Wittgenstein on being and time

JAAKKO HINTIKKA Boston Univerc,ity

1. Memory-time vs. information-time

TIIIS PAPER is an attempt to understand Wittgenstein’s conception of time during the different periods of his activity. This concept offers an interesting perspective on Wittgenstein’s philosophy and played a crucial role in the development of Wittgenstein’s philosophy. In particular, Wittgenstein’s ideas about time were instrumental in shaping his changing concept of object. I will also try to put Wittgenstein’s ideas about time in a wider (and deeper) theoretical perspective. First, what was Wittgenstein’s conception time at the different stages of his philosophical work? Well, what are the different possible answers to this question according to Wittgenstein himself! The main options are listed by him in his lectures, as recorded by G.E. Moore: Later on he made a distiction, as to the meaning of which I am not clear, between what he called ‘memory-time’ and what he called ‘information-time’, saying that in the former- there is only earlier and later, not past and future, and that it has sense to say that I remember that, which in ‘information-time’ i s future. (Moore 1959, p.319) But later on Wittgenstein introduced again the phrase ‘memory-time’, saying that a certain order of events might be so called, and then going on to say that all these events ‘approach a point such that it will have no sense to say “B occured after the present in memory-time”’; that ‘now’ ‘should be a point in an order’. (0”cit. p.322)

[...I when we say ‘The clock is striking now’, ‘now’ means ‘the present in memorytime’, and cannot mean, e.g., ‘at 6.7’ because it has sense to say ‘it is 6.7 now’. (QJ.cit. p. 321 ) What precisely is involved in Wittgenstein’s distinction between memory-time and information-time? His remarks are brief, but they are full enough to be put into a systematical and historical perspective. The prima facie easier of the two terms of the distinction to understand is information-time. Clearly, using information-time

4

JAAKKO HINTIKKA

means signifying time-references by reference to a public chronology, implemented by clocks, calendars, and other such time-keeping paraphernalia. In contrast, it is harder to see what precisely is meant by Wittgenstein by ‘memory-time’. There seems to be at least two different ideas involved in it. Memory is relative to the moment of time at which the remembering takes place. Hence memory-time involves a reference to the remember’s ‘now’. Indeed, if this were the whole story, Wittgenstein’s memory-time would simply amount to the system of time-references which are specified from the vantage point of some subject’s present moment. Statements using such time-referelices entail explicitely or implicitely a reference to ‘now’, i.e. to the moment at which the statement is made. But something else is involved in Wittgenstein’s notion of memorytime than a now-centered system of time-references. Wittgenstein obviously thinks that pure memory-time cannot rely on any external physical criteria for time-reference or time-measurement. All that it can rely on is the totality of one’s present memories and expectations. Such memories and expectations do not give me any way of measuring time-spans. This is the explanation why Wittgenstein says that in memory-time there is only earlier and later. The reliance of memory-time on memory serves to relate it to Wittgenstein’s notion of solipsism. In fact, Wittgenstein seem to have thought that an exclusive reliance on memory-time leads to what might be called temporal solipsism. Striking confirmation of this connection is provided by Wittgenstein himself in his lectures, as recounted by G.E. Moore: As regards Solipsism and Idealism he said that he himself had been often tempted to say ‘All that is real is the exprience ? / t h e present moment’ [emphasis added] or ‘All that is certain is the experience of’tlze present moment’ [emphasis added]; and that anyone who is tempted to hold Idealism or Solipsism knows the temptation t o say ‘The only reality is my present experince’. (Moore 1959, p.311.)

The puzzling thing about Wittgenstein’s contrast is that it admits of two apparently different generalizations. The contrast between nowcentered and chronology-based methods of time-reference (time-identification) can be argued to instantiate a contrast between perspectival and public modes of identification. I have explained this contrast on a number of occasions and will do so briefly below.

WITTGENSTEIN ON BEING A N D T I M E

5

But Wittgenstein’s contrast can also be considered as a special case of another distinction. The contrast between memory-time and information-time is in effect assimilated by Wittgenstein to the broader dichotomy between phenomenological and physicalistic frameworks, called by Wittgenstein ‘systems’. There is in fact ample evidence showing that for Wittgenstein time-references involving ‘now’ are on a par with ostensive references to objects, that is, references mediated by ‘this’ and ‘that’ and perhaps even ‘1’. Now (sic) it is unmistakable that for Wittgenstein such ostensive reference was a paradigm case of reference to phenomenological objects, called by him ‘aspects’ (Aspekte). In fact, in the Brown Book (pp. 171-172) Wittgenstein describes what clearly is his rejection of independent phenomenological discourse as follows: The danger of delusion which we are in becomes most clear if we propose to ourselves to give the aspects ‘this’ and ‘that’ names, say A and B.

But what is the delusion he is speaking of? Wittgenstein’s preceding remarks provide an answer. The delusion in question is related to time and change. It arises in context in which we are considering ‘seeing as’ and aspect change. I might for instence ask the question: When I said to myself ‘What at one time appears to me like this, at another ...’, did I recognize the two aspects, this and that, as the same [ones] which I got on previous occasions?

Hence what is wrong in calling ‘this’ and ‘that’ names is that they could only be names of phenomenological objects, i.e. ‘aspects’, and for such phenomenological objects there are no criteria of identity through time except memory-which by this time Wittgenstein had rejected as the ultimate arbitrator of identity in time. Wittgenstein’s argument against phenomenological objects is hence based on a rejection of self-sufficent memory-time. The two contrasts I have mentioned, viz. the contrasts between perspectival and public identification as well as between phenomenological and physicalistic systems, are not unrelated. Their relation to each other (in Wittgenstein and elsewhere) will be one of the main systematic problems to which Wittgenstein’s theory of time leads us. There are other important ramifications of Wittgenstein’s contrast. Perhaps the most important one concerns the role of memory

6

JAAKKO HINTIKKA

in constituting time in the sense of memory-time and hence in constituting the external world. This point can be highlighted by assuming, for the sake of explanation, that one of the two concepts is alone fundamental. For instance, if memory-time were our only basic concept of time, past and future would have to be constructed logically from the contents of our present experience. Needless to say, memory will then play a crucial role in the construction of our concept of time and thereby our entire world. In contrast, the sole primacy of information-time would imply that memory is not conceptually speaking the rock bottom of our concept of time. On the contrary, the testimony of memory has to be tested in the last analysis against such public evidence as is involved in information-time, for instance in the regularities in natural events, calendars, clocks, etc. The most important aspect of the conceptual construction of the world is the constitution of the basic objects of our ontology. The specification of such objects involves crucially the specification on their identity criteria. Especially important is the constitution of the individuals (basic particulars) of one’s universe. One of the main roles of individuals is to serve as the values of our quantifiers. The contrast between the two concepts of the time implies a similar contrast between two kinds of individuals. The difference between the two kinds lies largely in their identity conditions. If infonnation-time were our sole basic notion of time, our basic individuals would have to be objects persisting in physical time. For if their identification depends on what happens at other moments of time than the specious present, for instance, if the identity criteria of our objects refer to continuity in time or on temporally persisting ‘essential’ attributes of objects, our objects must be temporally persisting in order to have their identity tested by such criteria. Conversely, if memory-time is our sole basis of our temporal discourse, the basic objects must be what from the reverse perspective would be referred to as time-slices of persisting objects, perhaps something like Quinean rabbit-stages as distinguished from rabbits. This will not make such objects any less objective than individuals (and entities of other logical types) than persist in time. The difference is a conceptual one. The difference lies in a different relation of S i n (as codified in our quantifiers) to Zeit. For this reason, my

WIT T GE NS T E IN O N B E I N G A N D T I M E

7

study of Wittgenstein’s idea of time becomes in effect a study also of his ontology. Hence the word ‘being’ in my title, over and above the word ‘time’. 2. Primary problems

But what did Wittgenstein think about the relative primacy of the two concepts of time? Here the timing of Wittgenstein’s sundry utterances and ideas enters importantly into the discussion of his philosophy of temporality. At first sight, the two conceptions of time seem to enjoy happy coexistence. At the time of the quoted utterances, Wittgenstein gives the appearence of thinking of memorytime and the information-time as two separate ways of using temporal concepts. Wittgenstein’s conciliatory attitude in 1932-35 to the contrast between memory-time and information-time nevertheless should not hide the massive fact that in his middle and later periods there was one and only basic conceptualization of time. The primacy of this concept was consistently manifested by Wittgenstein ever since he in 1929 abruptly rejected phenomenological (and hence in a sense solipsistic) languages as ‘absurd’ in favour for everyday physicalistic languages as the sole legitimate ones. The explanation of the apparent coexistence on an equal footing is that the different ‘notations’ of the Blue Book period are but different dialects (as it were) of our basic physicalistic language (or languages). 3. The notion of time in the “Tractatus”

All this pertains in the first place to Wittgenstein’s views in 1930-36 and not to his views on time at the time of Tractatus. What were they? Here one of the quotes above offers a clue. When Wittgenstein speaks of a temptation to think or to say this or that, he invariably expresses something he had himself been tempted to say. And, quite often, he is speaking of a temptation to which he had in fact fallen. Hence the obvious null hypothesis here is that in the Tractatus Wittgenstein represented a kind of solipsism of the present moment. Of course, he could not officially say that. But then he did say in the Tractatus that

8

JAAKKO HINTIKKA

What solipsism means is quite correct, but it cannot be said.

Would it really be surprising at all if we could expand this nonstatement and put into Wittgenstein’s mouth the words ‘What solipsism of the present moment means is quite correct’? I will argue that we can. One way of capturing that sense is just to say that in the Tractatus Wittgenstein was maintaining the primacy of memory-time. A great deal of prima facie evidence can be found for this attribution. Such evidence might nevertheless be found inconclusive. There is a different line of thought, however, that I consider absolutely convincing. Since this argument is of a kind not usually employed by philosophers, a few comments in its methodological status are in order. This argument is based on an overall interpretation of a philosopher’s line of thought, sometimes a line of thought in an entire work. If an assumption is absolutely indispensable for the success of this line of thought, we can safely attribute it to the philosopher in question even on the basis of relatively weak collateral evidence. For otherwise the entire line of inquiry does not even make sense. The kind of interpretation strategy I have described is admittedly quite tricky in some other respects, mostly because it presupposes a firm grasp of a philosopher’s overall position. Fortunately, in the case of the Tractatus, it is not difficult to see what the central vision of this book is. It is Wittgenstein’s idea of all meaning based on pictorial correspondence, and nothing else. Elsewhere, I have shown how totally single-minded Wittgenstein is in pursuing this vision. What matters here is that this picture idea was taken by Wittgenstein to be the whole story of the semantics of our language (any correct language), so to speak not only truth and nothing but the truth, but the whole truth. This completeness of the picture ‘theory’ implies that according to the Tractatus a propositional ‘picture’ has to be capable of being compared with reality directly, without the mediation of any further activity or ‘calculus’. For the rules of those activities would not only be over and above the logic of pictorial representation, but more basic than it. The logic of picturing could not take care of itself (cf. Notebooks 1914-16, p.1); there would have to be logical constants in our language guiding the language-world comparisons (cf. 5.4); the rules for comparing propositions with states of affairs would constitute a

W I T T G E N S T E I N ON B E I N G A N D T I M E

9

logic of facts that would have to be represented in language (cf. 4.0321); and so on. All these inevitable concomitants of any comparison process are firmly rejected by Wittgenstein. Instead he says that Reality is comparcd with propositions. (4.05)

Indeed, this is all Wittgenstein ever says about the actual links between reality and propositions, apart from the picture theory. He might as well have added: ‘Period, end of the story.’ But u proposition can be immediately compared only with present facts. ‘Our propositions are verified by the present’ (Philosophical Rrmurks sec. 48). The only reality that can be at the receiving end of Tractarian comparisons with language is the reality of the timeless now. Otherwise we would need such temporally extended comparison processes which the completeness of the picture view was seen to exclude. And the idea of momentary character of languageworld comparisons obviously is precisely the primacy of memorytime. In brief, Wittgenstein’s so-called picture theory of language presupposes the primacy of memory-time. Our language means what it means only by saying something about what is now the case. 4. Wittgenstein ’s change of mind

We have thus seen that in the Tractatus Wittgenstein represented the primacy of memory-time but that in his later philosophy he defended the exclusive primacy of information-time. One thing that remains is to examine how Wittgenstein came to change his position concerning the relative priority of the two concepts of time. Earlier in this paper, I argued that the completeness of the picture theory implied the primacy of memory-time. In Wittgenstein’s mind-and on the pages of his notebooks from the year 1929-an argument for the opposite conclusion was slowly and painfully articulated. It is known that Wittgenstein gave up the completeness of truth-functional logic sometime in 1928-29. Very little argument is needed to show this, for in Wittgenstein’s little 1929 paper ‘Some Remarks of Logical Form’ he makes this point in virtually so many words. At the same time, Wittgeristein still insists very strongly on the need of direct comparisons between language and reality, as witnessed by many eloquent passages. But arbitrary propositions are no longer

10

J A A K K C) I4 1N T I K K A

pictures that can be confronted with reality directly, according to the Wittgenstein of the first half of to 1929. Hence they can be compared with reality only indirectly, through the mediation of actual calculus-like operations (activities). These calculations must take place in physical time. The correctness of calculation is predicated on the persistence of linguistic symbols in physical time and on the calculator’s ability to recognize the identity of these physical objects (markers) throughout time. The criteria of such identity can only be physicalistic, not phenomenological. As a consequence, the sum of total of such computational aids, our actual working language, must be physical. As Wittgenstein put it in the Philosophical Remarks, language belogs to the second (1.e. physicalistic) system. But if so, language-world comparisons are put to a new light. Not only do they rely on antecedent operations on the symbol combinations of our language. The objects operated on, the sentences and other expressions of our language, belong to the physical world. A fortiori, whenever an expression is compared with the facts, with the world, this expression is a physical object or physical fact. Moreover, the comparison between the world and the language takes place at the moment of time different from the one in which the statement was originally made, because the intervening calculations inevitably take some (physical) time. They are not made within a specious present. But you cannot compare a physical fact with a phenomenological one. Y o u cannot put a physicalistic object like a proposition ‘on the top of a phenomenological state of affairs. The two are plainly and simply incommensurable. Immediate comparisons are only possible between propositions and physical facts. But if so, the aspect of reality that our language primarily speaks of must be the physical world. It cannot be a phenomenologically construed world. Hence a phenomenological language is, Wittgenstein asserted on October 22, 1929, ‘absurd’ (MS 107, p. 176). This result of Wittgenstein’s has to be understood correctly. It is not a thesis about the world, about the objects that exist or do not exist in the world. It is about what language must be like in order to be able to say something about the world. It is a thesis of what the objects are like that are referred to directly by the symbols of our language. As Wittgenstein puts it in his Cambridge lectures (in Lee, editor, 1980, p. 82),

WITTGENSTEIN ON BEING AND TIME

11

The world we live in is the world of sense-data, but the world we talk about is the world of physical objects.

Ergo, the time we live in is memory-time, but the only time we can directly speak of in our language is physical time, that is information-time. Thus Wittgenstein’s change of mind had clear-cut implications for his conception of time. It meant a total victory of the physical time. The moment of time which I call the present and which contains everything given to me belongs itself to physical time. (TS 215, p.494)

5. Perspectival vs. public identification

What more is there to be said of Wittgenstein’s treatment of time? A great deal, in fact. As was pointed out above, his distinction between memory-time and information-time can be considered a special case of two different general distinctions. They are the distinction between perspectival vs. public identification and between phenomenological and physicalistic language systems. I have not tried t o keep them strictly apart so far, and what has been seen in fact shows that Wittgenstein did not do so, either. However, the time has come to sharpen our ideas and have a closer look at one of the distinctions, viz. the perspectival vs. public distinction. It is a contrast between two kinds of discourse, involving two different kinds of objects identified. Very briefly, in our actual thinking and speaking we are tacitly using two different cognitive systems. The difference does not lie in a difference between different cognitive attitudes, for instance between two different kinds of knowledge or two kinds of memory. It lies in the way we identify the objects of our perception, knowledge or memory. I have called the two kinds of identification perspectival and public. The contrast is clearest in the case of visual cognition, but it is in operation also on the higher cognitive functions. In visual cognition, one ean identify objeets of visual perception either by reference to the perceiver’s visual space or to an impersonal public frame of reference. Each mode of identification creates its own quantifiers and its own who-, what-, where-, when-questions. Consider, for example, a question like

12

JAAKKO HINTIKKA

Who around here is Alonzo Church? asked by someone who knows a lot of the famous logician and editor but has never met him or even seen pictures of him. Here the questioner knows who Alonzo Church the public person is, but is trying to find a slot for him in his or her visual space, among his o r her ‘visual objects’, as one is tempted to say. A satisfactory (conclusive) answer here is one which completes a perspectival identification. The typical answer is accordingly an ostensive one: ‘That man over there!’ Conversely, someone might point to a massive white-haired gentleman and ask: Who is that man? Here the questioner is taking a ‘visual object’ well entrenched in his or her visual space and trying to find a slot for him among his or her public objects. The success of such an answer means successful public identification. An answer might be, ‘Alonzo Church’, but it need not contain a name. It might instead contain e.g. definite description, as long as questioner knows who is referred to by that description. This distinction is one of the most fundamental features of our own actual conceptual system, even though philosophers have paid scant attention to it. It is highly important in cognitive psychology and neuroscience. It is a major tribute to Wittgenstein’s conceptual acumen that he should have recognized the generality of the conceptual contrast between the public and the subject-centered discourse, even if he did not reach a complete clarity in the matter. The most explicit discussion of the distinction in Wittgenstein is toward the end of the Blue Book (pp.61-65). There Wittgenstein discusses two visual criteria of visual location, calling them ‘geometrical’ and ‘physical’. They determine different ‘grammars’. The grammar of the word ‘geometrical eye’ stands in the same relation to the grammar of the ‘physical eye’ as the grammar of the expression ‘the visual sense datum of a tree’ to the grammar of the expression ‘the physical tree’. (The Blue Rook p.64)

Systematically speaking, the contrast between sense-data and physical objects is an archetypal example of the contrast between the perspectivally and publicly identified objects.

WITTGENSTEIN O N BEING A N D T I M E

13

Another symptom of perceptually identified objects is their being identified ostensively. This clue we also find in Wittgenstein, and even find it in important contexts: One has been tempted to say that ‘now’ is the name of an instant of time, and this, of‘ course, would be like saying that ‘here’ is the name of a place, ‘this’ is the name of a thing, and ‘1’ the name of a man. (T/w Brown Book, pp, 108-109, the sequel is also highly suggestive.)

He also recognized that what he had himself done earlier is to yield to that temptation in the case of ‘this’ and ‘that’. The danger of delusion which we are in becomes most clear if we propose t o give the aspects ‘this’ and ‘that’ names, say A and B. ([bid. pp. 171 -172.)

The result to Wittgenstein’s earlier yielding to that temptation was the ontology of the Tractatus. In it, as in Russell, ‘this’ and ‘that’ in a sense were the only ‘logically proper names’ . Wittgenstein thus reached a major conceptual insight. Unfortunately the unsystematic character of his philosophizing and his avoidance of logical tools prevented him from giving a really sharp formulation to his insights and from using it effectively for the clarification of philosophical problems. 6. From the priority problem to the integration problem

This generalization puts in a new light the question of the relative priority of memory-time and information-time. Wittgenstein seems to have thought during his middle and late periods that the perspectivel mode of identification is as it were but one dialect among our many physicalistic languages. Even if this is correct, it does not answer the question concerning the precise relationship between the two methods of identification. In fact, prima facie it is now the perspectival mode of identification that has the strongest claim to primacy. Such identification is in an obvious sense local. In the case of visual cognition, it relies on one’s visual perspective, in the case of memory, on one’s personal memories of persons, objects, places, events, and so on. The problem about the relationship of the two modes of identification is how these local frameworks can be integrated into one common public framework-if they can be so integrated. A wealth of problems, concepts and ideas both

14

J A A K K O HINTIKKA

in the history of philosophy and in recent philosophy relate to this problem. One of the key questions that arises here concerns the relation of the two frameworks (systems) to our ordinary language. It is not difficult to develop a formal (but interpreted) notation to express either kind of identification, even though the perspectival notation would have to be relative to a particular person in a particular situation. But what is the relationship of the two kinds of frameworks to ordinary language and to its semantics? Here Wittgenstein reached an important insight. Wittgenstein realized that the ‘grammar’ we use in describing our immediate experiences is different from the ‘grammar’ (semantics) of ordinary discourse. The world we live in may in some sense be a world of phenomenological objects, but the world we speak of in our ordinary language is indeed the world of physical objects. Speaking more generally, he in effect realized, even though he expressed his point in a differnt terminology, that almost all of the semantics of our language relies in its operation on the public mode of identification. This is the real moral of Wittgenstein’s story about the primacy of physicalistic languages. Here Wittgenstein is crossing swords not only with the author of the Tractatus, but with the mainstream analytic philosophy of language in the twentieth century. In this tradition, thinker after a thinker from Russell and Husserl to Kripke has conceived of ultimate explanations of meaning as ‘reductions to acquaintance’ or more generally as specifications of the perspectival objects our names and other symbols stand for. For instance, Russell seems to have assumed almost as a kneejerk action that the relevant values of quantifiers are objects of acquintance. A paradigm of name-giving according to such ‘Logical Adamism’ is the biblical story of animals parading before Adam to be named (Genesis I:2, 19-20). Likewise, a Kripkean dubbing can in the first place reach only to a perceptually accessible, ostended object, in brief, to a perceptival objeet. Against such formidable adversaries, the later Wittgenstein maintains that the normal references of our words are public (physicalistic) rather than perspectival (phenomenological) objects. I believe that Wittgenstein is essentially and importantly right here, even though I cannot present much of real evidence in this paper. In spite of Wittgenstein’s brilliant insight into the tie between the

WITTGENSTEIN O N BEING AND T I M E

15

semantics of our language and the public method of identification, his views are in certain other respects far from satisfactory. Wittgenstein was in effect considering the contrast between memory-time and information-time as a special case of what for us are two different distinctions which he tacitly assimilated to each other. But Wittgenstein identifies the same distinction with a contrast between phenomenological and physicalistic ‘systems’. For instance, his change of mind i n October 1929 meant in effect a victory for the public system, but Wittgenstein formulated it at the time by speaking of a rejection of phenomenological languages. This assimilation of the two distinctions to each other has several consequenses for Wittgenstein. First, he treated the distinction as a contrast between two overall languages or ‘systems’, the phenomenological and physical one. This does not make much sense in terms of the perspectival vs. physical distinction or, rather, it makes sense only logically. In reality, there are innumerable perspectival systems of identification, and the real problem is not the contrast between any one of them and the public system. Rather, the basic problem is the integration of all the local perspectival systems into a single public system. This might not seem to be far removed from Wittgenstein’s actual views. As was seen above, during his middle period Wittgenstein considered phenomenological (solipsistic) discourse as a bunch of subsystems of an overall physicalistic language. But even though Wittgenstein was on the right track here, it does not close the issue. For one thing, Wittgenstein still insists that local systems are parasitic on an overall public system. There are no purely phenomenological languages in their own right, independent of physicalistic languages, not even locally, Wittgenstein does not spell out his reasons here, but one can discern at least two different reasons here. First, from the fact that our actual ordinary language is geared the public system of identification it might seem to follow that phenomenological systems cannot be independent even locally. This seem< to be a mistake. In our ordinary discourse, we can use a perspectival system perfectly well locally and express it linguistically, independent of the public system. Much of our descriptive vocabulary is neutral with respect to the perspectival vs. public

16

JAAKKO HINTIKKA

distinction. Some others, like quantifiers, we can simply relativize to locally accessible objects. One might even try to turn the tables on Wittgenstein. He insisted at one time very strongly that ‘our propositions are verified only by the present’. (TS 213, p. 496.) This is a special case of a more general feature of the semantics of our language, to wit, that the only language-world interfaces are the local perspectival systems. Even in the games of seeking and finding that are the ‘logical home’ of quantifiers, the end point of a search is conceptually like pointing to an actually present object and saying, ‘Here is one!’ One might argue that perspectival ‘dialects’can not only be independent of the overarching public language, but even epistemologically prior to it. Another way in which Wittgenstein seems to have been led to his position is so to speak take seriously the idea of perspectival languages as phenomenological languages. Of these two ideas, the idea of perspectival language is conceptually the clear one, whereas the notion of phenomenological language is wrought with difficulties. It is identified by Wittgenstein with a language employing merely internal criteria. For instance, phenomenological time would be a concept of time relying on one’s momentary memories, expectations, and suchlike. That this was indeed Wittgenstein’s view is reflected by his very term ‘memory-time’. In contrast, there is nothing whatsoever in the idea of perspectival time-reference that restricts it to the use of memory in determing time-references. There is no reason why the perspectival ‘local’ systems that have to be combined into one big overall system like jigsaw puzzle pieces must be momentary epistemic states of memory and other knowledge. Conceptually, they can for instance be all my different episodic (perspectival) memories. This depersonalization of memory-time will also have the consequence of relating perspectival (memory) time to space. In that question of information-time cannot all be answered without bringing in questions of space. This fact is most pronounced in the theory of relativity, but it comes much closer to home. In ancient Greece, different city-states might have different calendars, and when I say that I telephoned Moscow yesterday, my yesterday might be my Russian colleague’s today. In general, the integration of perspectival frameworks into a single public one involves the integration of different local (memory)times in one absolute time.

WITTGENSTEIN ON REING A N D TIME

17

I am not suggesting that Wittgenstein pursued this line in his own thought himself. Fairly obviously he did not follow it very far even if something like it had been in his mind. But as a thought-experiment, this line of thought is highly relevant to our understanding and evaluation of the conceptual background of Wittgenstein’s philosophy, As was pointed out above, in the idea of perspectival identification there is no restriction to identification by memory. If a perspectival identifier uses all the available means of identification, including physical ones like clocks and diaries, and if such an identifier is omniscient, doesn’t his or her perspectival system coincide with the physical one? No, because there still is a restriction to what is observable to one particular observer tracing one particular world line through space-time. The problem of integrating different perspectival worlds into one common public world will then become Einstein’s question in his special theory of relativity of whether local times can be combined into one common absolute time. This tacit relationship between Wittgenstein’s and Einstein’s ideas throws interesting light on the conceptual background of Wittgenstein’s philosophy. For one thing, we know that local times cannot be combined into one absolute time. The analogy I have pointed out means that we cannot expect that in general there is a unique public system to serve as the reality our language is about. And this in turn suggests that the public system cannot be the primary one epistemologically,even if it may be the basic one in the semantics of our language. Hence the entire problem complex Wittgenstein is dealing with is in need of further examination. What we can say even before the results of such a study are in is that Wittgenstein deserves an immense credit for calling our attention to that important group of questions. References AMBROSE,Alice (ed.): Wittgenstein ‘s Lectures, Cumbridge, 1932-1935, Basil Blackwell, Oxford, 1979. BRAND,Gred (ed.): Die grundlegenden Texte lion Ludwig Wittgenstein, Suhrkamp, Frankfurt am Main. 1975. HINTIKKA, Jaakko: “‘Die Wende der Philosophic’: Wittgenstein’s New Logic of 1928”, in the Proceedings of the TwCftk Internationul Wittgenstein

18

JAAKKO HINTIKKA

Symposium, pp. 380-396, Holder-Pichler-Tepsky, Wien, 1988. HINTIKKA, Jaakko: “The Cartesian cogito, Epistemic Logic and Neuroscience” and Merrill B. HINTIKKA, The Logic of Epistemolin Jaakko HINTIKKA ogy und the Epistemology of Logic, pp. 113-146, Kluwer Academic, Dordrecht, 1989. HINTIKKA, Jaakko: “Ludwig’s Apple Tree”, in F. STADLER (ed.), Scientific Philosophy: Origins and Developments, pp. 27-46, Kluwer Academic, Dordrecht, 1993. HINTIKKA, Jaakko: “An Anatomy of Wittgenstein’s Picture Theory”, in Robert Cohen and Carol Could (eds.), Artqucts, Representations and Sociul Pruct i c r , pp. 223-256, Kluwer Academic, Dordrecht, 1994. HINTIKKA, Merrill B. and Jaakko HINTIKKA: Investigating Wittgenstein, Basil Blackwell, 1986. KAILA,Eino: Reulity und Experience, D. Reidel, Dordrecht, 1978. LEE, Desmond (ed.): Wittgenstein ’s Lectures, Cambridge 1930-1932, Basil Blackwell, 1980. MOORE,G.E.: “Wittgenstein’s Lectures in 1930-33”, in G.E. Moore, Philosophical Pupers, Georg Allen and Unwin, London, 1959. RUSSLLL,Bertrand: The Problems of Philosophy, Oxford University Press, 1959. (Original in 1912.) R L J S S ~ LBertrand: L “Knowledge by Acquaintance and Knowledge by Description”, in Mysticism and Logic, pp. 209-232, Longmans, Green and Co., London, 1918. RIISSIIL.L, Bertrand: Our Knowdedge oJ the E:xternal World, Open Court, La Salk, 1914. Revised ed., Ceorg Allen and Unwin, London, 1926. TU f - viN C;, Endel: Elements ojEpisodic Memory, Oxford University Press, 1983.

Acknowledgment A longer version of this paper has appeared in my book, Ludwig Wittgenstein: H d f i Truths and One-and-a-HalflTrutlis. Kluwer Academic Publishers, Dordrecht, 1996, pp. 241-244. It is reprinted here by the kind permission of the publisher.