That formality is itself the condition of a purely functional analysis. The idea of a linguistic
function and of a purely linguistic unit—the glosseme—excludes then not only the con-
sideration of the substance of expression (material substance) but also that of the substance of
the content (immaterial substance). “Since language is a form and not a substance (Saussure),
the glossemes are by definition independent of substance, immaterial (semantic,
psychological and logical) and material (phonic, graphic, etc.) .” 21 The study of the function-
ing of language, of its play, presupposes that the substance of meaning and, among other
possible substances, that of sound, be placed in parenthesis. The unity of sound and of sense is
indeed here, as I proposed above, the reassuring closing of play. Hjelmslev situates his
concept of the scheme or play of language within Saussure’s heritage—of Saussure’s
formalism and his theory of value. Although he prefers to compare linguistic value to the
“value of exchange in the economic sciences” rather than to the “purely logico-mathematical
value,” he assigns a limit to this analogy.
An economic value is by definition a value with two faces: not only does it play the role of a
constant vis-à-vis the concrete units of money, but it also itself plays the role of a variable vis-
à-vis a fixed quantity of merchandise which serves it as a standard. In linguistics on the other
hand there is nothing that corresponds to a standard. That is why the game of chess and not
economic fact remains for Saussure the most faithful image of a grammar. The scheme of
language is in the last analysis a game and nothing more 22
((58))
In the
Prolegomena to a Theory of Language (1943), setting forth the opposition
expression/content, which he substitutes for the difference
signifier/signified, and in which
each term may be considered from the point of view of form or substance, Hjelmslev
criticizes the idea of a language naturally bound to the substance of phonic expression. It is
by mistake that it has hitherto been supposed “that the substance-expression of a spoken
language should consist of ‘sounds’:”
Thus, as has been pointed out by the Zwimers in particular, the fact has been overlooked that
speech is accompanied by, and that certain components of speech can be replaced by, gesture,
and that in reality, as the Zwimers say, not only the so-called organs of speech (throat, mouth,
and nose), but very nearly all the striate musculature cooperate in the exercise of “natural”
language. Further, it is possible to replace the usual sound-and-gesture substance with any
other that offers itself as appropriate under changed external circumstances. Thus the same
linguistic form may also be manifested in writing, as happens with a phonetic or phonemic
notation and with the so-called phonetic orthographies, as for example the Finnish. Here is a
“graphic” substance which is addressed exclusively to the eye and which need not be
transposed into a phonetic “substance” in order to be grasped or understood. And this graphic
“substance” can, precisely from the point of view of the substance, be of quite various sorts
23
Refusing to presuppose a “derivation” of substances following from the substance of phonic
expression, Hjelmslev places this problem outside the area of structural analysis and of
linguistics.
Moreover it is not always certain what is derived and what not; we must not forget that the
discovery of alphabetic writing is hidden in prehistory [n.: Bertrand Russell quite rightly calls
attention to the fact that we have no means of deciding whether writing or speech is the older
form of human expression
(An Outline of Philosophy [London, 1927], p. 47)], so that the
assertion that it rests on a phonetic analysis is only one of the possible diachronic hypotheses;
it may also be rested on a formal analysis of linguistic structure. But in any case, as is
recognized by modern linguistics, diachronic considerations are irrelevant for synchronic
descriptions (pp. 104-05).
H. J. Uldall provides a remarkable formulation of the fact that glossematic criticism operates
at the same time thanks to Saussure and against him; that, as I suggested above, the proper
space of a grammatology is at the same time opened and closed by The Course in General
Linguistics. To show that Saussure did not develop “all the theoretical consequences of his
discovery,” he writes:
It is even more curious when we consider that the practical consequences have been widely
drawn, indeed had been drawn thousands of years before Saussure, for it is only through the
concept of a difference between form and substance
((59))
that we can explain the possibility of speech and writing existing at the same time as
expressions of one and the same language. If either of these two sub-stances, the stream of air
or the stream of ink, were an integral part of the language itself, it would not be possible to go
from one to the other without changing the language. 24
Undoubtedly the Copenhagen School thus frees a field of research: it becomes possible to
direct attention not only to the purity of a form freed from all “natural” bonds to a substance
but also to everything that, in the stratification of language, depends on the substance of
graphic expression. An original and rigorously delimited description of this may thus be
promised. Hjelmslev recognizes that an “analysis of writing without regard to sound has not
yet been undertaken” (p. 105). While regretting also that “the substance of ink has not
received the same attention on the part of linguists that they have so lavishly bestowed on the
substance of air,” H. J. Uldall delimits these problems and emphasizes the mutual independ-
ence of the substances of expression. He illustrates it particularly by the fact that, in
orthography, no grapheme corresponds to accents of pronunciation (for Rousseau this was the
misery and the menace of writing) and that, reciprocally, in pronunciation, no phoneme
corresponds to the spacing between written words (pp. 13-14).
Recognizing the specificity of writing, glossematics did not merely give itself the means of
describing the graphic element. It showed how to reach the literary element, to what in
literature passes through an irreducibly graphic text, tying the play of form to a determined
substance of expression. If there is something in literature which does not allow itself to be
reduced to the voice, to epos or to poetry, one cannot recapture it except by rigorously
isolating the bond that links the play of form to the substance of graphic expression. (It will by
the same token be seen that “pure litera-ture,” thus respected in its irreducibilty, also risks
limiting the play, restrict-ing it. The desire to restrict play is, moreover, irresistible.) This
interest in literature is effectively manifested in the Copenhagen School. 25 It thus removes
the Rousseauist and Saussurian caution with regard to literary arts. It radicalizes the efforts of
the Russian formalists, specifically of the O.PO.IAZ, who, in their attention to the being-
literary of literature, perhaps favored the phonological instance and the literary models that it
dominates. Notably poetry. That which, within the history of literature and in the structure of a
literary text in general, escapes that framework, merits a type of description whose norms and