Certainly, the categories posited by a general semiotics can prove their power insofar as they provide a satisfactory working hypothesis to spe-cific semiotics. However, they can also allow one to look at the whole of human activity from a coherent point of view. To see human beings as signifying animals — even outside the practice of verbal language — and to see that their ability to produce and to interpret signs, as well as their ability to draw inferences, is rooted in the same cognitive structures, represent a way to give form to our experience. There are obviously other philosophical approaches, but I think that this one deserves some effort.
[I] SIGNS
1.1. Crisis of a concept
Current handbooks of semiotics provide us with different definitions of the concept of sign which are often complementary rather than con-tradictory. According to Peirce, a sign is «something which stands to somebody for something in some respect or capacity» (C. P. 2.228). This definition is a more articulate version of the classical definition aliquid stat pro aliquo. When dealing with the inner structure of the sign, Saus-sure speaks of a twofold entity (signifier and signified). Hjelmslev’s def-inition, which assumes the sign-function as a mutual correlation between two functives (expression-plane and content-plane), can be taken as a more rigorous development of the Saussurean concept.
However, in the same period at the turn of the century in which semiotics asserted itself as a discipline, a series of theoretical proposi-tions concerning the death, or at least the crisis of the concept, of sign was developed. Throughout the history of Western thought, the idea of a semiotic theory — however differently defined —was always labeled as a doctrine of signs (see Jakobson 1974; Rey 1973; Sebeok 1976; Todorov 1977). The disparity of meanings attributed each time to the notion of sign calls for a rigorous critique (at least in the Kantian sense of the word ‘critique’). We shall see, however, that the notion of sign had been seri-ously questioned in this sense since the very beginning.)
In the last few years, this reasonable critical attitude seems to have generated its own mannerism. Since it is rhetorically effective to begin a course in philosophy by announcing the death of philosophy, as Freud is pronounced dead at the opening of debates on psychoanalysis, many people have deemed useful to start out in semiotics by announcing the
Heath of the sign. This announcement is rarely prefaced by a philosoph-ical analysis of the concept of sign or by its reexamination in terms of historical semantics. The death sentence is therefore pronounced upon an entity which, being without its identity papers, is likely to be resusci-tated under a different name.
1.2. The signs of an obstinacy
Everyday language and the dictionaries which record its usages disregard theoretical discussions and insist on using the notion of sign in the most varied ways. Even too varied. A phenomenon of this kind deserves at-tention.
I.2.1.
First of all, we find a cluster of linguistic usages according to which the sign is a manifest indication from which inferences can be made about something latent. This includes the usage of sign for medical symptoms, criminal evidence, weather forecast, premonitory signs, presages, the signs of the coming of the Antichrist. . . . A sample of urine for analysis was called signum by the ancients, which leads us to think in terms of a synecdochic relationship, as if the sign were a part, an aspect, a periph-eral manifestation of something which does not appear in its entirety. But the relationship appears to be a metonymic one as well, since the dictionaries speak of sign also for any trace or visible imprint left by an imprinter on a surface. Therefore, the sign is also revelatory of a contact, in a way which tells us something about the shape of the imprinter. These signs, besides revealing the nature of the imprinter, may become marks of the imprinted objects— for instance, bruises, scratches, scars (identifying marks). Ruins belong to the same category: they are the signs of ancient grandeur, of human settlement, or of the flourishing trades of the past.
In all these cases, the fact that the sign is produced intentionally or by a human sender is not relevant. Any natural event can be a sign. Morris asserted that «something is a sign only because it is interpreted as a sign of something by some interpreter. Semiotics, then, is not concerned with the study of a particular kind of object, but with ordinary objects insofar (and only insofar) as they participate in semiosis» (1938:20). However, this first category of signs seems to be characterized by the fact that the ‘standing for’ relationship is based on an inferential mecha-nism: if red sky at night, then sailor’s delight. It is the Philonian mecha-nism of implication: p q. The Stoics were thinking about this sign category when they asserted that a sign is «a proposition constituted by a valid and revealing connection to its consequent» (Sextus Empiricus, Adv. Math. 7.245). The same sign category was the object of Hobbes’ and Wolff’s definitions. According to Hobbes, a sign is the evident antecedent of a consequent or the consequent of an antecedent when similar consequences have previously been observed (Leviathan 1.3). For Wolff, a sign is «an entity from which the present or future or past exist-ence of another being is inferred» (Ontology, p. 952).
1.2.2.
Common language, though, points to a second category of signs. The sign is a gesture produced with the intention of communicating, that is, in order to transmit one’s representation or inner state to another being. The existence of a certain rule (a code) enabling both the sender and the addressee to understand the manifestation in the same way must, of course, be presupposed if the transmission is to be successful; in this sense, navy flags, street signs, signboards, trademarks, labels, emblems, coats of arms, and letters are taken to be signs. Dictionaries and culti-vated language must at this point agree and take as signs also words, that is, the elements of verbal language. In all the cases examined here, the relationship between the aliquid and that for which it stands seems to be less adventurous than for the first category. These signs appear to be expressed by a relation of equivalence rather than by one of inference: p q. /Woman/ «femme or donna»; or /woman/ «animal, human, feminine, adult». Furthermore, these signs seem to depend on arbitrary
decisions.
1.2.3.
The clear opposition between the two categories mentioned above is upset by the use of the word sign in relation to those so-called symbols which represent abstract objects and relationships, such as logical, chem-ical, algebraic formulas, and diagrams. They appear as arbitrary as the signs of the second category; yet, through a structural formula or a dia-gram, the operations which I perform on the expression modify the con-tent. If these operations are performed following certain rules, the result provides me with new information about the content. By altering the lines of a topographical chart, I can predict the possible order of the corresponding territory; by inscribing triangles within a circle, I discover new properties of the circle. This happens because in these sign there are one-to-one correspondences between expression and content. There-fore, they are usually arbitrary and yet contain elements of motivation. As a consequence, the signs of the third category, even though emitted by human beings with the intention of communicating, seem to follow the same model as the signs of the first category: p q, even though they are not natural. They are called iconic or analogical.
1.2.4.
In a similar way, any visual procedure reproducing concrete objects, such as the drawing of an animal in order to communicate the corresponding object and concept, is considered to be an iconic sigh. What do drawings and diagrams have in common? The fact that I can perform on them certain operations: if I draw a moustache on my portrait, I know what I will look like if I wear a moustache. What makes them different? The (apparent) fact that the diagram responds to highly codified and precise rules of production, whereas the drawing appears more spon-taneous. Also, the diagram reproduces an abstract object, whereas the drawing reproduces a concrete object. But this is not always true: the unicorns of the British royal coat of arms stand for an abstraction, a fic-titious object; they stand at most for a (an imaginary) class of animals. On the other hand, Goodman (1968) discusses at length the problematic difference between a person’s image and the image of a person. What makes the difference between the two? Is it related to the intensional properties of the content reproduced by the drawing or to the exten-sional use that we decide to make of the drawing? The problem was present already (and not entirely resolved) in Plato’s Cratylus.
1.2.5.
However, common usage also considers as signs those drawings which reproduce something, but in a stylized form, so that recognizing the ob-ject represented is less important that recognizing a content ‘other’ for which the represented object stands. The Cross, the Crescent, the Hammer and Sickle stand for Christianity, Islam, and Communism, re-spectively. These signs are iconic because —like diagrams and drawings — they can be subjected to manipulations of the expressions which affect the content. They are also arbitrary because by now they are in a state of
