410
PARTISAN REVIEW
signifier to signified "arbitrary" proved a useful gesture in linguistic
analysis, but it does not necessarily follow that all we read about the
"arbitrariness of narrative, " for instance, or the lack of relation
between signifiers and signifieds in a poem, has any analytic value at
all; it is based on a moralized extension of Saussurian concepts beyond
their proper field. Too often (as Jonathan Culler has pointed out) the
" linguistic model " has in fact been rather the " linguistic metaphor";
and in the shift-as is so often the case with metaphorical usages-an
analytic procedure has been moralized, become a hidden metaphysics.
I do not mean to argue that structuralism is (or
was,
since as a
"movement" it is past) without an implicit ideology, but I do think it
is less than fruitful to look for a structuralist "doctrine," and an error
to attempt the extrapolation of a structuralist philosophy. Structural–
ism more than anything else has been a kind of activity, a form of
analysis, a mode of attention, and its contribution
to
literary criticism
has not been a body of doctrine, nor even a lexicon, but rather a new set
of gestures, or perhaps more accurately, a very old set of gestures
renewed. And here I find wholly pertinent the pattern, or paradigm,
which Denis Donoghue repeatedly finds in structuralist work: its
characteristic movement from entity to function, its choice to replace a
thing, or a presence, with a set of relations, a place within a system.
Saussure's decision that language is form and not substance, a syste–
matic set of relations in which what counts (because it is what is
analyzable) is not entities in themselves but the differences between
them, stands as the inaugural gesture of structuralism.
It
is of course
preeminently a formalist gesture, based on the explicit belief that
progress in analytic linguistics could be achieved only by formalization
of the field of study, which here (as elsewhere) meant defining the
limits of the field of study, closing its frontiers, abstracting it from
historical process-taking a horizontal slice through it-and
establish~
ing the interrelationship of phenomena in the field thus defined. To
proceed in such a manner is no doubt an inevitable fact of any true
formalism: I think of Valery's remark, "What is form for others is
subject matter for me. " Form becomes the matter to be treated (Valery
maintains that
"Le
Cimetiere marin" was written as a reflection on the
metric form of the
decasyllable),
which should not mean hypostatizing
form into substance, into an entity.
In
detecting the move from entity
1O
function to
be
a "paradigm" of structuralist thought, Denis Dono–
ghue is himself making a move in the best tradition of structuralism,
pointing out the formal pattern characteristic of a kind of thinking.
This way of describing structuralism is to my mind far more pertinent
