We certainly possess cognitive types of sound sequences, if we can usually distinguish the timbre and the rhythm of the telephone bell from those of the doorbell, military calls such as taps and reveille, and often the melodies of two songs we know well.
If we admit that semiosic primitives exist, they are certainly elementary experiences such as walking, jumping, or running. When we skip, we are aware (or could be aware, if we pay attention to what we are doing) whether we use the right foot twice and the left foot twice or always the same foot. Yet it so happens that these last two operations have two distinct terms in English but not in Italian. This table by Nida (1975: 75), figure 3.2, distinguishes between the content of some English terms for motor activities.
Figure 3.2
Anyone who wants to translate a text describing these operations from English to Italian must interpret the terms according to this table, which—even though it uses linguistic terms—provides instructions of a motor type (one might very well think of its translation into film footage, or into a series of diagrams that use signs improperly called «iconic»).26
3.4.6 «Affordances»
The CT ought to include those conditions for perception that Gibson calls «affordances» (and Prieto would have called pertinency):27 the various tokens of the type «chair» are recognized because we are dealing with objects that make it possible for one to sit down, while tokens of the type «bottle» are recognized because they are objects that allow us to hold and pour liquid substances. It is instinctive for us to recognize a tree trunk as a possible seat and not a column (unless someone is a stylite), due to the length of our legs and the fact that we find it comfortable to sit with our feet resting on the ground. On the other hand, in order to categorize a knife, fork, and spoon among Cutlery, or a chair and a cupboard among Furniture, we must leave aside this morphological pertinency and fall back on more generic functions, such as the manipulation of food and the preparation of a habitable environment.
Our capacity to recognize affordances is registered, so to speak, in linguistic usage itself. Violi (1991: 73) wonders why, when faced with a table with a vase standing on it, we are led to interpret verbally what we see as The vase is on the table and not The table is under the vase. She suggests that «the selection of linguistic expressions seems regulated by complex configurations of the intentional relations between the subject that moves in space and the objects that surround it.» But this is equivalent to saying that our CT of the common vase also includes the sequence of actions that it permits, and so a vase is something easily movable that usually stands on something. On the other hand, our CT of the table includes not only its morphological features but also the notion (I would say, nuclear) that it is used for putting something on (and never for being inserted under something).28
But Arnheim (1969: 139) suggests that language can block our recognition of pertinency. Quoting a remark made by Braque, he admits that a coffee spoon acquires perceptual saliency that differs according to whether it is set alongside a coffee cup or inserted between shoe and heel like a shoe horn. But often it is the name with which we indicate the object that highlights one pertinency at the expense of others.
In conclusion, we still have imprecise ideas about the extremely various ways in which our CTs are organized—and how they express themselves in NC. I would tend to follow Johnson-Laird’s (1983: 7) proposal, whereby from time to time different types of representation gradually offer themselves as options for the codification of different types of information, and in general we move from real images to mental «models» (like Marr’s 3-D representation) and real propositions.29 Rather than talk of «double coding,» as is usual in these cases, I think we ought to talk of multiple coding, of our capacity to maneuver the same CT on different occasions by accentuating either the iconic component, or the propositional one, or the narrative one regarding our capacity to activate—within the ambit of a complex situation—more complex nuclear contents and information.30
All this induces us to review, I should like to say with indulgence, those fairly rigidified semantic representations (compositional analysis models, case grammar, contextual and circumstantial selections; see A Theory of Semiotics 2.10–2.12) that seem to be challenged by a reconsideration of the complex way (certainly not linear, but like a network^) in which our cognitive types are organized and of how we interpret them through nuclear content. These skeletal models are naturally stenographic forms that consider our NCs from a certain standpoint, according to what we wish to emphasize within the framework of a set theoretical discourse, or according to how we wish to indicate the ways followed in order to have a certain contextual disambiguation of the terms. With such models we interpret the quantity of NC we need as the need arises. They are metalinguistic (or metasemiotic) interpretations of interpretations rooted in perceptual experience.
3.5 Empirical Cases and Cultural Cases
Until now I have dealt with CTs that concern «natural kinds» such as mice, cats, and trees. But we have said that there are certainly CTs for actions such as walking, climbing, and skipping. The expression «natural kinds» is insufficient: CTs for artificial kinds, such as chair, boat, or house, obviously exist. Let us say, then, that I have considered CTs for all the objects or events we can know through perceptual experience. I cannot manage to identify a suitable term for indicating various objects of perceptual experience, and I choose the expression «empirical cases» (on the model of the Kantian empirical concepts): an empirical case would be the fact that I perceive or recognize a cat, a chair, someone sleeping or walking, and even that a certain place is a church and not a railway station.
It is a different matter with «cultural cases,» among which I would put a disparate series of experiences with regard to which we can certainly discuss whether it happens that what I name in a particular way is named correctly and whether I recognize something that others are also supposed to recognize. Nevertheless the definition of these «cases,» just as the instructions for their recognition, depend on a system of cultural assumptions. Among the cultural cases, I would put the functional genera (such as cousin, president, archbishop); a series of abstract concepts (such as the square root), which can also objectively «exist» on some Platonic Third World but which are definitely not objects of immediate experience); events, actions, relationships (such as contract, swindling, emphyteusis, or friendship). What is common to all these cases is that, if they are to be recognized as cultural, they require a reference to a framework of cultural rules.
This distinction could correspond to the one Quine made between those occasion sentences that are at the same time observation sentences and those that are not. One might agree with him. Except that, as we shall see, This is a bachelor is not completely nonobservational.
In the case of the bachelor, Lakoff (1987) would talk of Idealized Cognitive Models (ICM): it is hard to say when the term should be applied, but ideally it has a sense. Lakoff is thinking of the last phase of the debate on bachelors, which boasts a long history featuring an admixture of highly sensible observations and mere witticisms.31 It has been said that it is questionable whether the definition «unmarried adult male» can really circumscribe bachelors, because «unmarried adult males» includes Catholic priests, homosexuals, eunuchs, and even Tarzan (at least in the novel where he does not meet Jane), with the result that we cannot define them as bachelors unless our intentions are humorous or metaphorical.
The reply has been made, with a good deal of common sense, that bachelors are definable not only as unmarried adult males but also as adult males who have chosen not to marry (for a period marked by indefinite temporal limits) even though they are physically or socially able to do so; which therefore does not apply to the eunuch (unmatched as a result of a life sentence), to Tarzan (quite unable to find a partner, within the time limits), to the priest (celibate by obligation), and to the homosexual (unmarried out of a natural impulse toward other forms of union). In a situation where homosexuals can legally marry persons of the same sex, it would be possible to distinguish homosexual bachelors, who do not live in couples, from married homosexuals.
It is evident that, even once these specifications have been made, to be able to talk about bachelors we need other negotiations bound up with the circumstances. For example, a homosexual could marry a person of the opposite sex for social convenience (e.g., if he or she were heir to a throne) without ceasing to be a homosexual for all that, while a priest could not marry a person of the opposite sex without rejoining the ranks of the laity and ceasing to be a priest; and so—if we cared to—we could say that a homosexual bachelor is more of a bachelor than a priest. But since a priest who has not been defrocked but has been suspended a divinis may get married at a registry office in Reno, is a priest who
