After these two scientific events, the code wave crested, and the landslide effect of the new category did not allow enough time to make subtle formal distinctions. Thus one can record such expressions as phonological code, linguistic codes, semantic codes, kinship codes, codes of myths, literary and artistic codes, cultural codes, genetic code, discrete vs. analogical coding, coded vs. uncoded communication, be-havioral codes, gestural kinesic, paralinguistic, proxemic, physiognomic, perceptual codes —to give only some prudent examples.
Even though a suspicion of indulgent metaphorization was authorized, it was impossible to avoid the question as to whether this code boom represented a sort of epistemological trend, something similar to the common ‘formative will’ that, in the domain of arts, Panofsky called Kunstwollen. Was the code boom an instance of Wissenschaftwollen, the proof of the active, underlying presence in human affairs of the Hegelian Zeitgeist? Cultures know such kinds of terminological pollution: a given term, extrapolated from a precise disciplinary framework, quickly be-comes a password, a shibboleth, and, not only for cliquish reasons, comes to designate a cultural atmosphere, an era. ‘Barocco’ was the medieval name of a syllogism and became the precise designation of a way of making art and poetry, of thinking, of behaving, of acting politi-cally, of believing in God (as happened also with the term ‘mannerism’, coming from painter’s jargon). In such cases these metaphors have a cognitive power and frequently announce a sudden switch from one sci-entific paradigm to another, a scientific revolution.
5.3. Codes and communication
I have said that, underlying the three canonical definitions of code (paleographic, correlational, and institutional) there is a communicational purpose. What is rather curious in our story is that, from its very begin-ning, this purpose is more concealed or alluded to than asserted. As we shall see in 5.4.2, the communicational purpose is implied, by no means focused, in the mathematical theory and in Jakobsonian phonology, and the same happens with the early proposal of Levi-Strauss. I insist par-ticularly on Lévi-Strauss because the code boom starts, in the French structuralist milieu, with Levi-Strauss’ cultural anthropology, even though Lévi-Strauss elaborated the idea of a code because of his former contacts, through Jakobson, with the information theorists.
Les structures elementaires de la parente was written in 1947, but there the word code only turns up incidentally, never as a technical term (for example, “many contemporary codes” are vaguely mentioned; Lévi-Strauss 1947, 1.1.3). The basic categories of Lévi-Strauss are those of rule, system, and structure. Even when the same author in 1945 pro-posed his comparison between linguistics and anthropology (Levi-Strauss 1945), he spoke of phonological “systems” (not of phonological codes). Code appears as a category only with the analysis of myths in “La geste d’Asdival” (Levi-Strauss 1958-59).
In the concluding chapter of Les structures el’ementaires, an equivalence is posited among rule, communication, and sociality: “Linguists and sociologists not only utilize the same methods, but they apply them-selves to the study of the same object. From this point of view, in effect, ‘exogamy and language have the same fundamental function: communi-cation with others and the integration of the group”‘ (1947, ch. 29). Lévi-Strauss, however, is not saying that both kinship and language rep-resent cases of communication; he suggests that society communicates also at the kinship level because there is a more general code (I am interpreting, since the word is not yet used here) which rules kinship, language, architectonic forms, and other phenomena.
The point is: where there is rule and institution, there is society and a deconstructible mechanism. Culture, art, language, manufactured ob-jects are phenomena of collective interactions governed by the same laws. Cultural life is not a spontaneous spiritual creation but, rather, is rule-governed. These rules represent an object of investigation, since they probably are something deeper and more universal than their transi-tory and superficial instantiations.
As confirmation that the concept of code serves not so much to suggest that everything is language and communication as to establish that every cultural production is rule-governed, there is the first text in which (as far as I know) Lévi-Strauss introduces explicitly the term ‘code’: his essay “Language and the Analysis of Social Laws” (1951), in which he takes up again the thesis of Les structures elementaires, dwelling in particular on the analogies between kinship and language. Conscious of the daring of his hypothesis, he points out that it is not sufficient to limit investigation to one society alone, or even to many, unless one identifies a level at which a transition is possible from one phenomenon to another. Thus the problem is how to devise a ‘universal code’ capable of expressing the properties common to different phenomena: a code, whose use would be legitimated both in the study of an isolated system and in the comparison between systems. The final aim is to find “un-conscious similar structures . . . a truly fundamental expression . . . . formal correspondence” (pp. 155—63).
Therefore, as in Jakobsonian phonology, the code is not so much a mechanism which allows communication as a mechanism which allows transformations between two systems. It is irrelevant whether these are systems of communication or something else; what matters is that they are systems which communicate among one another.
Thus, at its very birth, the idea of code appears wrapped in am-biguity: bound to a pancommunicative hypothesis, it is not a guarantee of communicability but, rather, of structural coherence and of access be-tween different systems. An ambiguity rooted in the twofold meaning of communication: communication as a transfer of information between two poles, and as accessibility or passage between spaces. The two concepts imply one another. Their confusion can be fruitful: maybe there are common rules for two distinct operations and these rules are not ineffa-ble but can be expressed (maybe) by an algorithm. In other words, they are coded.
Most of the resistance against the notion of code was due to this fear of hyperrationalization as if these code-oriented theories wanted to put human minds into a computer. On the other hand, the popularity of the new category had all the characteristics of an exorcism: it consti-tuted an attempt to force order upon movement, structure upon events, organization upon earth tremors. Speaking of codes meant for many to identify ‘scripts’ where, previously, only random, blind impulses, un-speakable creativity, dialectic contradictions were recognized. It was perhaps a short ‘rationalistic’ season; as soon as it was possible, poststructuralism replaced codes with drives, dèsirs, pulsions, drifts.
However, we are not interested in this new instantiation of the eternal struggle between Apollo and Dionysus. Let us, rather, follow the tech-nical history of our concept: the present problem is that the early avatars of the notion of code were closer to the semantic field ‘rule’ than to the semantic field ‘communication and/or signification’.
5.4. Codes as s-codes
5.4.1. Codes and information
In the texts of the theoreticians of information, there is a sharp distinc-tion between information, as the statistical measure of the equiprob-ability of events at the source, and meaning. Shannon (1948) distin-guishes the meaning of a message, irrelevant to an information theory, from the measure of information that one can receive when a given mes-sage (which can also be a single electrical signal) is selected among a set of equiprobable messages.
Prima facie the problem of information theorists seems to be how to encode a message according to a rule of this type:
A 00 В 01 С 10 D 11
On the contrary, the true concern of the information theorist is not the correlation between signals (as if they were expressions of something) and their correlated content. The specific concern of the theory is the most economic way of sending a message so that it does not produce ambiguity. For instance, the problem can be solved by inventing a code that allows for more redundant messages, for example:
A 0001 В 1000 С 0110 D 1001
It must be clear that the real problem of the theory is the internal syntax of the system of 1’s and 0’s, not the fact that the strings generated by this syntax can be associated to another sequence (for instance of alphabetic letters) so to correlate them (as expressions) to a ‘meaning’.
Thus the code of which information theorists speak is a monoplanar system, a noncorrelational device, and as such I defined it (Eco 1976) as an s-code.
5.4.2. Phonological code
Phonological code is also an s-code. The distinctive features which make uP phonemes are elements of a mere system of mutual positions and oppositions, pure paradigm. They make up a structure, in the sense defined by Lalande as “an ensemble of elements, be they material or not, reciprocally dependent on each other, that form an organized sys-tem” (1926, “Structure”). A phoneme is distinguished from another by the presence or absence of one or more among the features that form the phonological system. A phonological system is governed by a structural rule, but this rule does not correlate anything to anything else.
It was Saussure who spoke of “code de la langue,” but it has undoubt-edly been Jakobson who extrapolated from information theory the notion of code, and the like, and extended them to linguistics and semiotics at large. At first glance, Jakobson seems to be responsible for a confusing generalization by which the term code indicates both a syntactic system of purely differential units devoid of any meaning and a correlation of two series of elements systematically arranged term to term or string to string, the items of the first standing for the item of the second. As a matter of fact, just when proposing the acceptance of this notion, Jakob-son (1961) appeared clearly conscious of this difference: there is a code only when
