The Rhetoric of Jacques Derrida I: Plato's PharmacyAuthor(s): Yoav RinonSource: The Review of Metaphysics, Vol. 46, No. 2 (Dec., 1992), pp. 369-386Published by: Philosophy Education Society Inc.Stable URL: http://www.jstor.org/stable/20129336 .Accessed: 16/05/2011 07:41

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unlessyou have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and youmay use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at .http://www.jstor.org/action/showPublisher?publisherCode=pes. .

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printedpage of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact support@jstor.org.

Philosophy Education Society Inc. is collaborating with JSTOR to digitize, preserve and extend access to TheReview of Metaphysics.

http://www.jstor.org

THE RHETORIC OF JACQUES DERRIDA I: PLATO'S PHARMACY*

YOAV RINON

-Although a great variety of topics are discussed in Derrida's

philosophical writings, a central theme recurs in many of them: the

relationship between speech and writing. Derrida consistently uses

the same methods to deal with this topic, and my reading aims to

expose the regulation of these methods. This essay tries to point out the blurring moments of the strategy which lead to one of Der

rida's most outrageous outcomes, which is that writing precedes

speech. This notion, however, is only the starting point; its con

sequences are the impossibility of communication and the collapse of the Platonic maxims. Such successful moments of deconstruction

are traced back to their origins so as to leave bare the devices on

which they are based. It will then be possible to discern a specific

recurring stage during which occurs an illegitimate movement ac

cording to the Derridian rules of the game.

Derrida's discussion of the Phaedrus begins at the "geograph ical" center of the dialogue (275c) with the deprecation of the

profession of logography. The logographer, who writes orations for

trials in which he himself does not appear, represents, for Derrida, the intersection of two crucial phenomena: the presence of the ab

sence (the writer of the speech is present only by means of his own

cited words, while being physically absent from the trial), and the

gap between writing and truth. The logographer, he says,

in the strict sense, composes speeches for use by litigants; speeches which he himself does not pronounce, which he does not attend, so to

speak, in person, and which produce their effects in his absence. In

writing what he does not speak, what he would never say and would

* This is the first of a series of two articles. The second article will appear in the March 1993 issue of the Review of Metaphysics.

Review of Metaphysics 46 (December 1992): 369-386. Copyright ? 1992 by the Review of Metaphysics

370 YOAV RI?ON

never think in truth, the author of the written speech is already en trenched in the posture of the sophist: the man of non-presence and non-truth. (76; 68)1

At this point Derrida follows Plato, who temporarily abandons

the topic of writing (274b), and also leaves the problem of absence

and truth for a different subject, the kidnapping of Orithyia in the middle of a game with Pharmacia (229b). Pharmacia is a link (une

maille) between the kidnapping, which ends in rape and death, and

the reappearance of writing in a later stage of the dialogue. Here,

the connection between Pharmacia and the Greek word (fr?ppanov is

important:

Pharmacia is also a common noun signifying the administration of the $ap\jLOLKov, the drug: the medicine and/or poison. ... A little farther on, Socrates compares the written texts Phaedrus has brought

with him to a drug [^?ppaKov]. This 4>?ppo?Kov, this "medicine," this

philter, both remedy and poison at the same time [? la fois], already introduces itself, with all its ambivalence, into the body of the dis course. (78; 70)

For Derrida, the fyappaKov is only an element in the chain of

significations (108; 95), whose interplay constitutes the textual phe nomenon. It is impossible, however, to try to analyze each of the

elements in isolation. This kind of interpretation, according to

Derrida, would damage the subtle texture of the literary object in

a most vulgar way. Sharp distinctions, he claims, are unacceptable

in dealing with language:

It is always possible to think that if Plato did not realize [n'a pas pratiqu?] certain options [passages] and even actively barred them from being realized [les a m?me interrompus], it is because he perceived them but left them in the domain of the potential [dans l'impraticable]. Such a formulation is possible if one avoids all reference to the dif ference between conscious and unconscious, voluntary and involuntary, a most vulgar means [instrument fort grossier] when one comes to deal

with language. (109; 96)

The borders, however, between the conscious and the uncon

scious, the voluntary and the involuntary, are stressed here, only to

1 Page numbers refer to Jacques Derrida, La diss?mination (Paris:

Seuil, 1967); and to Jacques Derrida, Dissemination, trans. Barbara Johnson

(Chicago: University of Chicago Press, 1981). The left number refers to the French original, the right to the English translation. The citation

given above is a modified version of Johnson's translation, and all the translations of Derrida in this essay (unless otherwise noted) are of the same type.

THE RHETORIC OF DERRIDA 371

be blurred later on. It is not that the limits are important in and

of themselves; their only significance stems from being targets for

deconstructive assaults. In other words, from a Derridian point of

view, the meaning of borders lies in the realization of their destruc

tive potential. This aspect of his reading is vividly expressed in his discussion of L?on Robin's French translation2 of the Greek word

(j)?ppoLKov in the Platonic myth of Theuth.

Toward the end of the dialogue (274c5-275b2.), Socrates utilizes the Egyptian myth of Theuth to illustrate his own arguments against

writing. In this myth, a conversation is held between the king and

one of the minor deities, Theuth. At a certain point, the new in

vention of writing is defined by the latter as "a (p?ppanov of wisdom

and truth." When Robin translated this expression he chose the

word remede (remedy) as the French equivalent of the Greek

(j)6?ppaKov. At first glance, this choice seems quite reasonable; the

presentation of a new invention should concentrate on its beneficial

aspects if the inventor wants it to be accepted. Derrida, however, stresses that such a translation erases the ambiguity of the Greek

original and with it the possibility of understanding the context

(109; 97). In this dialogue, the meaning of (fr?ppanov as poison is no

less important than its opposite, because it represents the point of

view of the other interlocutor, the king. Writing will be a means

of forgetting by giving the illusion of memory and wisdom. People will not know more as a result of the new instrument; on the con

trary, they will know less, because of the false beliefs which are the

inevitable outcome of writing. Each of the participants in the dia

logue about writing thus emphasizes one of the polar signifieds of

the signifier cfr?ppaKov, while the text, which consists of both atti

tudes, defers from choosing either the one or the other (109; 97).

Thus, in that which is named "text," the border between op

posites is blurred, and what seemed to be an either-or situation

turns out to be a both-and situation. The cpappaKov is at the same

time (d la fois) remedy and poison, good and bad, beneficial and harmful. The meaning of the text can be extricated only on the

basis of this mutual coexistence of contradictions. The dialogue between the king and Theuth, held within the unity of the same

signifier (dans Vunit? du m?me signifiant), is proof of its being a

2 Plato, Ph?dre, trans. L?on Robin (Paris: Les Belles Lettres, 1947).

372 YOAV RI?ON

unity of opposites (111; 98). Consequently, any intention to choose

between the different signifieds (which means a reestablishment of order and limits) is ipso facto a neutralization of the unique textual

quality of the object under discussion:

When a word inscribes itself as the citation of another sense of the same word, when the textual predominance [l'avant-sc?ne textuelle] of the word (fr?ppanov, although signifying "remedy," cites, re-cites and activates the possibility of reading that which within the same word

signifies on a different level and in a different profundity of the scene, poison . . . , the choice of only one of the renditions by the translator has as its first effect the neutralization of the citational play, of the

"anagram," and, in the end, quite simply of the very textuality of the translated text (111; 98).

The above citation indicates an important point in Derrida's

approach. When one (for example, Robin) erases "poison" on behalf

of "medicine," when one chooses a single option and gives it, by

definition, even a limited predominance within a section of a given

text, what is named "text" becomes a hierarchical phenomenon.

According to Derrida, the inevitable outcome of the decision to

choose is the deprivation of the text's own textuality, since in such

a delicate object even the slightest hint of hierarchy means an im

mediate loss of the text's most precious characteristic: the plurality

of its potentials. This statement, however, gives an important hint

to the critic of Derrida. If one would be able to prove that the

deconstructive strategy is generally based on the notion of hierarchy

and that Derrida himself assumes the existence of hierarchy and

utilizes it for his own needs, the very possibility of a deconstructive

reading would have to be seriously modified, to say nothing of the

process of reading itself. To indicate not only the existence, but

also the importance, of that assumption in Derrida's philosophy, I

shall now proceed to the <j>appaKos.

The (frappoiKos, the scapegoat in Greek religion, is intimately

connected to recurring themes in Derrida's interpretation of the

Phaedrus. Like the (fr?ppanov, it represents a both-and phenomenon:

it is both inside the city, being raised and nourished by it; and outside, since it must be exiled at a certain time. It is both a remedy, as

the city's existence in a time of crisis, especially in time of plague,

depends on it; and a poison, since that kind of existence is an outcome

of the scapegoat's expulsion and, sometimes, death. Moreover, the

(frappan'os is strongly connected with the dialogue's protagonist, So

THE RHETORIC OF DERRIDA 373

crates, both metonymically and metaphorically: metonymically,

since Socrates was born on the day of the (frappaicos's expulsion;

metaphorically, because Socrates was an essential part of Athens

and as such was executed. Thus, he was both citizen and outsider,

and his personality was the realization of the unity of contradictions.

At the same time he was ugly (in his features) and beautiful (in his

soul), honored and despised, loved and hated, close and at a distance,

knowing all and nothing, alive and dead, a remedy and a poison.

The main problem with the above link is its absence from the Pla tonic text. The signifier, the word (frappaubs, is absent not only from

the dialogue, but also from the Platonic corpus as a whole. Derrida,

who is quite aware of the significance of this kind of movement,

gives a thorough explanation for it:

But what does absent or present mean here? Like any text, the text of "Plato" could not not be in connection with [?tre en rapport], at least in a potential [virtuelle], dynamic, lateral manner, all the words that

composed the system of the Greek language. Certain forces of as sociation unite?at diverse distances, with different strengths and

according to disparate paths?the words "actually present" in a dis course with all the other words in the lexical system, whether or not

they appear as "words," that is, as relative verbal units in such dis course. (148; 129-30)

This explanation does not stem merely from a polemical need.

The insertion of a word which is absent from the text can be accom

plished only with great difficulty, if one remembers how delicate the Derridian tissue is. Since the absence of the "presence" of a word

in the text stopped functioning as a barrier for the external signifiers,

there is a serious risk of opening the text to an unrestricted inter

pretation that might enable the entrance of any word into the fragile

web. In other words, since the mere fact that Plato did not write

a certain word in a certain text?for example, (frappotKos?is not a

sufficient condition for preventing the word from functioning in the act of reading, the whole dictionary is poised to flood the text with an endless stream of words. To avoid demolition, a new principle

must immediately be substituted for the old one, and this principle concerns the nature of the connection between different elements

of what Derrida calls "text."

Not every word, according to Derrida, is a legitimate participant in textual activity. Only those words which, though actually absent from the text, are connected with it associatively and united with

its present words by means of combining forces, can participate in

374 YOAV RI?ON

the reading process. This condition can also solve the problem of

absence resulting from unconscious trends which cause the suppres

sion of words like cfrappaKos. The deconstructive movements shed

light upon this involuntary activity and, by reconstituting the sup

pressed elements, better explain the object under discussion.

It is this limitation, however, which demonstrates the impos

sibility of the already mentioned necessary deconstructive condition,

namely, that the text must be treated as an ahierarchical phenom enon. The mere admission of the fact that not every word may

enter into the process of interpretation is an admission of the fact

that the text is a hierarchical phenomenon. If there is exclusion,

there must also be hierarchy; some words are more important than

others, more "in" than "out." They stand on one side of the border,

while others stand on the other side. This hierarchy is implied by the very words Derrida chose to realize in his own text: "potential

manner," "diverse distances," "different strengths." True, many

readers might consider the existence of the limits as obvious, if not

banal; it is exactly this kind of banality and obviousness, however, that hinders Derrida's critics from seeing the possibilities of op

position. From a hierarchical perspective, which is that of Derrida's

rivals, the border exists by definition, and therefore there is no need

to prove its existence. Such a perspective, however, turns out to be

one of the great misunderstandings of the rules of the game. Noth

ing exists by definition since nothing is and nothing can be defined.

Only upon complete acceptance of this rule is it possible to see that

Derrida himself can not play according to the rules of his own game.

What is more, the difficulty in exposing an internal contradiction

between two necessary deconstructive conditions (the text as an

ahierarchical phenomenon and the importance of hierarchy to any

Derridian reading) stresses the fact that the existence of the border

is not so banal and obvious as it seems to be at first.

In order to present more vividly the existence of the limits in

Derrida's reading, I shall now turn to another of Plato's dialogues,

Phaedo, in which the fyappaKov plays a central role. The dialogue

deals with the last hours of Socrates and with his death by means

of the (frappaKOP. Let us examine all the places where this word

appears in the dialogue:

Have you been with Socrates yourself, Phaedo, when he drank the

(frappaKov in prison? (57al-2) . . . except that after drinking the fyappaKov he [Socrates] died. (57b2

3)

THE RHETORIC OF DERRIDA 375

"Well, definitely nothing, Socrates," said Crito, "except that for some time the one who is about to give you the fyappaKov says that you should be told to speak as little as possible. For he claims that those

who speak get warmer, and this impairs the <?)?tppaKov's activity." (63d6-e2)

For it seems to me that it would be better to drink the fyappaKov after

having a bath, so that the women will not have the burden of bathing the corpse. (115a7-8)

Despite the fact that for a long time I have claimed for many reasons that after I will drink the fyappaKov, I will not stay with you at all . . . (115d2-3)

Socrates, he said, as far as I am concerned, I will not condemn you, as I condemn all the others who are angry with me and curse me after I inform them that, according to the rulers' decisions, it is time for them to drink the fyappaKov. (116cl-4)

Well, Crito, let's obey him, and that somebody will bring the (frappaicov if it is ready. (116d7-9)

And the slave went out . . . and came back bringing the one who is about to give the (fr?ppanov. (117a5-6)

. . . and the one who gave the (frappanov. (117e6)3

Robin's reasons for translating <\>appaKov as poison in this con

text can easily be detected. The (frappanov caused Socrates' death,

and therefore it is reasonable to concentrate on its poisonous aspects.

Derrida, however, calls his readers' attention to the duality of the

(fr?ppoiKov even in this case. The (?xxppaKov which is given to Socrates

as poison is also responsible for the immortality of his soul (144-5;

126-7). Thus, the (p?ppaicov brings death and enables immortality

at the same time; it is poison-medicine and not poison or medicine.

The justification for stressing this ambiguity is based on the context

within which the word appears: Phaedo is a dialogue about the im

mortality of the soul, held on the verge of Socrates' death. In other

words, the unity of contradictions is reaffirmed by the unity of the

living speech and the death scene. The use of this kind of argu

mentation is neither surprising nor uncommon; on the contrary, the

importance of the context is quite conventional where matters of

interpretation are concerned. Something is hidden behind the in

nocent veil of convention, however, and not without cause.

3 The translations are based on Plato, Opera, ed. John Burnet, 5 vols.

(Oxford: Clarendon Press, 1900-1907). All the Greek translations in this

essay, unless otherwise noted, are my own.

376 YOAV RI?ON

The same context which allows the addition of another signified

(medicine) to the one given by the French translator Robin (poison) excludes yet another signified of the word fyappaKov: "paint." This

option, which appears in the LSJ dictionary4 as a possible meaning

of (?)?ppaKov, does not take a place in the Derridian text, not as a

result of a deconstructive repression, but due to the indifference of

the context. In the Phaedo the Platonic text creates an environment

which excludes the realization of "paint" as a signified. In other

words, within the boundaries of this dialogue, the signified "paint"

does not have the same intensity as the signifieds "poison" and

"medicine." Thus, the context serves as the borderline, and con

sequently hints at a crucial internal contradiction within the Der

ridian strategy: either the context is indispensable, and the text is

revealed as a hierarchical phenomenon, or the context is irrelevant,

and the fragile texture of the text is crushed by vulgar intrusions

of signifieds. Neither of these cases, however, is allowed according

to the Derridian rules of the game, which demand that the text be

both ahierarchical and confined within the limits of the context.

To strengthen and illuminate my critique, I shall now proceed

to a more complex example taken from another Platonic dialogue,

Cratylus. In this dialogue, Socrates talks about the connection be

tween words and their meanings, or, to use modern terminology,

between signifiers and signifieds. At a certain stage of the discus

sion, Socrates depicts a possible confusion:

However, a variety [iroLKLkeiv] of syllables is possible so that to the

layman they [the names] might seem different from each other, al

though ontologically they are the same [r? avr? ovra]; just as for us

the (j)?ppaKa of the physicians, being disguised [ireKOLKLkpeva]hy paints [Xp paoLv] and smells seem different, although ontologically they are

the same [r? avr? ovra]. However, as far as the physician is con

cerned, since he examines the power of the (fr?ppana, they seem the same [r? avr?], and he is not misled by the additions. (394a5-b2)

In this passage the physician represents the man who is a

professional in ontology. The cfrappaica are portrayed as identical

in their essence, in their "onticity" (r? avr? ovra), yet they are

colored by paints which make them seem different from one another.

Unlike Socrates and Cratylus, who cannot perceive the identical

4 Henry George Liddell and Robert Scott, Greek-English Lexicon, re

vised by Henry Stuart Jony.

THE RHETORIC OF DERRIDA 377

essences, being misled by the colors, the physician can recognize the

activity of the 4>appaKa regardless of their appearance. The profes

sionally of the physician will later have a central part in this dis

cussion. At present, I prefer, like Plato, to use it as a model for the

professionality of the linguist, who is able to differentiate between

the variety of signifiers referring to the same signified.

The signifier 4>appaKov has several signifieds, among which are

"drug" and "paint." When Socrates chooses to speak about the

"(j)?ppaKa of the physicians," the additional words "of the physi cians" create a hierarchical context within which "drug" has a higher

position than "paint." That is, speaking "of the physicians" gives

priority to drugs over paints. The Cratylus is, however, a dialogue

about linguists and not physicians, and its essential distinctions

concern words (see, for example, 423-4) and not drugs. The above

quotation is, therefore, located in a verbal context; this phenomenon cannot be ignored by any Derridian reading, as will be shown by the following.

The signified "paint" has an indispensible meaning within this

verbal context. Socrates explicitly states that the main danger faced

by nonprofessionals in the field of linguistics is that the signifieds, colored by different signifiers, will create ontological confusion, and

that one will not be able to perceive the identity hidden under the

cover of variation. At first, one may have the impression of an

unavoidable deconstruction. The message of the above citation is

centered upon the possibility of distinguishing between the different

signifieds of the different signifiers. On the one hand, ttolkl\ lv,

TceiroLKiXpeva, and Xp paaiv signify "paint"; on the other hand,

(?)?ppaKov signifies "drug." The context, however, tends to impede the realization of this distinction by loudly citing the association

between (jyappanov and "paint." An association of this type would

blur the boundaries not only between the different signifieds as such

("paint" would become a possible signified of (frappanov as well) but

also between the hierarchical values attached to them. The minute

that "drug" equals "paint," the minute one cannot assign a higher

position to "drug" than to "paint" within the context of "(?appana of the physicians," the whole hierarchical construct collapses, drag

ging with it the possibility of meaning. It seems that I have come to a dead end. At the beginning of

the discussion of Cratylus I claimed that it is possible to discern the boundaries and hierarchies both within and between signifiers and

378 YOAV RI?ON

signifieds; yet at this stage the associative connections (whose rel

evance is beyond any doubt) weave a web of contexts which refute

the above claim. Although it is not an easy task to point out the

misleading aspects of such an ambiguous reading, it is, nevertheless,

not impossible. If one concentrates on the signifieds' importance,

the force of the selection of the signifiers is forgotten; the citation,

which is loaded with signifiers hinting at "paint," never utilizes the

<j)?ppaKov as a signifier for "paint," and not without reason. The

selection of signifiers helps to avoid confusion; the variety of sig

nifiers for "paint" proves the significance not only of the present

realized options but also of the potential absent ones. Choosing not

to utilize (p?ppaKov as a signifier for "paint" preserves the hierarchy

between the inside and the outside, the signifier and the signified,

and reasserts the boundary of difference. The different signifieds,

"paint" on the one hand and "drug" on the other, have different

signifiers: ttolklXuv, TteiroiKikpeva, Xpicpaaiv for the one, and (pappanov

for the other. Plato thus reveals himself as being open to the risk

of a deconstructive reading, and as being able to resist it at the same

time. Further backing for this notion is found in another passage

from the same dialogue, where Socrates says, "Like the painters

who want to portray something, sometimes they add only the purple

pigment and sometimes another of the (p?ppana" (424d7-el). There

is almost no possibility of mistake here. It is clear that the context

is dominated by "colors," and Plato does not hesitate to use (p?ppaKov as the signifier for "paint." The firmer the limits between the sig

nifieds ("paint" is predominant over "drug," "poison," and "medi

cine"), the wider the liberty of choosing within the group of asso

ciated signifiers (fyappaKov, Xpcopa, -KeKouaXpeva).

The deconstructive reading is based upon an ahierarchical per

spective on the text. Although the above example tries to show the

impossibility of that perspective, it is possible that Plato is an ex

ception, that is, that the Platonic text qua text is unique in having hierarchical aspects. This possibility could be confirmed by giving as many examples as possible of the ahierarchical perception of the

text. Despite being a rather conventional way of verification, how

ever, this is also a defective one: it overextends the limits of discus

sion, and, even worse, it is always prone to refutation by unexamined

texts, the number of which is always higher than that of the ex

amined texts. A different method must therefore be put into prac

tice. It is the giving of one example of an ahierarchical text and

THE RHETORIC OF DERRIDA 379

proving its uniqueness; by proving the ahierarchical text to be the

exception, the normativity of hierarchy will be reaffirmed.

At this point in the analysis, the reader might wonder whether

such a text indeed exists, and, from the point of view of meaningful

differences and limitations, if there is a possibility of realizing such a text within the boundaries of a book. It is certainly not an easy

task to find a text completely fulfilling this deconstructive condition, but a consideration of the most recurring text within the Derridian

corpus will give us the answer immediately. This text, which is

ahierarchical by definition, is not philosophical but linguistic; it is the dictionary. The order of the words in the dictionary is mean

ingless from an interpretative point of view; the fact that signifiers

beginning with "a" precede those beginning with "b" does not say

anything about the precedence of the word "apple" over the word

"bee." The same can be claimed about the signifieds: their order

reflects the frequency of use but not any meaningful priority ("poi

son" is not ipso facto more significant than "paint"). The dictionary

is the deconstructive text par excellence. It is the emblem of equal

ity, the unity of signifiers and signifieds, the origin and the arche of

all the texts; and therefore it has unlimited fertile possibilities. Yet it is also the most barren of all texts because it never stops being a

potential and only a potential. The dictionary can never be a re

alization of its own copious options because any realization requires

the abandonment of at least one of the alternatives. The dictionary

is, no doubt, the Derridian both-and phenomenon. It is everything and nothing, all existing contexts yet not a text in itself; it is full

of citations without being able to say something of its own. If the

dictionary "means" something, it means no more than the possibility

of meaning.

Thus, in being ahierarchical, in being extremely both-and, the

dictionary is fundamentally different from all other texts. Any

text, qua text, is sl choice, is the exclusion of some contexts and the

inclusion of others. The text is an inevitable selection, for good or

bad, from the choices given in the dictionary. The text, therefore,

is an either-or phenomenon, a hierarchical phenomenon. But let

us now go back to the Derridian characterization of the (pappanov.

In a short passage Derrida refers to the appearance of the (pappanov

in the real world:

Sperm, water, ink, paint, perfumed dye: the fyappaKov always pene trates like a liquid; it is drunk, absorbed, introduced into the

380 YOAV RI?ON

inside.... In liquid, opposites are easily mixed. Liquid is the element of the <j)?ppaKov. And water, pure liquidity, is most easily and dan

gerously penetrated then corrupted by the </>?ppaKov, with which it mixes and immediately unites. (175; 152)

At first glance, the deconstructive rigidity tends to evade the reader's

eye; a second examination, however, reveals a curious signifier within

the textual tissue: "always." If, in the works of Plato, the 4>appaKov

does not always penetrate like a liquid, then a meaningful difference

exists between the Derridian (pappaicov and the Platonic one. The

word "always" hints at a total condition whose importance to Der

rida's reading is mentioned in the above passage. It gives another

indication of the unity of opposites. It sustains the fact that bound

aries tend to blur in Plato's writings, and it leads Derrida toward

the simultaneous interpretation of the cp?ppaicov as poison-medicine.

Any flaw in the integrity of this signifier would generate a crack in the whole of the deconstructive argumentation, as it would signal the possibility of another strategy of reading based on premises other than the Derridian ones. This different (p?ppaicov does appear

in the Platonic corpus, as the following passage proves:

Anyhow, when he asked me [Socrates] whether I know the cpappaKov for the head, I answered with great difficulty that I knew it. "Well,

what is it?" he [Charmides] asked. And I said that the thing itself

[avro] is a leaf, and that there is some kind of an incantation in addition to the (pappaKov which, being said with the practice of the thing itself

[avrb], the fyappanov completely heals the man, and without this in

cantation, the leaf is completely useless.5

The deconstructive liquidity realized in sperm, water, ink, paint

and perfumed dye stands in opposition to the Platonic solidity of the leaf. The firmness of the leaf, however, can easily be dissolved

in water, as in tea, and therefore, one might comment, my passage

does not contradict the deconstructive reading. My answer goes

back to the difference between potentiality and realization. It is

always possible to liquify a substance; but one does not always decide to put that option into practice. When Plato wrote that for the leaf to be a 4>appaKov it should be used substantially, he explicitly declares his preference not to liquify it. Of course, Derrida can always mix

the Platonic substance with some kind of liquid, but this will be in contrast to the Platonic text. It will be a different (pappanov, a

5 Charmides 155e3-8.

THE RHETORIC OF DERRIDA 381

Derridian and not a Platonic one, inasmuch as the consolidation of

the deconstructive fluidity creates a profoundly different text from that of Derrida. Such a difference is impossible according to the

rules of the game which demand the complete dissemination of all

the elements in order to achieve penetration.

The substantiality of the leaf, however, suggests another con

trast between Plato and Derrida. In his argumentation, the latter

defines the cp?ppanov as follows:

The "essence" of the fyappaKov is that having neither stable essence, nor individuality "proper" [ni de caract?re "propre'], it is not, in any sense of that word (metaphysical, physical, chemical, alchemical) a substance. The <j>?ppaKov has no ideal identity; it is aneidetic. . . .

This "medicine" is not a simple phenomenon [un simple]. But neither is it a composite phenomenon, a sensible or empirical avvderov par taking of several simple essences. It is rather the prior medium in

which differentiation in general [diff?renciation en general] is produced; this medium is analogous to the one that will ... be reserved for transcendental imagination. (144; 125-6)6

The Derridian rp?ppanov is thus defined as a nonsubstantial phe nomenon. It is deprived of any connection with "is-ness," with being as an appearance within the real world, and it has the attributes of

an origin. It is the "prior medium." Again, the contrast between

the (p?ppaKov of Charmides and the deconstructive rp?ppanov is

prominent. Whether metaphysical or not, the latter is completely

beyond the limits of reality. It is not a substance, and therefore it

is not a leaf. It is different from and alien to the Platonic fy?ppaKov. It is defined as le pharmakon en g?n?ral, the unity of multiple pos

sibilities. The moment of its realization in the physical world?

the moment it becomes a drug, a poison, a medicine, or a paint?is the moment this unity breaks. The realm of en general is left, the

border between what is an "is" and what is not an "is" is crossed,

and this is the moment of departure from that which is signified by "deconstruction."

In this respect, the (p?ppanov in general is very similar to the

dictionary. It is medicine, poison, and paint, while it is not any of

them as far as realization is concerned. As with words, which have

6 In Johnson's translation it is not clear whether "that word" refers

to "is" or to "substance." The order of the words in the French original, however, gives the impression that the first of the two options is the cor rect one.

382 YOAV RI?ON

to leave the randomness of the dictionary and choose a context in

order to start the activation of communication and meaning, so the

(p?ppaKov in general has to give up its potential variability in order

to become a poison or a medicine, to be absorbed or drunk, to be, and not "in general." In this respect, the (pappanov in general is

also very similar to "writing in general," which is the center of our

next step in the analysis.

At the end of his Platonic discussion, Derrida characterizes

"writing in general" in the following way:

That writing [is] eireKeiva ri]s ovalas. . . . Nonpresence is presence.

Diff?rance, the disappearance of any originary presence, is at once [? la fois] the condition of possibility and the condition of impossibility of truth. At once. "At once" means that the being-present [ v] in its truth ... is doubled as soon as it appears, as soon as it presents itself. It appears, in its essence, as the possibility of its most proper

non-truth, of its pseudo-truth reflected in the icon, the phantasm, or the simulacrum. What is is not what it is . . . unless it adds to itself the possibility of being repeated as such. And its identity is hollowed out by that addition, withdraws itself in the supplement that presents it. (194; 168)

It is quite clear from the above that the emphasis of the deconstruc

tive definition of "writing in general" lies in the meaning of the "at

once" (? la fois). The former expression, however, is still obscure

despite Derrida's efforts to illuminate it, and thus an examination

of the context within which this phrase occurs is required. "At

once" is connected with presence, with appearance, with essence,

with addition, and with the possibility of repetition. As for "writing in general," it is somehow connected with a place beyond being

(eireiceiva rr)s ovalas). Now, the possibility of repetition can "take

its place" beyond being, and so can the possibility of presence, the

possibility of appearance and essence. What can not be "there" is

presence itself, the realization of repetition within a certain context.

The moment of doubling is the moment of leaving the place which is beyond being. Onticity, being a part of what is v, means the

abandonment of the ? la fois, of the both-and which is possible only in the place beyond being. Generality as such, whether pharma

ceutic or written, is imprisoned behind the invisible walls of the

"beyond being," and therefore is barred from actively participat

ing in all that is real, such as presence, appearance, cpappaKov and

writing.

Deconstruction is based upon the deferral of choosing; this is

its great strength, but also its great weakness. One can always,

THE RHETORIC OF DERRIDA 383

like Bartleby, "prefer not"; but one can not prefer not for text which

is, by definition, preference and selection. Derrida, in characterizing

his notions as multiple choice phenomena, makes them both potent

(since they lead him to forceful and highly compelling readings) and

impotent (since they are practically impossible outside the context

of the dictionary) at once. The moment of doubling, unlike the

potentiality of being doubled, occurs in the world, which is defined

by Derrida himself as the neutralizer of all the force of his concepts.

Thus, reading as an activity, as an interpretation of an object that

exists, is excluded from the Derridian realms.

There is but one step left before leaving "Plato's pharmacy,"

and that is the unraveling of its rhetoric. The connection between

the vehement criticisms of Derrida's opponents and the power of

his argumentation can not be overemphasized. This, however, is

merely an indication and not an explanation. Wrath is always a

reliable sign of emotional intensity, but almost never a trustworthy

guide to the reasons for its existence. To achieve the latter, we

shall make a slight digression, which will illuminate the roots of

the deconstructive strategy.

In Philosophical Investigations, Wittgenstein develops his anal

ogy between language and games in the following way:

Instead of producing something common to all that we call language, I am saying that these phenomena have no one thing in common which

makes us use the same word for all, -but that they are related to one another in many different ways. And it is because of this relationship, or these relationships, that we call them all "language." I will try to explain this.

Consider for example the proceedings that we call "games." I mean

board-games, card-games, ball-games, Olympic games, and so on.

What is common to them all? -Don't say: "there must be something common . . . but look and see whether there is anything common at all. -For if you look at them you will not see something that is common to all, but similarities, relationships, and a whole series of them at that. . . . Look for example at board-games, with their mul

tifarious relationships. Now pass to card-games; here you find many correspondences with the first group, but many common features drop out, and others appear.. . . And the result of this examination is: we see a complicated network of similarities overlapping and criss-cross

ing; sometimes overall similarities, sometimes similarities of detail. I can think of no better expression to characterize these similarities than "family resemblances." . . . And I shall say: "games" form a

family.7

7 Ludwig Wittgenstein, Philosophical Investigations, trans. G. E. M.

Anscombe (Oxford: Basil Blackwell, 1958), no. 65-67.

384 YOAV RI?ON

The main reason for the comparison between language and

games is the impossibility of finding an overall characteristic for

each. There is no one single aspect shared by all the manifestations

of language and games. What they do share is associative connec

tions; the absence of identity is filled by the presence of resemblance.

Thus, in the category "games," one would not find an amalgamatic

phenomenon, but rather a group of diverse elements. The compar

ison between Wittgenstein and Derrida should not surprise the

reader, who can easily recognize many points of similarity (asso

ciations, links, diversity, and, obviously, games). The illumination

of the rhetoric, however, demands concentration on the difference

between the two.

Deconstruction reveals the connection between the different

links of the textual net. In itself, this movement is quite legiti

mate; its usage, however, is open to criticism. For Derrida, the

unveiling of hidden interconnections is a means of treating them

alternatively. In other words, if differ anee is related to (fiappaicov,

it is possible to apply either one or the other to a given context,

according to the requirements of deconstruction. Thus, "writing in general" can be an apposition to the diff?rance (194; 168), and

writing and the cpappaKov can be regarded as identical options

(118; 101-2). The Derridian links, however, like those of Witt

genstein, cannot be completely alternative; they can only serve as

partial supplements. Through most of the deconstructive itiner

ary one may easily ignore the incompleteness (for example, that

despite its being a cpappaKov, writing is not a leaf like the 4>appaKov

of the Charmides) of most of the Derridian supplements. At a

certain stage, however, such neglect is no longer possible, and it

is here that the deconstructive strategy both veils this impossi

bility and brings into light its most compelling outcomes. Hence,

the success of deconstruction lies in the possibility of taking al most invisible steps, so that in the end a prominent contrast may

be displayed. It is thus easier to concentrate on the forceful iden

tifications of truth and falsehood, of nonpresence and presence,

of origin and repetition, and, of course, of the priority of writing over speech, than to perceive the movements upon which these

conclusions are based.

A concrete example will clarify the above complexity. Let us

imagine the (p?ppaKov and the writing to be two links in our game.

According to Wittgenstein, they cannot be congruent, but only

THE RHETORIC OF DERRIDA 385

partially overlapping. This is reasonable if one remembers that

writing is not identical with the cpappaKov, yet has a lot in common

with it. Graphically, it would have the following shape:

writing (papiiaKov

The center of the diagram, which signifies the common aspects of

both writing and (p?ppanov, is also the center of Derrida's discussion, that is, the metonymical and metaphorical connections between

writing and the (pappaKov as two kinds of drugs. At the beginning of the deconstructive reading the margins can be identified with

marginality, since the fact that the congruity of the elements is

metaphorical and not concrete can be neglected. Then, to achieve

the decomposition of the logos, another move is required, from writ

ing to speech. Again, similarity (mimesis) is emphasized, while

difference (distance from the origin) is expelled to the margins. In

a diagram it would look as follows:

speech (p?ppaKov

writing

It is now, when identity between (p?ppaKov, writing, and speech seems

to be achieved, that Derrida forcefully deconstructs the logos by means of difference. Suddenly, writing as a (p?ppanov is not close

to speech but something alien to it, trying to destroy it in one stroke

(en coup); it is no longer a friend but an enemy, and, what is worse, an internal one. Thus, both inside (because of the similarity) and

outside (because of the difference), it cannot be either expelled or

386 YOAV RI?ON

accepted without causing great damage. Deconstruction has once

again won the battle.

I prefer not to focus here on the meaning of getting closer in

order to create the greatest possible distance, despite this being a

recurring, if not an essential, move in Derrida's strategy of reading.

At this stage of the analysis, the flaws in the rhetoric are my main

concern. The diagrams precisely exemplify that the move from

writing as (p?ppanov to speech is based upon the concealment of the

insurmountable gap between writing and the (p?ppaKov. The chasm

between the concrete and the metaphorical stops being marginal

when Derrida tries to create an identity between the above two

elements in order to deconstruct the logos. Writing is not (p?ppanov,

it is merely a (frappaicov, something similar but not identical to the

(fr?ppaKov. Again, one may discern that the resemblance between

the links creates both the possibility of an easy transition from

(p?ppaKov to speech and the impossibility of a complete liquidation of the latter by the former at once.

In conclusion, I have tried to show that deconstruction cannot

fulfill its own precepts without generating an internal contradiction.

The text is required to be both an ahierarchical and a hierarchical

phenomenon. The rhetoric of the strategy which successfully blurs

this contradiction is exemplified by the definition given by Wittgen stein to his language game. As for the deep and complex relation

ships between writing and speech hinted at in the last section, these

are dealt with in the next essay on the Phaedrus.8

The Hebrew University

8 This article has benefited immensely from the criticism of and dis

cussion with Shlomith Rimmon-Kenan, Elizabeth Freund, and Shuli Bar zilai. I am also indebted to Menachem Brinker for his comments on earlier versions of this essay, and to Andrew Lang for his careful editing of both

essays.