I I 6
C O N C E P T S
As the varied examples I have adduced suggest, the notion o f text
seems to serve, above all, to foreground the complexity o f the semiotic
productions that we undertake to study. Preserved in these vicissitudes is
stress on the etymological connotations o f wovenness— multiple strands
that can be pursued and whose relation needs to be considered— whether
these strands are considered to be things like the codes that Barthes de
scribes or the inextricability o f language from the reality that it performa-
tively helps to structure. Whether there are incompatibilities between the
complex concepts o f text in Derrida and de Man is not clear; for both, text
is a structural relationship between ineluctable but incompatible perspec
tives. What Derrida and de Man do certainly share, though, in contrast to
Barthes, is a preference for monadic rather than dualistic models: the text
as impossible object in its integration o f incompatible modes o f function
ing— an impossibility we generally succeed in ignoring.
Both the New Critical notion o f text as an organic totality and
Barthes’ conception o f text as a highly charged impossible ideal have been
eclipsed, I believe, and rightly so, by Derrida and de Man’s aporetic struc
ture, but text can still function as in the early days o f the structuralist enter
prise, as a relatively neutral way o f naming objects o f inquiry whose mean
ing cannot be taken for granted and where, as Geertz puts it, one needs
to focus on the fixation o f meaning from the flow o f events. Text, then, is
both one o f the most complex theoretical constructions in theory and an
incomparable interdisciplinary operator, offering analytic possibilities for
a wide range o f fields.
The Sign: Saussure and Derrida
on Arbitrariness
The arbitrary nature o f the sign lies at the root o f modern theory:
Saussure’s Cours de linguistique générale makes this the first principle o f
part
I
o f the Course, “Principes généraux.” “This principle dominates the
whole o f the linguistics o f la langue-, its consequences are innumerable.” ^
One o f these consequences is broached in the chapter on the immutabil
ity and mutability o f the sign: “the arbitrariness o f its signs theoretically
entails the freedom o f establishing just any relationship [n’importe quel
rapport] between phonic substance and ideas. The result is that each o f
the two elements united in the sign maintains its own life to a degree un
known elsewhere, and that language changes, or rather evolves, under the
influence o f all the forces which can affect either sounds or meanings” (F
no; E 76). The essential nature o f the history o f languages depends on the
arbitrariness o f the sign.
But o f course the most important consequence comes in the chap-
I.
Ferdinand de Saussure, Cours de linguistique générale, ed. Tullio de Mau-
ro (Paris; Payot, 1973), 100 (my translation). Henceforth I will cite page references
to this French edition, as well as to Wade Baskin’s English translation. Course in
General Linguistics (London: Peter Owen, 1974), in the form (F 12; E 13), though I
sometimes silently modify the English translation. Roy Harris’s more recent trans
lation, Course in General Linguistics (London: Duckworth, 1983), is astute but ten
dentious and idiosyncratic. Since his translation gives the page numbers of the
French edition, my references can lead to the correct page in his volume.
ii8
C O N C E P T S
ter on linguistic value, one o f the most difficult o f the Course. If the sign
were not arbitrary, one could not say that “dans la langue il n’y a que des
différences” [in the linguistic system there are only differences], but “ar
bitrary and differential are correlated qualities” (F i66, 163; E 120,118). In
the linguistic system, instead o f ideas given in advance, we have “values
emanating from the system.” These values are entirely relative (a conse
quence o f the arbitrariness o f the sign), and it is therefore the articulation
by language o f the plane o f ideas and o f sound that instantiates the arbi
trariness o f the sign in its most “radical” form.^ It is not just the relation
ship between the signifier love and the signified “love” that is arbitrary but
also the distinction between the signified “to love” and the signified “to
like,” both o f which are rendered in French by aimer. Language is not a
nomenclature, and the articulations o f each plane are themselves arbitrary
and conventional.
Thus for Saussure the arbitrary nature o f the sign determines what
is most distinctive about language both synchronically and diachronically.
Moreover, the Course seems determined to insist on the essential character
o f this arbitrariness. At the beginning o f part i, an objection to the prin
ciple o f arbitrariness is raised and swiftly rejected:
On pourrait s’appuyer sur les onomatopées pour dire que le choix du signifiant
n’est pas toujours arbitraire. Mais elles ne sont jamais des éléments organiques
d’un système linguistique. Leur nombre est d’ailleurs bien moins grand qu’on ne
le croit. Des mots comme fouet ou glas peuvent frapper certaines oreilles par une
sonorité suggestive; mais pour voir qu’elles n’ont pas ce caractère dès l’origine, il
suffit de remonter à leur formes latines {fouet dérive de fagus, “hêtre,” glas de clas-
Do'stlaringiz bilan baham: 
