Notice that semiotics is not strictly obliged to answer positively to all the questions raised above; it can also decide, for instance (as many semioticians did), that the way in which a cloud signifies rain is different from the way in which a French sentence signifies or is equivalent to —an allegedly corresponding English sentence. Semiotics is char-acterized by its interest in these problems, not by a prerecorded set of answers.
To be interested in these problems requires a philosophical curiosity; according to Aristotle, it is by an act of wonder that men began, and begin, to philosophize; and, according to Peirce, all new discoveries start when «we find some very curious circumstances which will be explained by the supposition that it was the case of a general rule and thereupon adopt that supposition» (C. P. 2.624). The concept of sign —or every other concept a general semiotics decides to posit as its own theoretical object—is nothing but the result of a supposition of this sort. Signs are not empirical objects. Empirical objects become signs (or they are looked at as signs) only from the point of view of a philosophical decision.
0.7.
When semiotics posits such concepts as ‘sign’, it does not act like a science; it acts like philosophy when it posits such abstractions as sub-ject, good and evil, truth or revolution. Now, a philosophy is not a sci-ence, because its assertions cannot be empirically tested, and this im-possibility is due to the fact that philosophical concepts are not ’emic’ definitions of previously recognizable ‘etic’ data that display even mini-mal resemblance in shape or function. Philosophical entities exist only insofar as they have been philosophically posited. Outside their philo-sophical framework, the empirical data that a philosophy organizes lose every possible unity and cohesion.
To walk, to make love, to sleep, to refrain from doing something, to give food to someone else, to eat roast beef on Friday—each is either a physical event or the absence of a physical event, or a relation between two or more physical events. However, each becomes an instance of good, bad, or neutral behavior within a given philosophical framework. Outside such a framework, to eat roast beef is radically different from making love, and making love is always the same sort of activity inde-pendent of the legal status of the partners. From a given philosophical point of view, both to eat roast beef on Friday and to make love to x can become instances of ‘sin’, whereas both to give food to someone and to make love to у can become instances of virtuous action.
Good or bad are theoretical stipulations according to which, by a philosophical decision, many scattered instances of the most different facts or acts become the same thing. It is interesting to remark that also the notions of ‘object’, ‘phenomenon’, or ‘natural kind’, as used by the natu-ral sciences, share the same philosophical nature. This is certainly not the case of specific semiotics or of a human science such as cultural anthropology. Anthropologists elaborate the notion of brother-in-law to define emically a series of etic occurrences, where different persons play the same social function and they would play this function etically even though no science had previously defined their emic role.
A brother-in-law exists independently as a male human being who, like other male human beings, has a sister who has married another male human being; like other male human beings in the same position, a brother-in-law performs (during certain ceremonies) certain ritual acts, allegedly because of his relationship with a given woman and a given man. Anthropologists can fail in detecting the true reason he performs these ritual acts or in selecting certain features of his behavior as rele-vant, disregarding other phenomena (or can overdo in asserting that the opposition brother-in-law/sister-in-law is analogous to the phonetic op-position voiced/unvoiced . . .). But the anthropologists start from the unquestionable fact that there are nuclei of three persons each, forming both a couple of siblings of the same parents and a couple of persons of different sex living and having sex together.
In philosophy things go differently. What is ‘true’ for Hegel is radi-cally different from what is ‘true’ for Tarski, and, when the Schoolmen said that truth is the adaequatio rei et intellectus, they did not describe entities that were recognizable as such before that definition. The defi-nition decides what a thing is, what understanding is, and what adaequatio is.
This does not mean that a philosophy cannot explain phenomena. It has a great explanatory power, since it provides a way to consider as a whole many otherwise disconnected data —so that, when a scientific approach starts with defining an observable datum and a correct (or true) observation, it starts by positing philosophical categories. A philosophy cannot, however, be true in the sense in which a scientific description (even though depending on previous philosophical assumptions) is said to be true. A philosophy is true insofar as it satisfies a need to provide a coherent form to the world, so as to allow its followers to deal coherently with it.
In this sense, a philosophy has a practical power: it contributes to the changing of the world. This practical power has nothing to do with the engineering power that in the discussion above I attributed to sciences, including specific semiotics. A science can study either an animal species or the logic of road signals, without necessarily determining their trans-formation. There is a certain ‘distance’ between the descriptive stage and the decision, let us say, to improve a species through genetic engineering or to improve a signaling system by reducing or increasing the number of its pertinent elements.
On the contrary, it was the philosophical position of the modern no-tion of thinking subject that led Western culture to think and to behave in terms of subjectivity. It was the position of notions such as class struggle and revolution that led people to behave in terms of class, and not only to make revolutions but also to decide, on the grounds of this philosophical concept, which social turmoils or riots of the past were or were not a revolution. Since a philosophy has this practical power, it cannot have a predictive power. It cannot predict what would happen if the world were as it described it.
Its power is not the direct result of an act of engineering performed on the basis of a more or less neutral descrip-tion of independent data. A philosophy can know what it has produced only apres coup. Marxism as a philosophy displays a reasonable explana-tory power and has had, indeed, a consistent practical power: it contrib-uted to the transformation, in the long run, of many ideas and some states of the world. It failed when, assuming to be a science, it claimed to have a predictive power: it transformed ideas and states of the world in a direction it could not exactly foresee. Applying to globality, a phi-losophy does not play its role as an actor during a recital; it interacts with other philosophies and with other facts, and it cannot know the results of the interaction between itself and other world visions. World visions can conceive of everything, except alternative world visions, if not in order to criticize them and to show their inconsistency. Affected as they are by a constitutive solipsism, philosophies can say everything about the world they design and very little about the world they help to construct.
0.8.
A general semiotics is philosophical in this very sense. It cannot work on concrete evidence, if not as already filtered by other specific semiotics (which depends on a general semiotics to be justified in their procedur-es). A general semiotics studies the whole of the human signifying activity —languages —and languages are what constitutes human beings as such, that is, as semiotic animals. It studies and describes languages through languages. By studying the human signifying activity it influ-ences its course. A general semiotics transforms, for the very fact of its theoretical claim, its own object.
I do not know, as yet, whether a pragmatic theory of speech acts is a chapter of general semiotics or a chapter of a philosophy of language. It should be clear, from the whole of this introduction, that such a question is, to me, devoid of any interest. Undoubtedly, a theory of speech acts starts from the observation (although never innocent) of certain empirical behaviors. In this sense, many of its discoveries could be ranked as items of a specific semiotics. However, I doubt whether a notion such as the one of performative sentence is a neutral one. One says /I promise you/ and bets one’s shirt on this promise; in other cases, one utters the same expression without being aware of the fact that one is ‘doing things with words’. But a theory of speech acts provides us with such an organized knowledge of our linguistic interaction that the future of our linguistic behavior cannot but be profoundly influenced by the sort of awareness it provides. So a theory of speech acts is explanatory, practically powerful, and not fully predictive. It is an instance of philosophy of language, perhaps a chapter of a general semiotics, not a case of specific semiotics.
I am not saying that philosophies, since they are speculative, speak of the nonexistent. When they say ‘subject’ or ‘class struggle’ or ‘dialec-tics’, they always point to something that should have been defined and posited in some way. Philosophies can be judged, at most, on the grounds of the perspicacity with which they decide that something is worthy of becoming the starting point
