This is why it is said that artists enhance our capacity to perceive the environment. An artist (and this is what the Russian formalists meant by their concept of defamiliarization) continuously tries to revise current CTs, as if everything were perceived as a hitherto unknown object. Cézanne or Renoir trained us to look in a different way, in certain circumstances of particular felicitousness or perceptual freshness, at foliage, fruit, or the complexion of a young girl.
There are lines of resistance in the stimulating field that oppose uncontrolled artistic invention (or that oblige the artist to portray objects not of our world but of a possible world). This is why the artist’s suggestions are not always completely absorbed by the Community. It would be hard for us to conceive a CT of the female body inspired by Duchamp’s Mariée mise à nu, and yet the work of artists always tries to call our perceptual schemata into question, if in no other way than by inviting us to recognize that in certain circumstances things could also appear to us differently, or that there are alternative possibilities of schematization, which make some features of the object pertinent in a provocatively abnormal way (the skeletal lankiness of bodies, for Giacometti; the uncontrollable tendencies of the flesh and muscle, for Botero).
I recall an evening in which people were playing parlor games, including a variant of «statues,» in which the onlookers had to guess which work of art the players were miming. At a certain point a (well-composed) group of girls presented themselves with their limbs and faces twisted into distorted positions. Almost everybody recognized the reference to Demoiselles d’Avignon. If the human body can interpret Picasso’s representation of it, then that representation had caught certain possibilities of the human body.
Chapter Four
The Platypus between Dictionary and Encyclopedia
4.1 Mountains and Mountains
As usual, let us imagine a situation. When Sandra tells me she is going to cross Australia from north to south by car, I tell her that she must not forget to visit Ayers Rock, which stands in the center of the continent and is one of the world’s many Eighth Wonders. I add that if, en route between Darwin and Adelaide, she passes through Alice Springs, she should then head southwest into the desert until she sees a mountain, hard to miss because it rises in the center of the plain like Chartres cathedral in the middle of the Beauce: this is Ayers Rock, a fabulous orographic formation that changes color according to the time of day and is stunning at sunset.
I have given her instructions not only for finding but also for identifying Ayers Rock, and yet I feel slightly ill at ease, as if I were deceiving her. And therefore I tell her that while I was telling the truth when I told her that (ia) Ayers Rock is a mountain, I am nonetheless also telling the truth if at the same time I state that (iia) Ayers Rock is not a mountain. Obviously, Sandra reacts by reminding me that a minimum of truth-functional good breeding requires that if (ia) is true, then (iia) must be false, and vice versa.
So I reiterate the difference between NC and MC (in this story, Sandra has already read this book, apart from this paragraph), and I explain to her that Ayers Rock displays all the characteristics that we attribute to mountains, and that if we were asked to divide the objects we know into mountains and nonmountains, then we should certainly put Ayers Rock in the first category. It is true that we are accustomed to recognize a mountain as something that rises to a great height after being preceded by hilly slopes that get steadily steeper and steeper, while Ayers Rock rises solitary and precipitous from the middle of the plain; but the fact that we are dealing with a curious, atypical mountain should not worry us more than the fact that the ostrich, insofar as it is a bird, is equally curious and atypical, without its being perceived as less of a bird for this reason.
Nevertheless, from a scientific point of view, Ayers Rock is not a mountain, it is a stone: it is a single stone—in other words, a monolith planted in the ground as if a giant had hurled it down from the sky. Ayers Rock is a mountain from the point of view of the CT, but it is not from the point of view of the MC, i.e., of a competence definable as petrological or lithological or what have you.
Sandra understands very well why I did not tell her to proceed southwest until she saw a stone—because in that case she would have gone on with her gaze fixed on the ground without looking up. However, she might say that, as I am in the mood to play with logical paradoxes, I would do better to rewrite (ia) and (iia) in this way: (ib) Ayers Rock is a mountain and (iib) Ayers Rock is not a MOUNTAIN. In this way it would be clear that (ib) asserts that Ayers Rock has the perceptual qualities of a mountain, while (iib) would assert that it is not a MOUNTAIN in a categorial system. Naturally Sandra would use her voice to emphasize, with suprasegmental features, the use of small capitals, precisely to show that terms written in such characters stand for what compositional semantics calls dictionary properties, which are semantic primitives for some, and which in any case imply a categorial organization, in the sense of the expression as used in the preceding chapter.
But at that point she would have me notice a curious paradox. The supporters of a dictionary representation maintain that such representations take account of relations within the language, leaving aside elements of knowledge of the world, while knowledge in an encyclopedic format presupposes extralinguistic knowledge. In order to provide a rigorous explanation of the functioning of language, the supporters of a dictionary representation maintain that we must turn to a package of semantic categories that are organized hierarchically (such as OBJECT, ANIMAL VS. VEGETABLE, MAMMAL vs. REPTILE) and that are of such a kind that—even when we have no knowledge of the world—various inferences can be made, of the type If mammal, then animal; If this is a mammal, then it is not a reptile; It is impossible for something to be at once a reptile and not an animal; If this is a reptile, then it is not a vegetable; and many other pleasant apothegms that, according to the experts, we habitually utter when, for example, we realize we have picked up a viper instead of an asparagus spear.
Encyclopedic knowledge, on the other hand, would be uncoordinated by nature, with an uncontrollable format, and the encyclopedic content of dog would have to include practically all that is and could be known about dogs, even details such as the fact that my sister has a bitch called Best—in short, a knowledge that would be too much even for Borges’s Funes el Memorioso. Naturally it is not quite like this, because we can consider as encyclopedic knowledge only those items that the Community has in some way registered publicly (and moreover it is maintained that encyclopedic competence is shared across sectors, according to a sort of linguistic division of labor, or activated in different ways and formats according to the context). But there is no doubt that, with regard to the events and objects of this world, not to mention those of other worlds, there are always new facts to learn, and therefore those who find the encyclopedic format hard to handle are not wrong.
Nonetheless the curious accident has occurred whereby, given that the repertoires that succinctly record the properties of terms are called «dictionaries» while those that indulge in complex descriptions are called «encyclopedias,» everybody thinks that dictionary competence is the indispensable one for the use of language. Instead, what the story of Ayers Rock tells us is that, in order to recognize that object and to be able to talk about it every day, the perceptual (not linguistic) characteristic of appearing like a mountain (on the basis of many factual properties) counts for a very great deal, while the fact that it is not a MOUNTAIN but a STONE is a datum reserved only for an elite that shares a vast encyclopedic competence. Therefore Sandra would point out to me that people, when speaking plainly, run on encyclopedia mode, while only the learned turn to the dictionary. Nor would she be wrong.
The whole business could be confirmed in historic terms too. If we take a look at Hellenistic and medieval encyclopedias, we find only descriptions that either tell us what something looks like (for Alexander Neckham, the crocodile was a serpens aquaticus bubalis infestus,
