Although Deleuze grants, in What Is Philosophy? (5), that the production or creation of sense is not a privilege of the arts, in Proust and Signs he argues the contrary: “The fact remains that the revelation of essence (beyond the object, beyond the subject himself) belongs only to the realm of art. If it is to occur, it will occur there. This is why art is the finality of the world” (33). This claim is softened somewhat by the fact that Deleuze describes this revealing character of art, and of Proust’s Search in particular, as its philosophical bearing (PS, 60), thereby diminishing the difference between art and philosophy. But we can escape this question of privileges by rephrasing it: Why is art capable of expressing sense? For one thing, because art resists the objectivist and subjectivist reductions to which perception, desire, and thought fall prey. The sense of a work of art cannot be situated in the substance the latter consists of, or in the opinions that it would represent. According to Deleuze, in an artwork, “substances (matières) are ductile, so kneaded and refined that they become entirely spiritual” (PS, 31). Substance is reduced to a minimum. By the same token, the artwork is not a riddle that is solved once the ideas and the opinions behind it are known. On the contrary, the sense of an artwork is situated in between the different ideas and opinions that intersect in the work. The artwork passes through its substance and subjective opinions rather than dwelling in them. The second reason why art is capable of expressing sense is that it recognizes the differential character of sence (PS, 27). This shows itself, among other things, in the fact that the sense of a work of art is always multiple: it can always be interpreted in several ways. Moreover, most of the time the interpretation changes as the reader changes: the Lewis Carroll one reads as a child is not the same Lewis Carroll one reads as an adult. However, the more fundamental point is that sense itself is constantly jumping during the interpretation process: when we notice certain details, our understanding of the personality of the main character, for example, can change, and that in its turn changes our interpretation of the whole book. Thus, sense is not only ambiguous but fundamentally ungraspable. And finally, art can be considered the expression of sense because it often happens in art that the unexplainable sense of one work is taken as the referent for another work, whose sense is again inexplicable but can be taken as the referent for a third artwork, and so on. In contemporary art, in particular, this referencing game is very present.
Thinking Thought as Learning How to Create Problems (Eighth and Seventh Postulates)
Let us recapitulate what we have found out about thinking thought so far. Thinking thought is instigated by the unforeseen confrontation with a sign whose sense it needs to unfold. During this process, thought is brought to confront its own limits: the sign causes problems because it forces us to reconsider our distinctions, ideas, and so forth. Eventually, thought even comes upon the unthinkable—which is, simultaneously, that which must be thought. The essence is empirically unthinkable because it is differential by nature; its differentiality makes a definitive “com-prehension” of it impossible. To know the essence is a contradiction in terms. The interpretation of signs, conversely, is an endless process of learning: the sense of a sign can only be learned, not known.
What does it mean to describe sense as the object of learning? Learning is not, as the eighth postulate, the postulate of knowledge, states, and, as the first postulate presupposes, a preparatory and temporary stage that finally dissolves when it reaches its goal, knowledge. To learn, for Deleuze, is to “enter into the universal of the relations which constitute the Idea, and into their corresponding singularities”; knowledge, conversely, “designates only the generality of concepts or the calm possession of a rule enabling solutions” (DR, 204). In order to understand this definition, we have to look at the seventh postulate, the postulate of solutions, which holds that representational thought traces problems from supposedly preexistent propositions and evaluates them according to their solvability, to their susceptibility for a solution. Representational thought poses a problem in function of the propositions it already has, and in function of the propositions that can possibly solve the problem. The problem itself is secondary. Deleuze, for his part, wants to think problems in themselves. Hence, he understands thinking thought not so much as the process of finding solutions to problems, but as creating problems. Deleuze agrees with the representational idea that “a problem is determined at the same time as it is solved” (DR, 203), but for different reasons, namely, because a problem is explicated in the solution. Thus, the problem, rather than disappearing in the solution, insists and persists in it (DR, 203). The problem is immanent and transcendent in relation to its solutions because its persistence does not imply that it is deduced from the solutions.
Now, thinking the problem in function of the propositions we have at our disposal means thinking in terms of particularities, since propositions are always particular. These particular propositions can be examined for what they have in common so as to be able to establish general principles. Tracing problems from propositions implies thinking them in terms of generalities and particularities. However, if we think the problem in itself, we think it in function of universality and singularities. Is this not a contradiction? No. We already saw how the Idea synthesizes in a way that is disjunctive, which means that it unites not by looking for what there is in common—that is what a generality does—but by playing out the differences between what is subsumed under this unity. Universality thus comprises a distribution of singular and distinctive points (DR, 202). Hence, the difference between knowledge and learning is that in knowledge, the solution “lends its generality to the problem,” whereas in learning, the problem “lends its universality to the solution” (DR, 202). Learning is a matter of penetrating the Idea (DR, 243), that is, the problem. Learning is about starting to see where the problems are. When we learn to swim, for example, our “knowledge” of water and of our own body changes because certain things, such as the ability to breathe, for instance, lose their self-evidence. We need to explore the singular points of the water (its weight, its movement, etc.), of our body (how to regulate breathing, etc.), and of the combination of both (under what circumstances do we need to lift our head higher so as not to swallow water, for instance?). Learning is not about acquiring something that already exists but about making your world more complex, about creating more distinctions and relations.
This description of thought in terms of learning, of creating problems that do not presuppose solutions, also has implications for the notion of truth. Whereas representational thought connects truth with the solution—the solution is true or false with respect to the problem posed, which is understood in function of the solution—Deleuze connects it to the problem. A problem is false when it is overdetermined; when it mistakes banalities for profundities, ordinary points for singular points (DR, 191); or when it is underdetermined, in which case it fails to identify the singular points. In other words, the truth of a problem has to do with its meaningfulness. As such, it determines the nature of its solution: “The problem always has the solution it deserves in proportion to its own truth or falsity—in other words, in proportion to its sense” (DR, 198). A problem always has the solution that corresponds to the way the problem is posed.
Stupidity as the Highest Finality of Thinking Thought (Fifth Postulate)
I have already mentioned, in passing, how representational thought understands error: an error occurs when one faculty appropriates an object actually intended for another faculty. For example, one might say, “Good morning, Theodorus,” when, in fact, Theaetetus is the one passing by, because one confuses Theodorus, whom one saw yesterday (memory), with Theaetetus, who is right now before one’s eyes (perception). An error is due to a failure of good sense. Representational thought also offers another explanation for error, namely, the false recognition or representation resulting from a false evaluation of opposition, analogy, resemblance, and identity (DR, 186). Thus, one says “Theodorus” instead of “Theaetetus” because of an overestimation of the resemblance between the two. Both explanations, however, see error as a temporary blinding, occasioned by external forces, of a thought that is by nature upright and thus in perfect alignment with the other postulates. Representational thought considers error to be an empirical fact, an extrinsic attack on the natural affinity with the truth.
According to Deleuze, the negative of thought