Much of Derrida’s analysis of Saussure’s (and others who follow him making writing secondary to a fully present vocal language) treatment of writing as the representation of words depends on the notion of the arbitrary sign:
Writing being defined as a “system of signs,” there is no “symbolic” writing (in the Saussurian sense), no figurative writing; there is no writing as long as graphism keeps a relationship of natural figuration and of some resemblance to what is then not signified but represented, drawn, etc…
…[T]his refers, beyond the nature/culture opposition, to a supervening opposition between physis and nomos, physis and techné, whose ultimate function is perhaps to derive historicity; and paradoxically not to recognize the rights of history, production, institutions etc, except in the form of the arbitrary and in the substance of naturalism.
What exactly is the point of belaboring arbitrariness? And what does that have to do with the distinction between nature and institution? It seems this returns again to the privileging of the voice, to the imputed natural unity between the voice and sense; but that will become clear later. First, it must be made apparent that linguistics does in fact inherit, uncritically, a large amount of baggage from the metaphysical tradition that is quite unscientific and quite untenable.
Linguistics (that follows Saussure) would like to simply disregard writing, but it cannot, because writing has permeated society. Because writing is perfidious, because it inverts the natural hierarchy of speech by not respecting its place as the tool of a full speech and usurps the natural unity of sound and sense, because it is parasitic on speech and confuses the order of things, writing is dangerous. Saussure demonstrates how writing interferes with what some people might call, using the popular science rhetoric currently in vogue, the natural evolution of pronunciation and articulation of speech by confusing people with archaic orthography1. The fact that the tool of writing can alter the thing which it is supposed to serve is seen as an aberration, something bordering on sin, and the discussion of the instance of this is couched in moralistic rhetoric. What comes about is situation where what reflects or represents language does so imperfectly — according to the assumptions of a full speech:
Representation mingles with what it represents, to te point where one speaks as one writes, one thinks as if the represented were nothing more than the shadow or reflection of the representer. A dangerous promiscuity and a nefarious complicity between the reflection and the reflected which lets itself be seduced narcissistically. In this play of representation, the point of origin becomes ungraspable. There are things like reflecting pools, and images, an infinite reference from one to the other, but no longer a source, a spring. There is no longer a simple origin. For what is reflected is split in itself and not only as an addition to itself or its image. The reflection, the image, the double, splits what it doubles. The origin of the speculation becomes difference.
This situation creates a certain anxiety for those who presume the natural hierarchy of speech/writing, for the proper study of language would be a study of the natural unity of speech and sense and how that changes in itself. Effects of writing upon it are effects from the exterior, something like pollution. This of course comes down to a poor definition of the field of language: the metaphysical baggage received without criticism. What does it mean that so many people devote so much time to worrying about the effects of writing? Why do they? The anxiety stems from necessity, follows unavoidably from the conception of speech as language which Saussure and his intellectual brethren espouse, from the unity of the word. The holding of this belief, besides being accountable to the inertia of the tradition, might be seen as depending “expressly, and in contradiction to the other levels of the Saussurian discourse, upon a psychology of consciousness and of intuitive consciousness.” The possibility of nonintuition is not even broached.
These are errors and they delimit the functioning of any linguistics that follows their model.
1. Saussure uses the example of the French last name Lefèvre, but we could find all manner of English or American examples should we look.