This is how institutional s-codes can act as correlational codes.
So far, in our examination of the internal logic of both correlational and institutional codes, we have ascertained the difficulty of clearly separating equivalences from inferences. It seems as if there were neither pure correlational codes nor pure institutional codes — except the rare and blatant cases of transciption tables, such as the Morse code and the most elementary ciphers equating expression with expression. In every other case, we have met an inextricable web of pseudocorrelations that involve instructions and inferences on one hand and, on the other hand, sets of instructions that can generate relationships of signification and processes of communication. We can say that, every time the word code has been pronounced in the semiotic milieu to indicate both s-codes and semiotic codes, it was this ‘strong’ sense of the term which was really in play.
To find perhaps more reason to justify this attitude, we should now consider a case where the term code seems to have been used in a naive metaphorical way, as referred to a process or to a series of processes that by no means can be called ‘communicative’, since they represent cases of sheer physical stimulation.
5.7. The genetic code
Signs always request an interpretation. Stimuli, on the contrary, produce or elicit a blind reaction (see Eco 1976, 0.7.1). However, the expression genetic code has been largely used in the biological milieu, since the dis-coveries of Crick and Watson (in the 1950s) and at the time when Jacob and Monod discovered the transcriptional process from DNA to RNA, in 1961.
Obviously, we should first distinguish between what happens in the cell and the metalanguage used by a scientist in order to describe it. The equivalences reproduced below, insofar as they are used by a scientist in order to know or to say what happens when RNA reproduces the mes-sage conveyed by DNA, sound like ciphers. In this sense, we could speak of a genetic code only apropos of the metagenetic device used by geneticists when they speak to each other about genetic phenomena. But once again one is compelled to ask why that ‘metaphor’ and not another one has been so successfully used.
As is known, proteins are defined by a sequence of amino acid residues in the polypeptide, and this sequence is determined by a sequence of nucleotides in a fragment of DNA. Since the fragment of DNA is in the cell while the protein synthesis takes place in the ribosome, then the information conveyed by DNA must be brought to the ribosome, and this transfer requires a series of ‘translations’. Thus the message of DNA is translated in terms of messenger-RNA, this one in terms of transfer-RNA, where special enzymes catalyze the covalent association of the amino acid with the RNA molecule. ‘Translation’ and ‘transcription’ are metaphors; as a matter of fact, the elements in play are coupled together because of a stereochemical complementarity, for the same reasons (so to speak) for which a given key fits a given keyhole.
In any case, in order to show which keys fits which keyhole, one can express this phenomenon by a sort of cipher. The nucleotides of DNA being adenine, guanine, cytosine, and thymine (A, G, С, Т), whereas in the messenger RNA thymine is substituted by uracil (U), the code ruling the transcription from DNA to messenger-RNA can be expressed as fol-lows:
A- U T A G С С G
In order to determine an amino acid, a triplet of nucleotides is needed; four nucleotides being in the position of producing 64 three-letter ‘words’, and the amino acid residues to be specified being only 20, then many different triplets can specify the same amino acid residue (thus realizing a sort of synonymy). Certain triplets, which do not desig-nate any amino acid, play the role of punctuation signals to establish the beginning and the end of a given sequence of nucleotides.
We have now a second code, where the expression (for the sake of simplicity) is written in the ‘language’ of the messenger-RNA and where the corresponding ‘content’ (or result) is the amino acid residues:
GCU GCC GCA GCG
GCU CGC CGA CGG AGA AGG
AUU AAC
Alanyl Arginyl
Asparagyl
(and so on)
As it is written, this code looks like a system of equivalences (though between a unique content and synonymous expressions). But since the transcription takes place through a process of steric stimuli, one could describe the process as an instructional one. The protagonists of the whole process ‘know’ (by a sort of blind material wisdom) that, if a given series of stimuli is provided, then a given insertion must be performed.
Prodi (1977) maintains that such a basic phenomenon represents an elementary, but by no means metaphorical, example of interpretation in Peirce’s sense. Every element in the process interprets a previous one and, in doing so, makes the process grow. A case of semiosis, even though not unlimited.
Thus the genetic code (but this time we can speak of the one of the organism, not only of the one of geneticists) seems to be an s-code made up with minor s-codes, in which every element is definable in terms of its (steric) position and opposition to other elements, but also a code in the strong sense of the term, both correlational and institutional, where not only x correponds to y, but where also if x then у must be realized. More similar to a mathematical system than to a judicial deontic code, ruled by necessity, susceptible obviously to errors (mutations, cancer), but not optional.
The fruitfulness of the genetic metaphor is not due to the fact that it can say whether the genetic processes are semiotic or not. What the metaphor reveals is that, even at the elementary level of these biological phenomena, there is no sensible difference (a) between s-codes and codes and (b) between correlation and instruction — that is, there is no sensible difference between equivalence and inference, each equiva-lence being a quasi-automatic inference.
Maybe it is too hard to assume (as Prodi suggests; see 1983) that the bio-logic represents the model, the source, and the materialistic founda-tion of the ‘cultural’ logic, and therefore of every semiotics. It is certain, however, that when studying both bio-logic and conceptual logic we are in trouble when we try to distinguish correlation from instruction, s-codes from codes. Or, to put it in more reasonable terms, we can out-line theoretical distinctions, by elaborating different abstract models, but we are obliged to recognize that in the actual semiosis these models are instantiated all together at the same time. Which explains (even though maybe it does not completely legitimate) the ‘generous’ use of code made by so many disciplines in the last decades.
Undoubtedly, the notion of encyclopedia is more flexible and de-scribes better than the one of code the kind of competence needed to express and to interpret texts in a natural (not necessarily verbal) lan-guage. But, from the point of view of their internal structures, codes and encyclopedias are not radically different. Both are complex networks of complex pseudo-equivalences and of more or less cogent and constructive instructions. The notion of encyclopedia may differ from the one of code insofar as it also comprehends, among other instructions, systems of frames and scripts. But, structurally speaking, a code in the strong sense of the term does not exclude this kind of instruction. Institutional codes
do exactly this.
5.8. Toward a provisional conclusion
We are now in a position to disentangle a lot of apparent inconsistencies and contradictions in the current literature on codes. Many of the usages that looked rather metaphorical, self-contradictory even within the same theoretical framework, can now reveal a more profound coherence.
In The Raw and the Cooked (1964) Lévi-Strauss gives the impression of speaking both of s-codes and of correlational codes. For instance, he says that he is interested in “the system of axioms and postulates defining the best possible code, capable of conferring a common significance on unconscious formulations which are the work of minds, societies, and civilizations chosen from among those most remote from each other” and that “as the myths themselves are based on secondary codes (the pri-mary codes being those that provide the substance of language)” his book “is put forward as a tentative draft of a tertiary code, which is intended to ensure the reciprocal translatability of several myths” (1964; Eng. tr., p. 12). Here he clearly means an s-code as a system of trans-formations. He is pursuing the same project later outlined in The Naked Man: the mistake of other mythologists was to try to understand the myths by means of a single and exclusive code, when in fact several codes are always in operation simultaneously.
It is im-possible to reduce the myth to any one code, nor can it be explained as the sum of several codes. It would be truer to say that a group of myths consti-tutes in itself a code, the power of which is superior to each individual code it uses to decipher manifold messages. It is tantamount to an ‘intercode’ — if I may be pardoned the neologism — which makes possible the reciprocal conversion of messages in accordance with rules, the range of which remains immanent in the different systems which, through its operation, allow the emergence of an overall significance distinct
