characteristic; not because of the theological foundation which ordained its possibility for the
infinite under-standing or logos of God, but because it seemed to suspend the voice.
“Through” this condemnation can be read the most energetic eighteenth-century reaction
organizing the defense of phonologism and of logo-centric metaphysics. What threatens is
indeed writing. It is not an accidental and haphazard threat; it reconciles within a single
historical system the projects of pasigraphy, the discovery of non-European scripts, or at any
rate the massive progress of the techniques of deciphering, and finally the idea of a general
science of language and writing. Against all these pressures, a battle is then declared.
“Hegelianism” will be its finest scar.
The names of authors or of doctrines have here no substantial value. They indicate neither
identities nor causes. It would be frivolous to think that “Descartes,” “Leibniz,” “Rousseau,”
“Hegel,” etc., are names of authors, of the authors of movements or displacements that we
thus designate. The indicative value that I attribute to them is first the name of a problem. If I
provisionally authorize myself to treat this historical structure by fixing my attention on
philosophical or literary texts, it is not for the sake of identifying in them the origin, cause, or
equilibrium of the structure. But as I also do not think that these texts are the simple effects of
structure, in any sense of the word; as I think that all concepts hitherto proposed in order to
think the articulation of a discourse and of an historical totality are caught within the
metaphysical closure that I question here, as we do not know of any other concepts and
cannot produce any others, and indeed shall not produce so long as this closure limits our
discourse; as the primordial and indispensable phase, in fact and in principle, of the
development of this problematic, consists in questioning the internal structure of these texts as
symptoms; as that is the only condition for determining these symptoms themselves in the
totality of their meta-physical appurtenance; I draw my argument from them in order to
isolate Rousseau, and, in Rousseauism, the theory of writing. Besides, this abstraction is
partial and it remains, in my view, provisional. Further on, I shall directly approach the
problem within a “question of method.”
Beyond these broad and preliminary justifications, other urgencies should be invoked. In
Western and notably French thought, the dominant discourse—let us call it “structuralism”—
remains caught, by an entire layer, sometimes the most fecund, of its stratification, within the
metaphysicslogocentrism—which at the same time one claims rather precipitately to have
“gone beyond.” If I have chosen the example of the texts of Claude Lévi-Strauss, as points of
departure and as a springboard for a reading of
((100))
Rousseau, it is for more than one reason; for the theoretical wealth and interest of those texts,
for the animating role that they currently play, but also for the place occupied in them by the
theory of writing and the theme of fidelity to Rousseau. They will, therefore, in this study, be
somewhat more than an exergue.
((101))
1. The Violence of the Letter: From Lévi-Strauss to
Rousseau
Shall I proceed to the teaching of writing? No, I am ashamed to toy with these trifles in a
treatise on education.—Emile
It [writing] seems to favor rather the exploitation than the enlightenment of mankind....
Writing, on this its first appearance in their midst, had allied itself with falsehood.—“A
Writing Lesson,” Tristes Tropiques.*
Metaphysics has constituted an exemplary system of defense against the threat of writing.
What links writing to violence? What must violence be in order for something in it to be
equivalent to the operation of the trace?
And why bring this question into play within the affinity or filiation that binds Lévi-Strauss to
Rousseau? Another difficulty is added to the problem of the justification of this historical
contraction; what is a lineage in the order of discourse and text? If in a rather conventional
way I call by the name of discourse the present, living, conscious representation of a text
within the experience of the person who writes or reads it, and if the text constantly goes
beyond this representation by the entire system of its re-sources and its own laws, then the
question of genealogy exceeds by far the possibilities that are at present given for its
elaboration. We know that the metaphor that would describe the genealogy of a text correctly
is still forbidden. In its syntax and its lexicon, in its spacing, by its punctuation, its lacunae, its
margins, the historical appurtenance of a text is never a straight line. It is neither causality by
contagion, nor the simple accumulation of layers. Nor even the pure juxtaposition of
borrowed pieces. And if a text always gives itself a certain representation of its own roots,
those roots live only by that representation, by never touching the soil, so to speak. Which
undoubtedly destroys their radical essence, but not the necessity of their racinating function.
To say that one always interweaves roots endlessly, bending them to send down roots among
the roots, to pass through the
xxx fotnote start xxx
•
Claude Lévi-Strauss, Tristes Tropiques (Paris, 1955), pp. 344, 345, translated as
Tristes Tropiques by John Russell (New York, 1961), pp. 292, 293. io1
xxx fotnote slutt xxx
((102))
same points again, to redouble old adherences, to circulate among their differences, to coil
around themselves or to be enveloped one in the other, to say that a text is never anything but
a system of roots, is undoubtedly to contradict at once the concept of system and the pattern of
the root. But in order not to be pure appearance, this contradiction takes on the meaning of a
contradiction, and receives its “illogicality,” only through being thought within a finite
configuration—the history of metaphysics—and caught within a root system which does not
end there and which as yet has no name.
The text’s self-consciousness, the circumscribed discourse where genealogical representation
is articulated (what Lévi-Strauss, for example, makes of a certain “eighteenth century,” by
quoting it as the source of his thought), without being confused with genealogy itself, plays,
precisely by virtue of this divergence, an organizing role in the structure of the text. Even if
one did have the right to speak of retrospective illusion, it would not be an accident or a