4.7.1 The Meaning of Terms and the Sense of Texts
Some have concluded that, if meaning is negotiable, then it is no longer of any use in explaining the way we understand one another.
There are two ways to avoid talking of meaning. The first consists in stating (e.g., as in Marconi 1997: 4) that we cannot talk of meaning, because it is an entity whose whereabouts is unknown, whereas we can talk about the various kinds of lexical competence, which are “families of skills.” But in such a case, it seems to me, in order to establish that such competence exists, all we can do is try a behavioral test: that the speakers share the same degree of competence would be proved by the fact that they understand one another, making the same inferences from the same premises, or in referring to something, and making what I have called acts of felicitous reference. Now, in what way does this proof of the existence of shared competence differ from what I understand as proof by interpretation of the public existence of a content (or meaning) that in its turn proves the private existence of cognitive types? Let us remember that for Peirce even certain ongoing behavior can be seen as a dynamic interpretant (the fact that at the command Attention! all soldiers assume a determined position is a possible interpretant of the verbal command). Therefore to speak of meanings as content does not lead to any hypostatization of elusive entities, or no more at least than it does with concepts of competence or lexical abilities.
The second way consists in saying that the understanding of language happens simply by attributing beliefs to our interlocutor that may coincide more or less with our own. But I have the impression that the introduction of belief does not exorcise the ghost of meaning (and of the CT it expresses), at least not in the sense of content I have been using so far. To take an example from Davidson (1984: 279), if a boat sails past rigged as a ketch, and someone beside me says, Look at that fine yawl!, I assume (i) he perceived the rigging of the boat as I did and erred only in using the linguistic term that belongs to a simple slip; (ii) he does not know the content of the word yawl; or (iii) he has committed a perceptual error. But in all these cases I must postulate that he may know various types of boat just as I do and that he associates a term with these types that expresses their NC, otherwise I could not even suppose that he has (i) simply confused the use of the words, (ii) confused the meaning of words, or (iii) erred in associating a given token with an idea of boat that he is conceiving somewhere. Without the assumption that the two interlocutors must in some way share a system, no matter how asystematic, of directories and files, interaction is not possible. I might be moved by a principle of charity so generous as to attribute to the other person directories organized differently from my own, and try to adapt myself to them. If this means comparing “beliefs,” well and good. But then we are dealing with a purely terminological issue. The tree of directories, and that which ought to be regis tered, is postulated as that organization of content, no matter how idiosyncratic, which others call “meaning.”
I maintain that these discussions lack a distinction that many theories of semiotics have been making for a long time, even though I admit that it is difficult to come to an agreement regarding the sense to be assigned to the terms. The notion of meaning is internal to a semiotic system: it has to be admitted that in a given semiotic system there exists a meaning assigned to a term. The notion of sense, on the other hand, is internal to utterances or, rather, to texts. I do not think anybody would refuse to admit that there exists a fairly stable meaning of the word dog (to the point that we can even—an extreme act of semiotic imprudence—assume that it is the synonym of cane, chien, perro, and Hund) and that nevertheless the same word can assume different senses within different utterances (we need only think of metaphorical cases).23
Readers are enjoined not to think in terms of a total parallelism with the difference as posited by Frege between Sinn and Bedeutung. In any case it seems clear to me that the dictionary can assign a meaning to term X, and nonetheless within different utterances the same term can assume different senses (if nothing else, in the most trivial sense of the term, and so the expression This pope is corrupt, pronounced by an anticlerical with reference to Alexander VI, may have a sense that is different from that pronounced with reference to John XXIII by a traditionalist prelate).
Now, it is evident that in order to determine the sense of an utterance, it is necessary to have frequent recourse to the principle of charity. But the same rule does not hold good with regard to the meaning of a term.
To say that understanding one another is the effect of infinite negotiations (and acts of charity in order to be able to understand the beliefs of others, or the format of their competence) regards the understanding of utterances, i.e., of texts.24 But it does not mean we can eliminate the notion of meaning by dissolving an old and venerable semantics in syntax, on the one hand, and in pragmatics, on the other. To say that meaning is negotiated does not mean that the contract springs from nothing. On the contrary, also from a juridical point of view, contracts are possible precisely because contractual rules are already in existence. A sale is a contract: if A sells a house to B, after the contract the house will be defined as the property of B, and it would never have been such had it not been for the sales agreement; but for the contract to be made, it was necessary for A and B to agree on the NC of sale. A and B can even negotiate the content of house (B could say to A that what he is trying to sell him is not a house but a farmhouse, a shanty, a cow shed, a skyscraper, a lake dwelling, or a ruin unfit for human habitation). But even in such a case they would start off from a shared notion of artifact originally intended to shelter living creatures or things, and if they were unable to have a regulated notion that at least allowed them to distinguish what could be defined as a house from what could be defined as a tree, they would not be able even to begin negotiating.25
Defining the meaning of the term sale is not the same as saying in what sense I must interpret the expression You sold yourself to the enemy.
It is one thing to say that we cannot formulate precise rules for the disambiguation of a concept (because it all depends on each individual’s beliefs), and another to say that the meanings of terms in a given language, which to some extent must be public, are nonetheless always negotiable, and not only in the shift from language to language but also within the same language, according to different pertinences.
Insofar as they are contents, meanings can always be identified even though they fluctuate and coagulate, and for some speakers shrivel until they all but prevent them from speaking appropriately or from recognizing something. But I see no reason why a contractual view of the sense of utterances must rule out, on the one hand, the existence of a grain that binds our cognitive types and, on the other, the linguistic conventions that register these bonds and supply the basis for successive interpretations and negotiations.26
There is no doubt that if I, sitting in a car beside the driver, urge her on, saying, You can go, the traffic light is blue, she will instantly understand that I meant to say green (or she will think that I am color-blind, or that it was merely a slip on my part). Perhaps this happens because the meaning of the words does not count and she understands me only because she attributes to me a belief similar to her own? And what would have happened if in that moment I had said You can go, because 7 is a prime number? Would she have thought that, as I am like her, I could only have been referring to the green light? Or would not the strength of the words, independently of the situation, have obliged her to try to understand what I meant to communicate, perhaps implicitly, because my observation was certainly mathematical and had nothing to do with traffic circulation?
4.7.2 Meaning and the Text
I have said that certain surprises regarding the flexibility of our semiotic instruments spring from the fact that, in almost all discourses on the elusive nature of meaning, there is confusion over the meaning of the terms and the sense of the utterance. But there is more to the problem than this. There is also confusion with regard to elementary utterances and texts.
In the traffic-light example, the dialogue cannot stop at that point. The driver has to
