" as poetic
has "taken over'" the sophist <perieilephen), taken him over from ktetilee. We are in fact,
now (232 b) at the point of that "take over", where it becomes clear that the sophist, as an
antilogikon, cannot be understood unless logos itself is understood as made.
This "take over" means that the dialogue now concerns itself with made images and im­
itation (mimetilt�e) explicitly. And, of course, once mimetics is the issue, the stranger can
divide mat tech:ne into a mimetics whose products attempt only to copy the original
1 92 Robert B. Pippin

(eikestike, or "eikastics") and those which deliberately distort the pmportions of the
original and are thus false images but ones which appear true because of the "unfavora­
ble" position of the observer. It is this latter case, defined later as "appearing and seeming,
but not being. and of saying things but nottrue ones" (236 e) that raises the awsome ques­
tion of how the false can "be", or finally, how it is possible for to Jl.T) ltv dvm.
It is this connection between the poetic status of images and their "othc:rness" that gen­
erates the different discussion of negativity in Plato. Given Wittgenstein's logical
isomorphism. a picture in a sense mirrors exactly the logical structure of an atomic fact. a
characteristic that, as we saw, made it hard to specify the sense in which a picture was
"other" than the fact. But now, the whole point of calling an image made is to stress that it
is other than the original. And. here we mean by "other-than" much more than not-the­
same-as. We mean that, while in a number of respects it is like the original (it is after all an
image of it), in other respects it is not like it (it is only an image, limited in its reproducing
power in ways familiar since the Republic). It is this likeness-within-difference that will
require the first class of images, and the image making art- Ebt.uOtlXi). This art makes true
images. but only under the qualification of otherness.

� yaQ no\) I4LI4"ms no ("oC; 1:(� tonv. £l&iJi.wv I4Ev1:0l. <p�tv, ill'oux autiilv btaatwv.
(For imitation is a kind of production. howe ve r of images. not of e ach of the things the mselves.)
265 b l-3.
Further, it is the difference within likeness that makes qJuvtUOtLXll nel;essary. the art o f
making images which seem to be true. In the language Plato is wont to use in explaining
this aspect of otherness. the image is, at the very least, one remove from reality.
But this all means that it is the problem of the mode of being of the image that sets the
stage for the classic, recently much discussed parricide of Parmenides, and the limited re­
ality the ES gives not-being. We now need to explain, not just the operation of negation in
the formal mode as a meaningful linguistic activity; we need to explain the specific way in
which what is "other" can be27; we need an account of the way in which the form of
otherness is woven together with "what-is" and especially with logos and opinion.
Without this intermingling the image would be indeterminately other th;;m or trivially dif­
ferent from its original. As we have stressed, it must be determinately other, either as the
image ofthe original, or the image of what is other than the original, but seems to be of it.
It is this characterization of false (not negated) images that raises the question of the being
of otherness in a primary "ontological" way. And it is this characterization that must be

27 On this point, cf. Owen's intricate study, Plato on Not-Being. Owen notes l:he problem of "im­
ages" (p. 259) , but in general his account of not-being is too "Wittgensteinean". The claim that
the ES's explanation of ne gating being is made "from analogy" with negating any predicate, once
the existential inte rpre tation of "is" is excluded in favor of a thoroughgoing "incomplete " sense,
ignores too much of the text. The alternatives are not simply the "existence" of some non-being,
or the denial, or elliptical denial, of some predicate. As Lee points out, Pla.to on Negation and
Not-Being in the Sophist, Philosophical Review, v. St, 1972 , p.270, Owen's analysis is "too
weak", it stops shon of discussing the Pans of Othe rness section (257c-258c) whe re a claim for
the dete rminate "reality" (although, of course, not the existence) of "other -than-X" is main­
tained.
 Negation and Not-Being 193

defended by showing how otherness itself is woven together with being throughout ev­
erything, especially with spoken and written images.
While it would require obviously more detail than can here be presented to follow
tb rough this conception of ontology, we can note that it results in what we can call a "du­

al" conception of being, of a thing's being-what-it-is, or its being "truly", if we borrow

Kahn's formulation of the primary sense of ElvaL 28. This requires that the determinacy or
oneness of any being is possible only if, on the same ontological plane, that being is self­
identical and "other", is what-it-is by also having parts or elements which determinately
exclude it from other beings.
We cannot here pursue such difficult matters as symploke, the megiste gene, and the
ES's full definition of on. Our initial point is to show that when some version of a corres­
pondence model of truth and meaning is invoked to explain language's relation to the
world. then, first, the question of intentionality (the question of "making" in the Soph­
ist) must arise, together with the topic of otherness, ofthe image's distortion as well as re­
production. Finally, this mode of otherness, not reducible to the indeterminate, direct
exclusion of nega.tion. requires on account of the being of otherness within an account of
the being (especially the "being true") of anything.
But, with the issue raised in this way, some indications can be given of what such a full
ontological account of otherness would be like. at least within the limits of the topic we
are pursuing. In short, we can ask, what can the ES mean, when he says at 258 a 8

. . . E1tEbtEQ � itaT€Qou <p�<no E<PUvl1 TWV IIvtwv ouoa, b�e[VTJt; li E oU011t; uv6YXl1 Iii} 'Kat to.
�6Qta aU't'I]t; !111IiEv6<; Irnov lIvTa 'nittvm
( . . . since the nature ofthe other is among the beings and, with it being, it is necessary to ascribe no
less being to its p arts .)

Even though the ES is clear at 2 5 8 c 6 ff. that he does not mean by this the "opposite of
being" exists. he is also clear that even though the other is (partakes of being) it is other
than being (not the same as being) and forms with it an intermingling dyad throughout all
the forms. In short. the first result of this characterization of the reality of otherness. is a
systematic deni;L! of atomism, of any semantic or ontological variety29.

'" Charles Kahn, The Greek Verb 'To Be' and the Concept of Being, Foundations of Language (2),
1966. especially p. 260.
19 This agrees with Moravcsik's analysis in Being and Meaning in the Sophist, Acta Philosophica
Fennica, vol. XIV. 1962, as well as with much of his clarification of the difference between the on­
tological and s<�mantic issues at stake in the Sophist. (Cf. especially his definition of falsehood on
p. 41). But, he shares a problem with Owen. Immediately after informing us that meaning must
be explained i�Ldependently of truth, Mo rav csik claims Fa, if false, can only mean that -Fa. As a
result, even though he grants the "reality" of negative predicates, it is not clear whether or not
they are just negated predicates (pp. 65-73), re-raising confusions between formal and material
modes. See especially p. 52, where 263 b 1 1-12. " . . . in relation to each thing, much is and much is
not . . . " is translated into "of each entity much can be predicated affirmatively and negatively".
Some of the s:une ambiguities are shared by Lee's "ope rational" sense for "other-than", again,
even though the other-than-X is given intensional legitima cy. Cf. Lee, Plato on Negation, p. 293
and especially p. 296.
194 Robert B . Pippin

This notion of the constitutive ontological role of otherness can be expanded by notic­
ing our results thus far. We have claimed that the poetic status of images (as opposed to
Wittgenstein's pictures), the difference between falsity and negation, the "seeming art",
and aporiai in Wittgenstein's own account, force us to consider the issue of notbeing as
involving more than the correct use of "not", more than distinguishing between denials
of identity and denials of predication. Throughout that account, one point was continu­
ally stressed - explaining the sense of negative expressions, while a difficult and tricky
task in itself, will not explain the sense of false expressions. Again, the Sophist does not
say -p and challenge us to figure out how; neither does he merely say p �, confident of
our inability to explain some error we know he has committed. As we have shown
throughout, what he does assert are "images", and their mode of being is the crucial issue.
They are certainly not "unreal", as "other than the real", nor are they just indetermi­
nately other, not the same as their originals. In short, the "image" is for logos what
doxa is for political reality and what phainomena are for the whole, the "mix" of being
and not-being, of determinacy and indetenninacy, the constant Platonic duality so enig­
matically entitled 1:0 ev and "to aOQlatas Mas in the Philebu$.
To summarize one final time, to say something false, according to th(: ES, is, but is not
only, to "do" something incorrectly. That is, to claim that otherness is woven together
with all the forms does not only mean that it can make sense to say that something isn't,
since we only mean by that that it isn't something else. For the referent of a false image to
make sense independent of negation, it must also mean that any thing's being what-it-is
cannot be determinate as a monadic, self-same unit, as a pure "one". Its being depends on
the way it excludes anything else from its place in "logical space". This other-than-x is
neither the trivial denial of identity, nor the positing of some specific contrary (the
non-X). It is not the former because X, by being what-it-is must specifically define and
exclude its other (just as in the Parts of Otherness section, the not-large is not meant to in­
clude "green" or "just", but just those things not-large). It is not the limer, since some­
thing need not be a contrary to be other-than-X in this detenninate :;ense ("medium"
does as well as " small" for not-large)3o.
Now, it should be immediately noted that this discussion of onte.logical otherness
needs a good deal more explanation, but we can note here some initial problems before
going any further. In the first place, as Rosen's recent Hegelian treatment of Plato points
out3 1, if this requirement for "detenninateness only by virtue of otherness" is applied to
Being itself, Being must be indetenninate, or at least unspeakable. There is no "other" for
being, no 1:0 !J.TJbaIJ.OJSOv, in Hegelian tenns, no Nichtsl2. Thus Being itself would not be

30 We are thus disagreeing strongly with P. Seligman in Being and Not-Being (rhe Hague, Nijhoff,
1974), p. 1 17. Far hom having no alternative to the Parmenidean to fJ.'ljl\al.l(il� QV or indefiniten�
gation, and far from confusing explanations of negation with explanations of falsity, the ES is
concerned precisely to avoid just such confusions.
3 1 See Rosen's chapter on "Contradiction in Plato and Aristotle", in G. W. F. Hegel: An introdll"
tion to the Science of Wisdom (New Haven, Yale University Press, 1974), pp. 64-88.
II Similar arguments against the possibility of a univocal or "fundamental" ontology are broached
in Aristotle, Metaphysics, 1001 aff.
 Negation and Not-Being 195

definable or delim itable. In line with this, any particular other-than-x, while genuinely
other, is still a being, indeed it is the form of other-than-X. If otherness in this sense is
just a further constituent of what it means to be, it is hard to appreciate how any danger­
ous parricide of Parmenides has taken place.
And, there are problems with this notion of "image". For, the image is still something
made, and, exactly as in Wittgenstein, once this function of its otherness is stressed, the
"conditions" for its being, for its determinacy, shift inevitably to the semantic rules for
that making, however much, since the image is still eikastic, still has "originals", the trad­
itional conception of those conditions or "forms" is maintained. Now this equivocation
in Plato between forms as ltUQUOEtYl1utU or being-itself, and forms as functions, or rules
lordiaeresis, between "being" and "truth", or even between the so-called middle and late
periods, is well known33• Both elements of this formal and material analysis are present in
the Sophist, and because of our comparison with Wittgenstein, we have stressed some
ontological aspeclts of the problem of not-being. But it should not be denied that the more
we stress the oth!:rness of images since made, the more difficult it becomes to retain any
imaging "originals" talk, or finally, any ontology.
This difficulty is dramatically underscored by a strange, and as far as I can tell un­
noticed inconsistc�ncy in the ES's overall account. That is, even if the mode of being of the
image and thus the possibility for iconic and phantastic is explained as it is in the Sophist,
we are still eft wondering about our " access" to the "original" for "images", and, indeed
what those originals might be. Throughout the dialogue, both our ability to compare
image and original, and the status of the original itself seem unproblematic for the
stranger. Early on in the discussion of the sophist as imitator, what he imitates are said,
unqualifiedly, to be "beings". (Em( tLs I1LI1'lTils rov t<'iw ovtrov 235 a 1). We wonder, im­
mediately, though, how the sophist could "see" ta onta in any sense in order to be able to
make an image of them, even a deceptive one34• This is especially odd in the light of the
last grand diaeresis of the sophist, which places him clearly in the class of those who do
not know (hoi d'ouk eidotes) and do not, in any sense, "have " the "beings". The stranger
will even cautiolllsly coin a new word for the sophist's unique kind of imitation - doxo­
mimetike (267 d), a name which clearly indicates that "what" the sophist imitates are not
at all ta onta, as is clearly stated at 235 a, but that he is an imitator of doxa, which itself
stands in relation to the true much as an image does, both like and unlike. In shon, from
235a to 267 d, the characterization of the sophist as a particular kind of image-maker has
shifted from that of a man who, in imaging the real, deliberately distorts it into phan-
145mata (said thc�re to be done for the sake of beauty, for kalas summetrias) to, now, a
man who does not know the real and whose images are defective not primarily because of

their distonion of the real, but because, as images, they only image doxa. In the first de­
finition of phantastike, the sophist still clearly " looked at" the original, to aA'I]{}ts, but

Jl Cf. Gerhart Huber's study of the problem in Das Sein und lias Absolute (Basel, 1955). Peck's re­
solution of the ambiguity in Plata and the Megiste Gene af the Sophist, Classical Quarterly,
No.46, 1952, seems far too one-sided.
34 CE. Republic, 598 a ff. and Philebus, 39 a ff.
1 96 Robert B. Pippin

produced, instead of an "exact" copy, one distorted for the sake of beauty . Now, of
course, the sophist's access to the original vanishes, leaving him, and perhaps us, with juSt
images of doxa35• In any event, this clear shift in describing the phanta.stic nature of the
sophist from him as a mimetes ton onton to him as a doxomimetes raises more than a few
problems with regard to what the stranger thinks the original is for any of our eidola
legomena and what access we have to them. (This is all not to say that the stranger's, and
W ittgenstein 's for that matter, ontology merely require a "critique of reason", but it is to
say that for both there may finally be no ontology, no possibility of ,I real doctrine of
primary "being" in any sense.)
Of course, if the above is correct, then, while the stranger's explanation might reveal
how false images are false, it does not go very far in explaining how we lcould ever know
or discover any given image were false, precisely because there is no explanation in the
dialogue of either what exactly is the original for spoken images, or how we could know
those originals except, again, through further images. In other words, the stranger's long
and tedious education of Theaetetus would seem, amazingly, to be of little use in instruct­
ing Theaetetus in "unmasking" any sophist, in being able to recognize, much less refute a
sophist. (In that sense, the stranger's JtuLOE(a might be a Jtmo(a.) To do so would re­
quire, to use the stranger's own example at the end of the dialogue, that Theaetetus could
know the true "form" (schema) of justice and "present" it discursively, a possibility ab­
out which we are nowhere enlightened. Finally, then, in a strange and '''fantastic'' way, it
could be said that this dialogue's doctrine of images, however superior in some ways to
Wittgenstein's account of "pictures", critically "lacks" a doctrine of originals, much as
the dialogue Sophist itself lacks its counterpart, mentioned but never presented through­
out this trilogy, or anywhere else in the Platonic corpus, the dialogue Philosopher, a final
incompleteness, or absence which may always be the beginning and end of any
"Platonism" .

35 In fact, this last diaeresis is even more confusing in its definition of phantastike, since we now
find under this dissection of the art, supposedly the class of all false images (266 e 3). an imitator
who does know the original, and produces. still under phantastike what seems to be an icon. not
a phantasm (267 b t l).