Andrei Rodin
Naming as Event
(about an Aristotelian interpretation and possible non-Aristotelian development of Kripke’s theory of meaning)
In the beginning of Posterior Analytic (71a10) Aristotle distinguishes between two sorts of "previous knowledge" or principles necessary for any inquiry: the knowledge "that it is" and the knowledge "what it is". Aristotelian science (episthmh) combines the two to grasp their unity that is desired essence. The procedure roughly is as follows. At first they suppose that there is something they want to investigate. Then they make some supposition about what this thing could be, ascribing to investigated thing as a subject certain predicates. Until it is not proven that the predicates are really inherent to the subject, they form some possible thing (with the condition that the predicate do not contradict to each other). To prove (through "middle term" of syllogism) that such-and-such predicates actually belongs to the subject of inquiry is to prove (1) what the investigated thing really is and (2) that the predicates relates to real thing, i.e. supposed possible thing (i.e. a "model") really exists. Proof gives us a thing that really exists being the same time "conceptually transparent". Such a thing Aristotle calls essence. For example we know that clouds sometimes thunder. Why does it happen, what is the reason? Aristotle takes as the reason the process of lightning’s going out (An.Post. 93b8). We could now add something about atmosphere electricity. Anyway the mechanism of thunder is "naively" treated as Aristotelian essence when we consider it as a theoretical construct that really exists (that is not only imaginary or abstract thing). If one can not prove inherence of a certain predicate to the subject it does not mean, generally speaking, that the predicate does not inherent to the subject nor that such a proof is not found yet. For there are contingent properties whose inherence can not be proven. It is interesting that Aristotle does not give us any criterion to distinguish between contingent "by nature" and contingent "for us", i.e. between situations when proof is impossible and when proof is unknown (1).
Kripke’s approach seems to be very similar. To designate an object rigidly according to Kripke [1] means to designate one and the same thing, even when nothing certain (necessary) about this thing is known. In Aristotle’s language it corresponds to "previous knowledge that it is" (without knowledge what it is). What is known about the thing a priori is not obligatory necessary and may be revised. According to Kripke, it is science that discovers necessary properties of its objects, for example that a molecule of water consists of one atom of Oxygen and two atoms of Hydrogen. According to Kripke as well as to Aristotle it is a matter of science to distinguish between necessary (essential) and contingent (superficial) features of investigated object.
We should not forget, however, that Kripke returns to Aristotelian views starting from modern transcendentalism and empiricism. He criticises empiricism when argues that meaning can not be treated subjectively as a certain cluster of sensorum data and insists on intersubjective nature of meaning. That is a sense of Kripke’s realism. Kripke’s controversy with transcendentalism is not so obvious but as I try to show below when he argues against the generally adopted way of speaking about "possible worlds" he actually argues against transcendentalism. An exposition of Kripke’s controversy with transcendentalism allows us to mention and to develop some non-Aristotelian moments of his theory. I think that to restore Aristotelian essences was not actually an objective of Naming and Necessity. I think that Kripke found himself on Aristotle’s side rather because after dismissing empiricism, transcendentalism and modern philosophy of Subject in general, he had no other background except the most influential predecessor of modern philosophy of Subject that is the ancient philosophy of Essence. I believe that a more accurate critic of modern philosophy would enable us to elaborate a new realistic perspective without restoration of essentialism. In the end of this paper I shall try to give a short sketch of such a perspective.
Kripke defines rigid designator as something that "in every possible world designates the same object" ([1], p.48). He notes then that in this definition "we don’t require that the objects exist in all possible worlds". Why Kripke does such a remark? What can make one to think that identity in all possible worlds implies existence in all possible worlds? Note, that Aristotelian interpretation of Kripke’s theory presupposes such an implication. I presupposed it myself, establishing a correspondence between Kripke’s rigid designation and Aristotle’s "knowledge that it is". In spite of the fact that therefore Aristotelian interpretation openly contradicts Kripke’s note mentioned above, I think that this interpretation is reasonable. I believe that Kripke himself had Aristotelian essences in mind when he elaborated his theory of meaning. Starting from possible worlds Kripke then left it for Aristotelian essences. He did not claim that he reduced a designation in possible world to designation of essences and so we can not accuse him in inconsistency. But he gave us no idea of how (actually) non-existent things could be rigidly designated. To include in Kripke’s theory the case of non-existent referent we should interpret it in a non-Aristotelian way. I think it is important since the idea of Kripke’s realism is to justify rather the social reality of meaning than the physical reality of "external world". I shall propose such an interpretation later but now let us consider a relation of Kripke’s theory to transcendentalism on the one hand and essencialism on the other hand in more details.
Kripke turns to Aristotle already on the pp.52-53 of his book discussing "transworld identification problem" that is the problem of criteria of being the same in different possible worlds. Kripke’s decision is as follows. Since we use the notion of possible worlds to describe different states of some objects, these object’s identity is presupposed from the very beginning and not need to be specially justified with respect to possible worlds construction. If you like (and/or if it gives you some technical profit) you may treat a possible situation where this table were in another room as if it would belong to "other possible world" but anyway you consider this table but not another one. Later you may discover that some properties of the table remain the same in all possible worlds (or, more accurate, in all possible worlds where this table exists) while others properties vary from one possible world to another. The former the call essential properties, the latter contingent ones. However it does not mean that criterion of identity of this table is a conjunction of its essential properties for to speak about some possible situation and moreover about possible worlds such a identity should be presupposed. This identity according to Kripke is given by rigid designation that may include an indication of some contingent properties of designated object (that is why a contingent a priori true is admissible) while (some of) its essential properties may be discovered with later scientific researches (that is why a necessary a posteriori true is admissible).
I think however that it is not only the matter of logical technique whether to speak about different states of the same thing or about different possible world where the thing is in different states. For the two approaches have different philosophical background. Roughly speaking we choose here between ancient essencialism and modern transcendentalism. What is common for the two that is a presupposition of identity prior with respect to every possibility. Such an identity includes every possibility into "bundle of possibilities" relating to the same thing. (This presupposition is obviously shared by Kripke.) However concepts of identity, necessity, possibility and contingency are very different in the two cases. What is identity of essence for Aristotle I just mentioned above. But what identity is presupposed when they speak about possible worlds? Dealing with transworld identification problem Kripke believed that notion of possibility in the case of possible worlds abstraction does not presuppose identity except the identity he described and that was similar to Aristotelian essence. I believe that it is not the case. For notion of possible worlds presuppose an extra-world identity that is Leibniz’s transcendent divine subject or Kant’s transcendental human subject. This is the identical transcendent(al) subject who "observes" (constructs) possible worlds. I do not think that it is right to use the notion of possible worlds purely technically ignoring this metaphysical background. Ignoring this context we rather find ourselves under its power than slip it out. The difference between essencialistic and transcendentalistic understanding of prior identity is of great logical importance. Nominally in both cases the prior identity is called "subject" but in different cases the term means different things. Essencialistic (Aristotelian) subject is an identity that remains the same while its contingent properties vary. (Essential properties also remain the same but it is rather a symptom than a foundation of identity. It exactly coincides with Kripke’s point of view.) Transcendentalistic (Leibnizian or Kantian) subject remains the same while all the world as a whole vary, or, more accurate, while different possible worlds change each other. Essencialistic subject is a "bed" of all its properties, while transcendentalistic subject is a person constructing possible worlds. In transcendentalistic perspective all the properties become in a sense contingent (with respect to transcendental subject) and Aristotelian notions of essence and of necessary property loose their sense. Every possible world appears to be contingent as a whole while relations of its elements are necessary. It can be clearly seen in Newton’s mechanic which was developed in transcendentalistic perspective (cf. the notion of "external observer"): world trajectories of particles are contingent in a sense that there is no reason for every particle to have a world trajectory it actually has than some other (that is modern giving the causa finalis up); the same time every world trajectory is completely determined so that the past with necessity causes the future. Paradoxically, the notion of contingency plays much more important role in deterministic Newton’s physic than in Aristotle’s physic that admits contingency in the world. For in Newton’s physic the world always remains fundamentally contingent while for Aristotle contingency is only a sort of veil covering the essence, some inevitable illusion that hides the truth. The concepts of necessity are also different in both cases. Essencialistic necessity is not differed from contingency distinctively. There is no strict boundary between the two (2). Necessary was what one managed to prove (of course it does not resolve the problem of false proof). Nobody knows what may be proved in the future. There are "verisimilar" things that take place "in majority of cases" and there is a special kind of verisimilar proof (so called dialectical syllogism) for such things (An.Post. 93a15, Top.100b18-30, 157a20-158a25). I guess that Nietzsche’s provocative idea that one thing may be more necessary than another [2] is much closer to Aristotle than to Kant. For Kant puts necessity and contingency in different domains - transcendental and empirical correspondingly. To confuse the two is a mortal sin for transcendentalist. While to confuse an essential feature with a contingent one is a particular error to confuse a transcendental and empirical spheres is a principal misunderstanding. Kripke’s anti-Kantian concept of a posteriori necessity (correlative to that of a priori contingency) seems to be very Aristotelian. Descriptions based on contingent features, conjectures and "established facts" all belong to the same sphere of scientific construction. To distinguish between them is a matter of scientists rather than of philosophers.
The problem of crossworld identification arises when transcendentalism is revised. For transcendentalism it is empirical content that fill transcendentally conditioned possibilities in to make things actual. In the frameworks of such an approach they speak not about different possible states of one and the same thing but about different possibilities constructed by the same transcendental subject that may be fulfilled in experience. In this case there is no need to look for some other crossworld identity except that of transcendental subject. Renouncing a transcendental subject but saving possible worlds we appear in a problematic situation. Are there other solutions except a return to Aristotelian essencialism?
A solution I would like to suggest here is inspired by Deleuze’s Difference et Repetition [3] and Logique de Sense [4]. Let as consider a bundle of possible worlds without identical extra-world subject. Let us also give up an idea that any identity should fasten a number of possibilities together. Possible worlds without presupposed identity make what Deleuze calls series . World series is not a variation of the same world while its sameness is not presupposed. That is why, according to Deleuze’s logic, it is rather repetition than variation. However it is not a repetition of the same world because such a repetition presupposes this "same" as a prior identity. (It is the crucial point of Deleuze’s argumentation.) It is rather a repetition of repetition (or "dressed repetition" as Deleuze calls it) that coincides with the most radical ("nude") difference because it means that the world changes such a way that no identity avoids changes, whether it is identity of transcendent(al) subject or identity of essence. Such a change Deleuze calls an event that is a moment of transformation of identities one into another, i.e. their "births and deaths". As a moment of transformation of all the world an event unites the world but unlike the cases of essencialistic and transcendentalistic doctrines such an "eventual" unification presupposes no prior identity.
This reasoning sounds very metaphysically but I believe that it can help us to develop Kripke’s theory in a non-Aristotelian way analytically. Note first of all that the notion of event actually appears in Kripke’s discourse. I mean his initial baptism that is a primary fixation of a referent by name and/or some kind of description (that may be based on contingent features of the referent). Putnam who developed in [5] a theory of meaning that was very close to Kripke’s directly called a corresponding notion "an introducing event". Although Kripke uses the word "baptism" in quotation marks I believe that his reference to Christianity here is essential. For Kripke after Christian theology (which strongly influenced transcendentalism) considers an event as an unique act of foundation - namely a foundation of identity of meaning. That is true that Kripke tries to consider an event rather socially than mystically, but nevertheless it remains "initial" and I believe mystical. Of course a nominalistic idea of linguistic convention is an idealisation, but this idealisation is basically confirmed with real everyday social practice of making agreements (and also is modelled with scientific terminological practice). "Initial baptism" seems to respond nothing but ceremonial practice (3). I think that it is true that ceremonial practice actually also belongs to our everyday experience but to demonstrate this fact it is not enough to put baptism in quotation marks. We have to elaborate notions of ceremony and of event more accurately.
A ceremony is essentially a repetition. It may be considered as a repetition of the foundational event. Kripke’s "initial baptism" seems to be such sort of event that founds use of a certain word as a ceremony. However taking baptism more seriously we should admit that its foundational status is relative. For such a particular baptism refers to certain "original" baptism that appears to be doubly foundational: accordingly the Christian tradition it is a baptism carried out by John the Baptist. Together with critics of foundationalism I claim that a ceremony actually has no unique foundation and it is rather Deleuzean "repetition of repetition". However I do claim not that Kripke’s "initial baptism" never occurs nor that foundation in general is an illusion (nor that it is a bad illusion). I claim on the contrary that every event in general and every speech act in particular, combines these two aspects: the aspect of reference to foundational event in the past and the aspect of creation of foundation for the future. Both aspects are in different senses foundational. We should distinguish not between "foundational" and "caused" use of words but between its two aspects: the first relating to the past and the second relating to the future.
Let me make the same claim in other terms. They distinguish between history and prehistory. And they usually suppose that prehistory historically precedes history that is obviously inconsistent. Jaspers who very good showed irreductibility of history and prehistory to each other [6] did the same thing. I think that we should change our understanding of both history and prehistory to consider every event as historical and prehistoric the same time. Every event is historical in the sense that it refers to its past prehistory as to its foundation and it is the same time prehistoric in the sense that it itself prepares a foundation for future events.
Let us to specify some notions. I think that the notion of possible worlds is hardly relevant to the case of absence of transcendent(al) subject. For in this case strictly speaking we can not use a notion of possibility nor can we use a notion of world. For as Kripke argues - and I am completely agree with him at this point - to speak about possibilities we should presuppose some identity . And if not to restore essences we can not speak about possibility here. A notion of world I believe is not in the case relevant either for it is only a extra-world subject who can "to complete" a world as a whole. Thus without such a subject we can rather speak about virtual medium than possible worlds. What is important is that events taking place in the medium correlate not only with existent but also with non-existent things. For an event is a transition from existent to non-existent and vice versa ("births and deaths"). It is an event in virtual medium that accordingly Kripke’s definition of rigid designation is identical to remain the same in different possible worlds independently of the fact whether it corresponds to existent or to non-existent thing.
Let me finally add one argument that does not directly relate to the subject of Naming and Necessity. I guess that in our contemporary situation it is not enough to be able to deal with an organisation caused by our prehistory. We must be able to deal with events themselves.
Endnotes:
For justification of my interpretation of Posterior Analytic and further details see my "Euclid’s Elements in the Context of Plato’s and Aristotle’s Philosophy" (Moscow 1995 thesis for PhD diploma, in Russian)
More accurately speaking, Aristotelian necessity remains absolutely different from contingency, while contingency is relative and may by "more or less contingent" up to being "almost necessary". The model for Aristotelian necessity-contingency opposition is Plato’s Form-Copy scheme. Copy can be arbitrary close to its Form but Form remains absolutely different from any Copy.
It seems to be interesting to compare the notion of ceremony with its aspect of "initiation" and "foundational event" with Humean notion of habit. Note, that a notion of ceremony is also a matter of psychoanalysis.
Literature:
1. S. Kripke Naming and Necessity, Oxford 1980
2. F. Nietzsche Also Sprach Zarathustra, Kehl 1994
3.G. Deleuze Difference et Repetition, Paris 1968
4. G. Deleuze Logique de Sense, Paris 1969
5. H. Putnam The Meaning of Meaning. — In: H. Putnam Mind, Language and Reality,
Cambridge Mass., 1975.
6. K.Jaspers Vom Ursprung und Ziel der Geschichte Zuerich 1949