But, as is clear from this last quotation, even in the machineries of such an intercode, significations emerge —in the sense in which I have demonstrated that also institutional codes, being in themselves s-codes, Permit the recognition of an immanent strategy of internal signification. In the same work (The Raw and the Cooked, at the beginning of the chapter “Three-Part Inventions”), Lévi-Strauss seems to change his mind, and calls “armature” a combination of properties that remain in-variant in two or several myths (the intercode?), and “code” the pattern of functions ascribed by each myth to these properties. Here the code seems closer to a notion of correlation. On the contrary, the many culi-nary, astronomical, vestimental, geographical codes that Levi-Strauss quotes so frequently in his works, look more similar to content structures, semantic s-codes, systems of values, as well as systems of prescriptions. When, however, he speaks of the “conversion of the culi-nary code into a vestimental one” (in The Naked Man, but with a refer-ence to a more comprehensive analysis in The Raw and the Cooked, Eng. tr., p. 334ff), he seems to speak of a correlation of mutual signification between the elements of two content systems.
Thus Levi-Strauss’ codes look at the same time like syntactic systems, like institutions prescribing norms that can be either obeyed or disre-garded, as bodies of textual functions eliciting forecasts about their pos-sible transformations and as systems of signs, since in the parental code, for instance, the choice of a given partner becomes significant of the obligations to be entertained with his or her relatives. Codes made up with codes, a flexible web of codes, an inextricable texture of internal and external significations, in which it is impossible to distinguish what is semantic from what is syntactic.
Even larger and more complex is the notion of code in the typology of cultures (Lotman and Uspenskij 1971). A code is a way of modeling the world: verbal languages are primary modeling systems, whereas secon-dary modeling systems are all the other cultural structures, from mythol-ogy to art. For these authors there is a clear difference (even though the term code seems to cover it) between equivalences (gemination of two or more chains of elements belonging to different semiotic systems) and pragmatic codes, whereas in their notion of text, as opposed to grammar, there appears a clear idea of the internal signification of instructional devices. Lotman studied abundantly different institutional codes (Lot-man 1969) which are systems of norms or of values, and once again the opposition ‘grammar vs. text’ mirrors the difference between cogent in-stitutions and textual models which suggest or prescribe by means of examples (a given behavior is proposed as an emblem, a sample, a speci-men, so that the instruction might sound like this: if you recognize the charismatic power of this text, then you should act in the same way). In Lotman’s semiotics, correlational and institutional aspects are hardly distinguishable precisely because in this line of thought every social ac-tivity reveals its profound communicative purpose.
Roland Barthes frequently uses the word ‘code’ (and indeed he was one of those responsible for the code boom at the beginning of the 1960s). The fashion code (Barthes 1967) is an s-code, a correlational and an institutional code at the same time. There are systematic links among vestimental units, correlations between types of clothes and social atti-
tudes, between words and clothes, and so on.
In S/Z (1970) Barthes lists five so-called codes: semic, cultural, sym-bolic, hermeneutic, and proairetic. It is difficult to avoid the suspicion of full metaphorization, and, as compared with more traditional definitions, Barthes’ codes seem to overlap each other.
Through the semic code the reader detects that the name Sarrasine has a connotation of femininity (because of the final /e/), and in this sense the semic code does not look so different from so-called linguistic codes. It is through the hermeneutic code that the same name, put forth as the title of the story, articulates a question, its possible response, the variety of chance events which can delay the answer; this title is an enigma and the reader is led to ask: “What is Sarrasine? A noun? A name? A thing? A man? A woman?” (Barthes 1970; Eng. tr., p. 19). The symbolic code seems to be an imprecise set of intertextual evocations suggested by the opening sentence (“I was deep in one of those daydreams”) and by the following antitheses (garden vs. salon, life vs. death, outside vs. interi-or). The proairetic code suggests a logic of actions, of possible narrative developments, and elicits forecasts about the further course of events. The cultural or reference code organizes a body of world knowledge re-ferred to by the text.
In this sense, stresses Barthes, a code is not a mere list of equiva-lences:
The code is a perspective of quotations, a mirage of structures; we know only its departures and returns; the units which have resulted from it (those we inventory) are themselves, always, ventures out of the text, the mark, the sign of a virtual digression toward the reminder of a catalogue (The Kid-napping refers to every kidnapping even written): there are so many frag-ments of something that has been already read, seen, done, experienced; the code is the wake of this already. Referring to what has been written, i.e., the Book (of culture, of life, of life as culture), it makes the text into a prospectus of this Book. . . . Each code is . . . one of the voices of which the text is woven. (Ibid.; Eng. tr., pp. 20—21)
Prima fade here Barthes mistakes codes for the infinite process of semiosis, or with what later will be called intertextuality. But, wrong from the point of view of the weak sense of code, Barthes is right from the point of view of the strong one. What he calls here code is the whole of the encyclopedic competence as the storage of that which is already known and already organized by a culture. It is the encyclopedia, and therefore the Rule, but as a Labyrynth. A Rule which controls but which at the same time allows, gives the possibility of inventing beyond itself, by finding new paths, new combinations within the network. If the code is not only a strict gemination of systems, or a correlation, but also a system of inference, its fate is exactly this.
A code is not only a rule which closes but also a rule which opens. It not only says ‘you must’ but says also ‘you may’ or ‘it would also be possible to do that’. If it is a matrix, it is a matrix allowing for infinite occurren-ces, some of them still unpredictable, the source of a game. It is not by chance or by a deliberate unfaithfulness that, in the 1960s, so-called poststructuralist simply started from the model of the code and of the linguistic sign to find out, beyond or beneath the rule and the organiza-tion, the ‘whirl’, the difference, beance. Since its very beginning, the idea of code was not necessarily a guarantee of armistice and peace, of law and order: it also signaled the coming of a new war.
Codes were introduced to put events under the control of structures, but very soon (maybe already with Levi-Strauss; see Eco 1968) the ulti-mate nature of the ultimate codes (or the Ur-Code) was looked for so deeply as to turn over as the concept of an unshaped ‘origin’. At this point the code became unmanageable, because it was suspected that we do not posit it; on the contrary, we — as thinking cultural subjects—are posited by it.
So the acknowledgement of codes (or of the Code) meant that we are not gods and that we are moved by rules. What was at stake, however, was the question as to whether we are not gods because we are deter-mined by rules socially produced in the course of human history or be-cause God is the Rule which stands behind us and our social history. In other words (and the story is a rather old one), the code can be either nomos or phusis either the Law of the Polis or the Epicurean clinamen. For many poststructuralists, if it was phusis and clinamen, it was not therefore structure, but the absence of every structure.
This conclusion was unnecessary and was determined by previous metaphysical assumptions. It is possible to think of an open matrix, of an unlimited rule, without assuming that it cannot be culturally produced. It is possible to think of the encyclopedia as a labyrinth without assum-ing necessarily that we cannot describe it, and explain its modes of birth and development.
Under the metaphoric usages of code there was at least a unifying ob-session, and the eternal dialectics between law and creativity or, in the words of Appollinaire, the constant fight between Tordre et l’aventure.
[6] ISOTOPY
In The Role of the Reader I devoted several pages to the notion of topic, defined as a cooperative device activated by the reader (usually in the form of a question) for the purpose of identifying the isotopy for inter-preting the text (Eco 1979, 0.6.3).1 I wrote: ” . . . topic as question is an abductive schema that helps the reader to decide which semantic prop-erties have to be actualized, whereas isotopies are the actual textual ver-ification of that tentative hypothesis” (ibid., p. 27), by which I meant to say that the topic as such is not expressed by the text, whereas the isotopy is a verifiable semantic property of it. In other words, the topic is a pragmatic device, whereas the isotopy is
