As we can see, the negotiations can continue indefinitely, and that is why today, customs having changed, the word bachelor is almost never used anymore (also, it has particular connotations of the free and easy life and evokes the complementary notion, equally in desuetude, of the unmarried lady or, even, the «spinster»). And so bachelors are now a part of the hazy archipelago of «singles,» which includes unmarried adults of both sexes, homosexuals or heterosexuals, divorcees, widowers, spouses on the rocks, and spouses still madly in love with their partner but obliged to work in New York while their partner has found a job in California. Lakoff’s notion of ICM is still valid in the sense that, while an idealized definition of bachelor does not always allow us to say whether someone is a bachelor, it certainly does allow us to say that he is not a bachelor if he is the happily married (and cohabiting) father of five children.32
Nevertheless the fact that notions of this kind require negotiation on the basis of conventions and behavior bound up with cultures does not allow us to exclude that the occasion sentences they permit have no observational basis.
Consider the difference between killing and murdering. Someone’s killing another is directly perceivable: in some way we have a CT of killing, in the form of a fairly elementary scenario; we recognize we are faced with a killing when someone strikes another living being and thereby causes its death. I think the experience of killing is common to different cultures. It’s another case with murder: a killing can be defined as homicide in self-defense or culpable or without malice aforethought, as ritual sacrifice, as an act of war recognized by international convention, or finally as murder, depending entirely upon the laws and customs of a given culture.
What is puzzling about this difference between empirical cases and cultural cases is that the first are without a doubt based on the testimony of the senses, but it cannot be said that experiential data are devoid of value in the second. Just for a start, an act cannot be recognized as murder unless there is experience (direct or indirect) of the fact that it was a killing.
Granted therefore that there is a difference between empirical and cultural cases and seeing that there are CTs for empirical cases, do we also have CTs for cultural cases?
One could avoid this perplexing question by saying that CTs concern the objects of perceptual experience and that’s all. For other concepts, ones expressed by linguistic terms, there are no CTs, only NCs, which would be the same as saying that some things are known to us on the basis of perceptual experience while others we know only through definitions, duly contracted within the ambit of a culture. Which brings us back to Russell’s distinction between object-words and dictionary words (see Russell 1940), except for our broadening of the concept of object-word to include natural genera and qualia as well as experiences of other kinds.
But, seeing that the CT has been defined as «something in the head,» which allows us to recognize something and name it as such, even though it has not yet been publicly interpreted in terms of NC, can we perhaps say that when we pronounce the word cousin or president, we have nothing in our heads, and certainly not anything remotely similar to the Kantian schema? Note that the question remains even if it is admitted that we do not think in images but only by processing abstract symbols. In this second case, the question ought simply to be reformulated as follows: Is it possible when we state that something is a cat, that we process something «in the head,» while when we state that X is Y’s cousin, we process nothing?
When I understand the meaning of cousin and president, I call up in some way a kinship or an organizational schema, a Peircean graph. What happens when I understand that, in correspondence with the Italian term nipote, there are two different positions in the kinship schema, expressed in English by nephew and grandson? It is true that I can express the difference verbally too (which brings us to the NC), and so there is a nipote who is the child of an uncle and a nipote who is the child of a child. But the question—which I do not feel like answering, owing to my intention not to stick my nose into the black box—is whether this verbally expressed NC is all I know about the difference, or whether it constitutes the ver bal interpretation of a difference grasped and understood via a diagram.
A supporter of the eminently visual nature of thought, such as Arnheim, seems to surrender before an example made by Biihler: asked to respond to the question «Should it be lawful or not to marry the sister of one’s own widow?» the subjects asserted they had come to understand that the statement was senseless without the help of images (1969, 6). Of course, and especially in the case of a person with a well-trained mind, the answer to the question can be arrived at propositionally. But on repeating the experiment, I also found someone who came to recognize the contradictory nature of the question by imagining a widow weeping, with her sister beside her, over the grave of her own husband (and intuitive evidence suggests that a husband in the grave is unlikely to get married).
The same holds good for the Italian word presidente, and even more so when I have to decide whether the apparent synonymy (in English) is a good translation. As a matter of fact, not only is an American president unlike (in constitutional terms) an Italian president (their relations of power are expressed by two different organization charts), but also in the world of business what Italians call the Presidente of a company is the equivalent of the Chairman of the Board in UK English, while the role of President of an American company is very like that of the Direttore Generale (Managing Director) in an Italian one.
In this case too the difference becomes evident on considering the position of the President in a company organization chart. Naturally the organization chart can be interpreted verbally, by saying that the president is the man or woman who gives orders to X or Y but not to K (who gives orders to him or her), but this would be the same as saying that expressions such as above or below may be interpreted only verbally (in terms of an NC), while we know very well that we translate them mentally in terms of CTs. And the fact that someone is the boss of a group of gangsters that we see in action can be inferred through perceptual experience. Does this mean, therefore, that there is a CT for boss while there is none for president?
Many people would be incapable of interpreting in words or with other signs the NC of the word murder, and yet on seeing someone cracking an old lady’s skull and then snatching her handbag before fleeing, they would realize they were witnessing a murder. Is there not a CT (a frame, or narrative sequence) for murder, therefore?
It would be puzzling to say that in order to recognize a triangle or a hypotenuse, or the fact that there are two onlookers rather than three, things are based on perceptual experience (and therefore there is a CT for these empirical cases), while it is not on the basis of a CT that we recognize 5,677 to be an odd number. Identifying an odd number, even a very large one, depends on a rule, and this rule is certainly an instructional schema. If there is a system of instructions for recognizing a dog, why shouldn’t there be one for recognizing that 5,677 is an odd number?
But if there is a system of instructions for recognizing 5,677 as an odd number, why shouldn’t there be a system of instructions for recognizing whether a certain agreement is a contract? Is there a CT for contracts?
It is agreed that the instructions for recognizing an odd number are of a different kind to those we have introjected in order to recognize a dog. But in the discourse on schematism in 2.5 we acknowledged that to characterize the schema as a system of instructions it is not indispensable for the instructions to be morphological in nature. We have already forsaken the idea of understanding CTs exclusively as visual images, and we have decided that they can also correspond to scripts or flowcharts for the recognition of a sequence of actions.
The quality of being a bachelor does not seem recognizable on the basis of experience. But does that of soccer referee? To be a referee is certainly not to belong to a natural kind: a camel is always a camel, but a referee is a referee only in certain moments or periods of his life. The functions of the referee are indeed expressed by verbal interpretations. But let us suppose we have been suddenly transported to the stands of a football stadium while a football match is in progress, even though no one, players included, is wearing a shirt that permits perceptual recognition. After a little we would be able to say, by inferring from each person’s behavior, which of the twenty-three people is the referee, just as we
