PARTISAN REVIEW
539
"on a scientific basis," a system for accurately measuring tensions be–
tween anticipation and surprise in poetry. ) Jakobson emphasizes that
parallelism should be " promoted to canon" - which seems to mean
that the correspondence between sound and meaning should no
longer be considered as doubtful or occasional or heavily ambiguous
events in poetry, but rather as an " inevitable" conjunction whi ch
characterizes and identifies poetry, and which can be expressed in
terms of "laws" of the poetic function of language.
How well does Jakobson's linguistic theory of poetry work when
it is actually applied to poetry? Nicolas Ruwet, in his contribution to
the Johns Hopkins symposium on structuralism, expresses some re–
servations about Jakobson's theory. First of all, Ruwet writes, the
transformational model in linguistics has shown the inadequacy of
the theory of the two axes and the need to postulate the existence of
other types of relations. Furthermore, Ruwet asks how the outlining
of relationships of equivalence can take into account the "infinite
gradation of equivalences in a poem ," as well as " the problem of the
distinction between obligatory linguistic elements and those which are
optional." More fundamentally, Ruwet points out that while lin–
guistics does provide materials for the precise description of poetic
language, " ... it is incapable, working alone, of determining how
pertinent these materials are from an aesthetic point of view." Indeed,
Jakobson 's " promotion to canon" of the princi ple of parallelism can
be defended only if we limit ourselves to poetry characterized by the
most obvious and uninteresting artificiality. I'm thinking of the ver–
sified writings of Edgar Allan Poe, who has been greatly admired by
Jakobson and by several generations of French writers. In "Linguis–
tics and Poetics;' J akobson quotes some famous lines from "The
Raven "; he makes a linguistic analysis of these lines in whi ch he
manages to say everything about them except the major fact that
they are atrocious. An adequate analysis of them would require some
explanation of the unintentionally comic or burlesque effect (and
not, as Jabokson claims, " the overwhelming effect") which, as is
frequently the case in verse, results from excessive sound and rhythm–
ic repetitions.
Jakobson writes disdainfully that " .. . no manifesto, foisting a
criti c's own tastes and opinions on creative literature, may set as sub–
stitute for an objective scholarly analysis of verbal art," and the
French structuralists ha\'e also proclaimed their indifference to value
