A curious position: if there is inference in the perception itself, then there is fallibilism, and in fact Peirce also deals with perceptual illusions (CP 5.183); and yet it seems that at the same time these perceptual inferences are not hypothetical but «apodictic.» A fine and explicit affirmation of realism, were it not uttered by one who never ceased saying that perception is also semiosis and therefore already abduction. And finally, if perceptual judgment could not be called into question, we would have an intuition of the singular, an idea Peirce always rebelled against, right from his anti-Cartesian writings. Moreover, if that which cannot be called into question, and is singular, is the «percept» (and the percept is identified with the Ground), it cannot set off inferential processes that have to do only with general terms (CP 5.298). If there is an abstractive moment in perception, then there is interpretation, even if rapid and unwitting (see Proni 1990: 1.5.2.4), and if there is interpretation, there is «possible criticism.»
If we were to forget these subtleties (and the inevitable contradictions found in writings from different periods), we might be able to cut a long story short this way: agreed, there is an unclear blend in that space lying between Firstness (Ground or non-Ground) and fully realized Thirdness; there is a first moment of reaction of the senses that is unquestionable; the moment in which the quality presents itself to me as the quality of something (Secondness), this something becomes the premise of every other inference, in the sense that I know that in any case there is a Dynamical Object that is triggering the chain of my responses. At this point the work of interpretation begins, and, when perceptual judgment establishes itself and takes shape, it resolves itself into the formation of the Immediate Object.
Some aspects of the Ground converge in the Immediate Object (it has the nature of an icon, of Likeness) as well as all the aspects of perceptual judgment (it presents itself as the point of departure of every subsequent interpretation). At most we can say that there are also Immediate Objects of something we do not know through perception (there must certainly be two Immediate Objects that correspond to the terms president and Alpha Centauri). But it should not trouble us too much if we think that an icon is not necessarily an image in the visual sense of the term, because also the melody we whistle, perhaps tunelessly, may be an icon of Beethoven’s Fifth; and because even a graph has an iconic nature—though it perhaps does not betray any morphological similarity with the situation represented.
We could therefore gain a breathing space by recognizing that, while the notion of the Ground and the very nature of perceptual judgment are still obscure, the same cannot be said of the notion of the Immediate Object. It is the object as it is represented (CP 8.343), in the respect in which it is thought (CP 5.286), it is the type of which the Dynamical Object that triggered the sequence of responses was the token (Prodi 1990: 265).4 To some extent it eludes the individuality of perception, because insofar as it is interpretable, it is already public and intersubjective; it does not tell us all about the object, but it is only by coming to it that finally we know and can say something about the object.
Now, in this process and in the moment of its first fulfillment, it appears to me that a problem arises that Peirce had already come across in Kant. Peirce is trying to reformulate, without deducing it transcenden tally, the Kantian notion of schema.
Is Peirce really thinking of Kantian schematism? Is it by trying to distinguish the categories (but which, his or Kant’s?) from the schema and the categories from the manifold of the sensible intuition that an apparently inextricable knot between the Ground and the Immediate Object is created?
Peirce always returns, almost parenthetically, to the Kantian notion of schema. In CP 2.385 he says without hesitation that the Kantian schema is a diagram; but he speaks of it in an abstract fashion, in reference to the postulates of empirical thought in general, and within a framework of modal logic. However, in 1885, he said that the doctrine of schemata must have come to Kant’s mind only late, when the system of the first Critique had already been laid down: «[F]or if the schemata had been considered early enough they would have overgrown his whole work» (WR 5: 258–59). It seems like a research program, the identification of a breach through which it should be possible to arrive at a nontranscendental Kantism. But what had Peirce understood of schematism, of which even Kant, as we shall see, had understood something only step by step?
2.3 Kant, Trees, Stones, and Horses
Is there a reason why Peirce, the future theoretician of semiotics, started by reading and rereading Kant, considering Kant’s table of judgments and categories as if they had been handed down to him from Sinai?5
Kant has been reproached for a radical lack of attention to the problem of semiotics. But as Kelemen (1991) notes, since Hamann’s and Herder’s day this lack has been attributed to the fact that Kant considered a very close nexus between language and thought to be implicit, and it has been suggested that this nexus presents itself precisely in the doctrine of schematism, so much so as to suggest that the schema was concept-word (Wortbegriff). On the other hand, it cannot be denied that there is an implicit semiotics in the distinction between analytic and synthetic judgments, that there is explicit discussion of the theory of signs in the Anthropology,6 and that it is possible to read the entire Logic in semiotic terms (see Apel 1972). Moreover, the nexus between knowing and communicating has been repeatedly underlined, a nexus of which Kant speaks in numerous passages, even though he does not dwell on the subject overmuch, as if he considered the question obvious (Kelemen 1991: 37). Finally, and we shall be coming back to this, there are the semiotic pages of the third Critique.
In any case it should suffice to consider the purely verbal origin of Kant’s categorial apparatus as much as Aristotle’s and quote a celebrated remark of Heidegger’s:
[B]eings equipped with an intuitive capacity must always be able to mingle in the intuition of the entity, but finite intuition, insofar as it is intuition, always remains in the first place anchored to the singular intuited from time to time. The entity intuited is known only if everyone can make it comprehensible to himself and to others, and succeeding in this way in communicating it. (1973, 1, 2)
To talk of that which is means rendering what we know communicable. But to know it, and to communicate it, implies recourse to the generic, which is already an effect of semiosis, and depends on a segmentation of the content in which the Kantian system of categories, bound fast to a venerable philosophical tradition, is a cultural product that is already established, culturally rooted, and linguistically anchored. When the manifold of the intuition is ascribed to the unity of the concept, the percipienda are by that time perceived as culture has taught us to talk about them.
That a semiosic foundation is implied by the general framework of Kantian doctrine is one thing, but whether Kant ever elaborated a theory of how we assign names to the things we perceive, be they trees, dogs, stones, or horses, is another matter.
Given the question «How do we assign names to things?,» in the way that Kant had received the problem of a theory of knowledge, the answers were in brief two. One was provided by the tradition we will call Scholastic (but begins with Plato and Aristotle): Things present themselves to the world already ontologically defined in their essence, raw material modeled by a form. It is of no importance to decide whether this form (universal) is ante rem or in re: it offers itself to us, splendid in its individual substance, is grasped by the intellect, is thought and defined (and therefore named) as a quiddity. The work of our mind amounts to what the active intellect (wherever it may work) does in the blink of an eye.
The second answer was provided by the British Empiricists: We do not know substances, and if there were any, they would not reveal anything to us. What we do have, according to Locke, are sensations, which propose simple ideas, both primary and secondary, but still disconnected: a rhapsody of weights, measures, dimensions, and then colors, sounds, tastes, and reverberations, which change with the time of day and the state of the subject. Here the intellect is active, in the sense that it works: it combines, correlates, and abstracts, in a way that is certainly spontaneous and natural, but only thus does it coordinate simple ideas into those compound ideas to which we give the name of man, horse, tree, and then triangle, beauty, cause, and effect. To know is to put a name to these compositions of simple
