This closure has emerged, says Derrida, with the latest developments in linguistics, the human sciences, mathematics, and cybernetics, where the written mark or signifier is purely technical, that is, a matter of function rather than meaning. Precisely the liberation of function over meaning indicates that the epoch of what Heidegger calls the metaphysics of presence has come to closure, although this closure does not mean its termination. Just as in the essay “On the Question of Being” (Heidegger 1998, 291-322) Heidegger sees fit to cross out the word “being,” leaving it visible, nevertheless, under the mark, Derrida takes the closure of metaphysics to be its “erasure,” where it does not entirely disappear, but remains inscribed as one side of a difference, and where the mark of deletion is itself a trace of the difference that joins and separates this mark and what it crosses out. Derrida calls this joining and separating of signs différance (Derrida 1974, 23), a device that can only be read and not heard when différance and différence are pronounced in French. The “a” is a written mark that differentiates independently of the voice, the privileged medium of metaphysics. In this sense, différance as the spacing of difference, as archi-writing, would be the gram of grammatology. However, as Derrida remarks: “There cannot be a science of difference itself in its operation, as it is impossible to have a science of the origin of presence itself, that is to say of a certain non-origin” (Derrida 1974, 63). Instead, there is only the marking of the trace of difference, that is, deconstruction.
Because at its functional level all language is a system of differences, says Derrida, all language, even when spoken, is writing, and this truth is suppressed when meaning is taken as an origin, present and complete unto itself. Texts that take meaning or being as their theme are therefore particularly susceptible to deconstruction, as are all other texts insofar as they are conjoined with these. For Derrida, written marks or signifiers do not arrange themselves within natural limits, but form chains of signification that radiate in all directions. As Derrida famously remarks, “there is no outside-text” (Derrida 1974, 158), that is, the text includes the difference between any “inside” or “outside.” A text, then, is not a book, and does not, strictly speaking, have an author. On the contrary, the name of the author is a signifier linked with others, and there is no master signifier (such as the phallus in Lacan) present or even absent in a text. This goes for the term “différance” as well, which can only serve as a supplement for the productive spacing between signs. Therefore, Derrida insists that “différance is literally neither a word nor a concept” (Derrida 1982, 3). Instead, it can only be marked as a wandering play of differences that is both a spacing of signifiers in relation to one another and a deferral of meaning or presence when they are read.
How, then, can différance be characterized? Derrida refuses to answer questions as to “who” or “what” differs, because to do so would suggest there is a proper name for difference instead of endless supplements, of which “différance” is but one. Structurally, this supplemental displacement functions just as, for Heidegger, all names for being reduce being to the presence of beings, thus ignoring the “ontological difference” between them. However, Derrida takes the ontological difference as one difference among others, as a product of what the idiom “différance” supplements. As he remarks: “différance, in a certain and very strange way, (is) ‘older’ than the ontological difference or than the truth of Being” (Derrida 1982, 22). Deconstruction, then, traces the repetitions of the supplement. It is not so much a theory about texts as a practice of reading and transforming texts, where tracing the movements of différance produces other texts interwoven with the first. While there is a certain arbitrariness in the play of differences that result, it is not the arbitrariness of a reader getting the text to mean whatever he or she wants. It is a question of function rather than meaning, if meaning is understood as a terminal presence, and the signifying connections traced in deconstruction are first offered by the text itself. A deconstructive reading, then, does not assert or impose meaning, but marks out places where the function of the text works against its apparent meaning, or against the history of its interpretation.