Page 165 - Encyclopedia of Philosophy of Language

on the literary (narrative and metaphoric) qualities of all texts.
Despite all these anticipations of later literary struc- turalism and semiotics, the structuralist work of Jakobson and Levi-Strauss remained an intrinsic structuralism, which did not theorize the reading or writing subject, and which did not theorize the relations of texts to their contexts, nor move beyond the denotative level of linguistic analysis. The function of connotative semiotics in making sense of myth or poetry as explored later by Barthes in S/Z, for exam- ple, was not even considered. Nor was what Barthes would later call the myth of the denotative level itself ever raised in this context: 'denotation is not the first meaning, but pretends to be s o . . . the superior myth by which the text pretends to return to the nature of language, to language as nature' (1974:9). The reader was assumed to know what Levi-Strauss and Jakob- son knew: structuralist facts were transparent and the same for everyone. Levi-Strauss could confidently say: 'We shall use the Oedipus myth, which is well known to everyone' (1963:213). And his readings of the bun- dles of relations he finds in the system of that myth, presented as if they were intrinsic to the text and the system, are of course connotative readings, structured by binarisms, and produced in relation to cultural and hermeneutic codes that only a reader constructed and positioned as a highly trained and educated structural anthropologist would produce.
3.5 Metaphor and Metonymy
Jakobson linked the principles of similarity and par- allelism that constituted the code or the paradigm, and metaphor, which always involves saying one thing in terms of another. These similarity relations were particularly in evidence in the case of the met- alinguistic function and its focus on the code through paraphrase, and in the poetic function with its focus on the message. On the other hand, he connected the constitutive role of difference in the syntagm, its patterns of contiguity which always involved relations of parts to wholes in the context of the linearity of the signifier, to metonymy.
As so often with Jakobson, the insights come from phonology. Speaking of the exchange between the cat and Alice in Lewis Carroll's Alice in Wonderland: '"Did you say pig or fig!" said the Cat. "I said pig," replied Alice'—he points out that in her response Alice makes a choice between 'the distinctive feature stop versus continuant,' and combines this choice with cer- tain other features combined into a bundle of dis- tinctive features, called the phoneme, in this case the phoneme /p/. This is then followed by the phonemes l\l and /g/, themselves bundles of distinctive features. 'Hence the concurrence of simultaneous entities and the concatenation of successive entities are the two ways in which we speakers combine linguistic con- stituents' (Jakobson 1957b: 242). Here in fact is what
he would define as the poetic function established as a constitutive feature of all language.
He speaks (Jakobson 1957b:243) in the same con- text of the 'ascending scale of freedom' with which the speaker operates as he moves from the phonemic level, where choices are fully established by the code, to the levels of word, sentence, or utterance where 'the freedom of any individual speaker to create novel utterances increases substantially.' Every sign is made up however of constituent signs and occurs in com- bination with other signs. This means that every linguistic unit is simultaneously a context for simpler units and has its own context in a more complex unit. 'Combination and contexture are two faces of the same operation.' This leads him to a critique of Saus- sure's concepts of the operation of syntagm and para- digm, as mutually exclusive, and of the linearity of the sign. He argues that Saussure recognized only the temporal sequence, and not the two varieties of con- catenation and concurrence in the linguistic signifier: he 'succumbed to the traditional belief in the linear character of language "qui exclut lapossibilite depro- noncer deux elements a lafois".' According to Jakob- son, the constituents of a context are in a state of contiguity (metonymy), while the elements of a sub- stitution set are linked by various degrees of similarity (metaphor). It is these two operations which provide every linguistic sign with two interpretants (he quotes Peirce's use of this term), one which links the sign internally to the code and the other which links it externally to the context. Every sign has a meaning that derives from the code, and meanings derived from its contextual relations with other signs in a sequence. It is the first that for Jakobson ensures the trans- mission of the message between addresser and addressee.
The context he was talking about was internal to the linguistic system. Halliday has used the same insights to explore the internal semiotics of texts, the metonymic and metaphoric relations between clause and text grammars (1980), and his concept of gram- matical metaphor also works to explore metaphoric and metonymic relations internal to the linguistic sys- tem (1985: ch. 10). His textual function of language however and his theory of the relations between texts and the semiotic constructs that are their contexts (a realizational relation) begins to provide a linguistic account of similar relations that are external to the system. Jakobson's important insights about the simultaneity of metaphor and metonymy to every linguistic sign could easily have been extended to describe the metonymic and metaphoric relations with contexts external to the text that are the stuff of Hjelmslev's connotative semiotics, and at issue in Levi-Strauss's reading of the Oedipus myth. Never- theless, once it was recognized that the paradigmatic was always constitutive of the syntagmatic, of textual- ity, as discussed above, the recognition of the essential
Literary Structuralism and Semiotics
143