It is obvious that Robinson, on his desert island, could reconstruct an analogue of society only by giving himself, all at once, all the rules and laws which are reciprocally implicated, even when they still have no objects. The conquest of nature is, on the contrary, progressive, partial, and advances step by step. Any society whatsoever has all of its rules at once – juridical, religious, political, economic; laws governing love and labour, kinship and marriage, servitude and freedom, life and death. But the conquest of nature, without which it would no longer be a society, is achieved progressively, from one source of energy to another, from one object to another. This is why law weighs with all its might, even before its object is known, and without ever its object becoming exactly known. It is this disequilibrium that makes revolutions possible. It is not at all the case that revolutions are determined by technical progress. Rather, they are made possible by this gap between the two series, which solicits realignments of the economic and political totality in relation to the parts of the technical progress. There are therefore two errors which in truth are one and the same: the error of reformism or technocracy, which aspires to promote or impose partial arrangements of social relations according to the rhythm of technical achievements; and the error of totalitarianism, which aspires to constitute a totalisation of the signifiable and the known, according to the rhythm of the social totality existing at a given moment. The technocrat is the natural friend of the dictator – computers and dictatorship; but the revolutionary lives in the gap which separates technical progress from social totality, and inscribes there his dream of permanent revolution. This dream, therefore, is itself action, reality, and an effective menace to all established order; it renders possible what it dreams about.
Let us return to Levi-Strauss’s paradox: two series being given, signifying and signified, there is a natural excess of the signifying series and a natural lack of the signified series. There is, necessarily, a "floating signifier, which is the servitude of all finite thought, but also the promise of all art, all poetry, all mythic and aesthetic invention." We would like to add that it is the promise of all revolutions. And then there is on the other side a kind of floated signifier, given by the signifier "without being thereby known," without being thereby assigned or realised. Levi-Strauss proposes to interpret in this way the words "gadget" or "what-not," "something," "aliquid," but also the famous "mana" (or, yet again, "it" [ca]). This is a value "in itself void of sense and thus susceptible of taking on any sense, whose unique funciton would be to fill the gap between the signifier and signified." "It is a symbolic value zero, that is, a sign marking the necessity of a symbolic content supplementary to that which already charges the signified, but able to take any value whatsoever, on the condition that it belong to the available reserve … " It is necessary to understand that the two series are marked, one by excess, the other by lack, and that the two determinations are interchanged without ever reaching equilibrium. What is in excess in the signifying series is literally an empty square and an always displaced place without an occupant. What is lacking in the signified series is a supernumerary and non-stiuated given – an unknown, an occupant without a place, something always displaced. These are two sides of the same thing – two uneven sides – by means of which the series communicate without losing their difference. It is the adventure in the Sheep’s shop or the story that the esoteric word narrates.
We may, perhaps, determine certain minimal conditions for a structure in general: 1) There must be at least two heterogeneous series, one of which shall be determined as "signifying" and the other as "signified" (a single series never suffices to form a structure.). 2) Each of these series is constituted by terms which exist only through the relations they maintain with one another. To these relations, or rather to the values of these relations, there correspond very particular events, that is, singularities which are assignable within the structure. The situation is very similar to that of differential calculus, where the distributions of singular points correspond to the values of differential relations. For example, the differential relations among phonemes assign singularities within language, in the "vicinity" of which the sonorities and significations characteristic of the language are constituted. Moreover, it seems that the singularities attached to a series determine in a complex manner the terms of the other series. In any case, a structure includes two distributions of singular points corresponding to the base series. And for this reason, it is imprecise to oppose structure and even: the structure includes a register of ideal events, that is, an entire history internal to it (for example, if the series include "characters," it is a history which connects all of the singular points corresponding to the positions of the characters relative to one another in the two series). 3) The two heterogeneous series converge toward a paradoxical element, which is their "differentiator." This is the principle of the emission of singularities. This element belongs to no series; or rather, it belongs to both series at once and never ceases to circulate throughout them. It has therefore the property of always being displaced in relation to itself, of "being absent from its own place," its own identity, its own resemblance, and its own equilibrium. It appears in one of the series as an excess, but only on the condition that it would appear at the same time in the other as a lack. But if it is in excess in the one, it is only as a supernumerary pawn or an occupant without a compartment. It is both word and object at once: an esoteric word and exoteric object.
It has the function of articulating the two series to one another, of reflecting them in one another, of making them communicate, coexist, and be ramified. Again, it has the function of joining the singularities which correspond to the two series in a "tangled tale," of assuring the passage from distribution of singularities to the next. In short, it has the function of bringing about the distribution of singular points; of determining as signifying the series in which it appears in excess, and, as signified, the series in which it appears correlatively as lacking and, above all, of assuring the bestowal of sense in both signifying and signified series. For sense is not to be confused with signification; it is rather what is attributed in such a way that it determines both the signifier and the signified as such. We can conclude from this that there is no structure without series, without relations between the terms of each series, or without singular points corresponding to these relations. But above all, we can conclude that there is no structure without the empty square, which makes everything function.
– Gilles Deleuze, The Logic of Sense, trans. Mark Lester, Continuum, London, 2004, pp. 58-61
