***************************************************************** * * * File: 15-1-95.TXT Dateilänge: 47 KB * * * * Autor: Michael Kober * * Freiburg, Br. (at present Berkeley) * * * * Titel: Wittgenstein in transition, or how to make use of * * the 'Nachlaß' - A review of David G. Stern: * * WITTGENSTEIN ON MIND AND LANGUAGE * * * * Erschienen in: WITTGENSTEIN STUDIES, Diskette 1/1995 * * * ***************************************************************** * * * (c) 1995 Deutsche Ludwig Wittgenstein Gesellschaft e.V. * * Alle Rechte vorbehalten / All Rights Reserved * * * * Kein Bestandteil dieser Datei darf ganz oder teilweise * * vervielfältigt, in einem Abfragesystem gespeichert, * * gesendet oder in irgendeine Sprache übersetzt werden in * * irgendeiner Form, sei es auf elektronische, mechanische, * * magnetische, optische, handschriftliche oder andere Art * * und Weise, ohne vorhergehende schriftliche Zustimmung * * der DEUTSCHEN LUDWIG WITTGENSTEIN GESELLSCHAFT e.V. * * Dateien und Auszüge, die der Benutzer für seine privaten * * wissenschaftlichen Zwecke benutzt, sind von dieser * * Regelung ausgenommen. * * * * No part of this file may be reproduced, stored * * in a retrieval system, transmitted or translated into * * any other language in whole or in part, in any form or * * by any means, whether it be in electronical, mechanical, * * magnetic, optical, manual or otherwise, without prior * * written consent of the DEUTSCHE LUDWIG WITTGENSTEIN * * GESELLSCHAFT e.V. Those articles and excerpts from * * articles which the subscriber wishes to use for his own * * private academic purposes are excluded from this * * restrictions. * * * ***************************************************************** * * * Kober, Michael (1995) Wittgenstein in transition, or how to * * make use of the 'Nachlaß' - A Review of David G. Stern: * * WITTGENSTEIN ON MIND AND LANGUAGE; in: * * Wittgenstein Studies 1/95, File: 15-1-95; hrsg. von * * K.-O. Apel, F. Börncke, N. Garver, B. McGuinness, P. Hacker, * * R. Haller, W. Lütterfelds, G. Meggle, C. Nyíri, K. Puhl, * * Th. Rentsch, A. Roser, J.G.F. Rothhaupt, J. Schulte, * * U. Steinvorth, P. Stekeler-Weithofer, W. Vossenkuhl * * (3 1/2'' Diskette) ISSN 0943-5727 * * * ***************************************************************** WITTGENSTEIN IN TRANSITION, OR HOW TO MAKE USE OF THE 'NACHLASS' - A Review of David G. Stern: WITTGENSTEIN ON MIND AND LANGUAGE, Oxford University Press: Oxford, New York 1995. (ISBN 0-19-508000-9) All references made simply by page numbers, e.g. "(p.13)", refer to pages of David Stern's book. Numbers in parentheses between asterisks, e.g."(*13*)", refer to endnotes. Contents: 1. Introduction 2. What can be expected from the hitherto unpublished parts of Wittgenstein's 'Nachlaß'? 3. Wittgenstein in transition - the systematic importance of his thinking around 1930 4. Closing Remarks 1. INTRODUCTION 1.1. David Stern clearly states his aim right at the beginning of his book: "This book is an exposition of Wittgenstein's early conception of the nature of representation and how his later revision and criticism of that work led to a radically different way of looking at mind and language" (p.3). He expresses the hope that "there is much to be learned by looking at the process of revision that leads to the rejection of the leading ideas of the TRACTATUS to the development of Wittgenstein's later philosophy of mind and language" (p.98). Consequently, Stern does not intend to focus merely on Wittgenstein's TRACTATUS, the PHILOSOPHICAL INVESTIGATIONS or other published 'major' writings, for "my reading of his philosophy of mind and language begins from the initial articulation of his thoughts in his first drafts, conversations, and lectures and the process of revision that led to the published works" (p.3). Indeed, Stern extensively quotes notes, paragraphs, or remarks stemming from different parts of Wittgenstein's Nachlaß, from both already published and still unpublished material, in the reasonable hope that he can reconstruct "the inner dialogue that was the driving force in the development of Wittgenstein's philosophical work" (p.5). The reason for doing this is well stated: " [...] while Wittgenstein's posthumously published books do contain much of his thought, they are often much more compressed and opaque than the drafts on which they were based and were written in the conscious expectation that very few readers would understand him" (p.4). As long as this is not supposed to mean that almost everything that appears to be obscure or difficult to understand in the published writings will become clear in the Nachlaß (and, no doubt, Stern does not want to hold that), this is certainly the thing Wittgenstein scholarship has to do in order to advance both a better presentation of Wittgenstein's views and the REASONS which led or even forced him to repeatedly change those views. Despite his notoriously anti- theoretical remarks, which can so easily be trivialized (PI 107- 133), Wittgenstein was aware of the fact that doing philosophy does not consist merely in advancing illuminating pictures or similes. He himself was constantly putting forward arguments. 1.2. "Studying the development of Wittgenstein's work by making use of his published and unpublished typescripts and manuscripts, records of what he said, and the like, in addition to his later and his earlier masterpieces, can make a substantial contribution to our understanding of those works. It can clarify the problems that led him to write those works, and it can illuminate aspects of his thought that have generated enormous controversy in the secondary literature, by making it clear which positions he opposed and what solutions he had to offer" (pp.5-6). Stern is eager to point out that this additional material in the Nachlaß is not of secondary importance, that is: that it should not regarded as preparatory work or rough notes only and therefore lacking the authority of the TRACTATUS or the PHILOSOPHICAL INVESTIGATIONS (p.6). For "[m]uch of the power of Wittgenstein's writing arises out of a struggle between opposing intuitions and his attemps to resolve that struggle" (p.7). I take this to mean that Wittgenstein's views can only be evaluated philosophically if we can reconstruct his arguments against those initially tempting intuitions which ultimately turned out to be untenable. And, according to Stern, this also contributes significantly to our understanding of Wittgenstein's later writings, because the objecting interlocutor in them often puts forward exactly those views which Wittgenstein himself had held earlier (p.7). In this sense, the decisive standards against which Stern's book has to be measured are set up by Stern himself, and I am glad to say that he has met them all. It should be added that Stern focuses on Wittgenstein's thinking around 1930 (with much emphasis on 1929) - a reasonable decision, since the most dramatic and most influential revisions opening the road to the PHILOSOPHICAL INVESTIGATIONS occured at that time. 1.3. I will proceed as follows: In section 2, I will present my views on what can be expected from consulting the hitherto unpublished material of Wittgenstein's Nachlaß - and of what should not be expected from this. In section 3, I will outline the systematic reasons that caused Wittgenstein to abandon his TRACTATUS-account, his attemps to develop new views involving the search for a phenomenological language, and why he dropped them later. I hope that this will make clear that Wittgenstein was moved by arguments and does not merely offer us perspicuous pictures in philosophy. Section 4 consists of a few closing remarks. 2. WHAT CAN BE EXPECTED FROM THE HITHERTO UNPUBLISHED PARTS OF WITTGENSTEIN'S NACHLASS? 2.1. In conversation with some Wittgenstein scholars and from articles in popular magazines (such as DIE ZEIT in Germany), one can hardly avoid the impression that many people suspect that Wittgenstein's Nachlaß contains material of high importance which has until now been kept 'inexcusably' secret by the Wittgenstein trustees. 2.1.1. However, this is not entirely true: Most of the writings constituting the Nachlaß have been available to researchers since the late 1960s, either on microfilm or as bound photocopies. Furthermore, the Wittgenstein Archives at the University of Bergen (Norway) have invited all interested scholars to use their work. The trustees even sent photocopies of manuscripts to interested graduate students upon request. This is, of course, all a bit inconvenient compared to desirable free access to the Nachlaß in the nearest academic library - a situation which hopefully will be attained soon (at least partially) due to the CD-ROM edition of the Nachlaß of the Norwegian Wittgenstein Project (cf.p.5). 2.2. Nevertheless, one question suggests itself: What can actually be expected by having access to hitherto unpublished material in the Wittgenstein Nachlaß? Several options come to my mind (which are, of course, not necessarily disjoint, but may complement one another): 2.2.1. If one is not acquainted with Wittgenstein's unpublished writings, one may hope to discover new philosophical problems, new arguments, or even new solutions - and "new" may be either new as regards Wittgenstein, or even new altogether. In either case, this would surely be a stunning discovery. However, if one recalls that it was Wittgenstein himself who collected those remarks which appeared to him to be the best ones and arranged them into the published typescripts of the PHILOSOPHICAL INVESTIGATIONS (parts I and II), the REMARKS ON THE FOUNDATIONS OF MATHEMATICS (part I), and the REMARKS ON THE PHILOSOPHY OF PSYCHOLOGY (volumes I and II), one may conclude that this is highly unlikely. Given my own familiarity with the Nachlaß - not too extensive, but relatively substantial, I believe - I would be surprised if something entirely new were actually to come up. It is my impression that much of the unpublished manuscripts is already known since its (more important) remarks appear repeatedly in later manuscripts or typescripts which have been made public. It appears to me today that the Wittgenstein trustees have not left any really important material unpublished. There may be some exceptions, it has to be admitted; for example, sections of the so-called 'Big Typescript' are still unreleased (TS 213 in the von Wright-catalogue; cf.p.94). That is to say: in my view, the former trustees - Rhees, Anscombe, and von Wright, themselves pretty good philosophers - have already published the most substantial or interesting remarks over the years (*1*). -- Of course, I hasten to add (in particular, in view of David Stern's investigations), Wittgenstein scholars still need access to all the preserved notes, drafts, writings, etc., and I am glad to hear that the Norwegian Wittgenstein Project is on its way (p.5, fn.3). 2.2.2. One may expect from the Nachlaß some elucidation of arguments, views, or remarks of Wittgenstein's in the published 'major' writings, which hitherto remain obscur. This is highly reasonable, since many remarks (for example, those on solipsism in the PHILOSOPHICAL INVESTIGATIONS, PI 398ff; cf.p.97) contain extremely condensed lines of thought which need to be disentangled. On other occasions, the manuscript sources indicate that Wittgenstein considers some issues not yet settled (like "Ich bin in einem Wirrwarr" ["I am in a muddle"]), or that he himself does not know what to say (*2*). But hopes for clarification from the Nachlaß may easily be disappointed. For instance, my own study of the sources of the remarks constituting the so-called private language argument (appearing in MS 129 and TS 228) has revealed them to be as condensed as they appear in the PHILOSOPHICAL INVESTIGATIONS, though they are arranged in a different order (see section 2.2.4. below)(*3*). 2.2.3. One may hope to reconstruct the evolution of Wittgenstein's thought. This is certainly reasonable, but a very difficult thing to do, since Wittgenstein was often working on several issues entangled in one another at the same time (see section 3.0. below). One may discover sources of problems he was struggling with during the genesis of his own thought (e.g. why he was committed to showing the community dependence on speaking a language), or one can watch him making detours in his philosophical development (e.g. the treatment of a phenomenological language). And one may hope to detect influences of other philosophers on him (especially Frege, Russell, Ramsey, and James), since in the first draft of a remark Wittgenstein sometimes adds names in parentheses which indicate from where he got ideas, problems, questions, or (alleged) confusions (e.g. the manuscirpt source of RPP I 1036 mentions Koehler's name). 2.2.4. One may realize connections or see links within Wittgenstein's philosophy which have remained hitherto unnoticed. There are remarks that occur in different contexts (e.g. compare RPP II 727, talking about psychological concepts, with Z 352, talking about colour concepts), the train of thought may develop in different directions (for example PI 189, which is also part of RFM I 1), or one may see closer connections (for example, PI 201 with the puzzling opener "This was our paradox" appears in MS 129 shortly after the source of PI 293 [the 'beetle in the box'], cf. PI 304). 2.2.5. Finally, one may hope to achieve a better support or confirmation for some readings of Wittgenstein's philosophy, and one may even be successful in disconfirming or even refuting other ones. Notoriously, this hardly happens in a 'knockdown' way, but more often shifts attention to other remarks. And it does not help in cases like 'Kripgenstein,' where the issue at stake is highly interesting independent of the question whether Kripke is offering a correct reading of Wittgenstein's thinking or not (Stern thinks that he is not; cf.p.178). 2.2.6. In the end, one can put it only as a triviality: Having access to and reading remarks or notes in Wittgenstein's Nachlaß does not put aside the need for the interpreter's own thinking. 2.3. Stern obviously aims at options 2.2.2.-2.2.5. (cf. my quotes of his in section 1.2.) - and he has succeeded. To be honest, I was afraid of merely being told the genesis of Wittgenstein's thinking, of what he was writing down when, without being shown a remark's systematic impact or without being provided with any argument. I was afraid of being bombarded with masses of unpublished quotations which merely repeated familiar remarks. I was afraid of being overwhelmed with quotations illustrating anew well-known views. I was afraid of merely being provided with the fact that remark R appears in MS x, MS y, and TS z as well. All that would have left me bored! I am genuinely pleased that David Stern has not succumbed to these temptations which are so near at hand when one is beginning to work on the Nachlaß. He does not merely offer philological facts. He succeeds in elaborating reasons why Wittgenstein so often changed so many of his views. Whereever possible, he quotes from already published writings in order to substantiate his exposition. The hitherto unpublished remarks he quotes are well chosen and to the point (they are in English in the main part of the book and in the original German in an appendix). David Stern's contribution is therefore of high value not only to Wittgenstein scholarship, but to the philosophical community in general. 3. WITTGENSTEIN IN TRANSITION - THE SYSTEMATIC IMPORTANCE OF HIS THINKING AROUND 1930 3.0. With the help of Stern's book, I am now going to reconstruct the main line of reasoning which caused Wittgenstein to give up his TRACTATUS philosophy (*4*), to suggest some intermediary positions in 1929 and 1930, and to change them again so that he could finally develop his later views in the PHILOSOPHICAL INVESTIGATIONS and the like. I am neglecting therefore many other issues Stern mentions (such as the 'metaphysical subject,' Wittgenstein's use of the simile of the magic lantern, his acceptance and later rejection of the idea that 'all is in flux,' a discussion of Kripke's reading of Wittgenstein, and others), and I am using the freedom of writing a review to remain unconcerned with having to justify my presentation with quotations from and references to Wittgenstein's writings, which can be found throughout Stern's book. In doing so, I am trying to give a more perspicuous view of Wittgenstein's struggles than Stern offers us. My one critical point about his contribution is that he tries to present so many of Wittgenstein's problems and their suggested solutions, and how they are related to each other, that a reader may have difficulties following the main road of Wittgenstein's development. As Stern provides several previews, summaries, and retrospects, and takes up issues from earlier sections in order to discuss them from a new perspective in later ones, repetitions in the text occur and some confusion on part of the reader may result. As I proceed, I will sometimes add comments on how compelling Wittgenstein's arguments are, or what the philosophical significance of some his changes is, though it should be clear that the force or the significance of what is supposed to be an argument cannot be evaluated independently of the practices and the background motives, interests, or convictions of those confronted with it. 3.1. It is useful to remember why the TRACTATUS philosophy was, and for many philosophers still is, so appealing: It is the so-called picture theory which proves to be extremely fruitful. The theory maintains that language can be looked at as an isomorphic model of the world due to the common logical structure of language and reality. This 'theory' makes several philosophical claims: (i) From an ontological point of view, it tells us that the world consists of simple objects. (ii) From a semantical point of view, it tells us (a) why linguistic expressions have a meaning (they refer to objects or states of affairs), (b) that a proposition is meaningful if and only if all its non- logical constituents of its logical form have a reference, (c) that the limits of language and the limits of thought can be determined by the boundaries of the range of meaningful uses of language, and (d) why it is possible that two speakers understand each other: the non-logical constituents of the propositions they use refer to the same objects, which exist independently of the speaker's subjective points of view. (iii) From an epistemological point of view, the TRACTATUS tells us that there is some kind of isomorphic projection between atomic states of affairs and elementary propositions, but leaves it to empirical psychology to describe how such a somehow mental projection is actually carried out. It has to be added that, from a semantical point of view, this mental process of projecting constitutes reference. It explains why dead signs, e.g. sounds or inscriptions, can have a meaning (cf.p.105), and why a speaker knows what he or she means by an linguistic expression. -- It should be clear that the picture theory is only a fruitful metaphor and lacks further justification: it is an elaborated simile based on the intial intuition that pictures and propositions may share some striking similarities (cf.pp.42-43). 3.2. I believe that the TRACTATUS is already a work of transition. Although it may initially appear as a closed philosophical system, the TRACTATUS contains some serious and finally irreconcilable antagonisms. That becomes obvious in particular if one recalls the open question of what the simple objects are supposed to be. Reading the TRACTATUS from the beginning, a somehow materialistic or physicalistic ontology emerges, where simple objects are physically indivisible particles forming the eternal substance of the world, whereas reading the TRACTATUS in reverse (in particular the 'the world is my world'-part around TLP 5.6), simple objects seem to be some kind of sensations or sense-data. Both readings have their shortcomings in light of other propositions in the TRACTATUS, and Stern shares my conviction that this issue is not decidable on the basis of the TRACTATUS text (p.63, with substantial confirmation by Wittgenstein himself, pp.63-64). The unresolved tension is typical for the empiricist tradition: it is the tension between trying to be guided by the natural sciences, mainly physics, on the one hand, and taking the starting point of sensations or sense-data as the empirical given on the other. That this tension is to be found in the TRACTATUS indicates that Russell's empiricistic thinking had much influence on Wittgenstein. Due to the pictorial relationship between language and the world, these two ontologies require two different kinds of language: an ordinary or physicalistic language talking about objects like watches on a table (*5*) on the one hand, and a phenomenological language talking about sensations or sense-data (with colour-sensations and - words as paradigmatic cases) on the other. Stern shows that Wittgenstein in 1929 was trying to combine these two languages into one account (cf.pp.98-100,128-135). 3.3. In 1929, Wittgenstein considered the phenomenological language, the language of immediate experience, to be the primary language, whereas he looked at the ordinary or physicalistic language as a secondary language. This choice or evaluation might be historically due to the fact that the physicalistic language was ready at hand during Wittgenstein's development of the picture theory until 1916, while the part closer to the phenomenological approach emerged due to his reading of Schopenhauer only after that (cf. his NOTEBOOKS). 3.3.1. This choice is certainly also systematically motivated by holding fast to the epistemological account of the TRACTATUS (cf. section 3.1.iii above): If the projection between names within elementary propositions and simple objects is supposed to be some kind of mental process (and does not need to involve any further theoretical or purely logical reduction from perceptible objects, such as watches, to theoretically posited entities, such as atomic particles), the simple objects cannot be 'smaller' than sense-data or sensations. For physically indivisable particles beyond the threshold of human perception would then be absurd. -- This move is, of course, not compelling: anyone intending to hold on to the picture theory's ontology and semantics may adjust this account by developing another kind of epistemology. As Stern never mentions this way out, one may conjecture that Wittgenstein never thought of it. 3.3.2. According to Stern, Wittgenstein motivates his choice of the phenomenological language as primary because of the fact that he has realized that he could not provide a finite analysis of everyday or physicalistic propositions like "my watch is on the table" in the TRACTATUS (cf.p.133). If that were so, that is, if the logical analysis of such ordinary propositons were infinite, Wittgenstein's philosophical account would be unable to explain why two speakers can in fact understand each other, for understanding can only be gained if the two refer to the same simple objects by their elementary propositions (cf. section 3.1.ii.d above). And it would also imply that a speaker does not understand even his own propositions until they have been analysed (p.136). However, this trouble can be overcome if the process of analysing ordinary propositions is conceived to end at elementary propositions expressing immediate experience, which means that the non-logical constituents of elementary propositions refer to sense-data as simple objects (cf.p.134). Such a language, describing 'directly' immediate experience, is called "PHENOMENOLOGICAL LANGUAGE" by Wittgenstein. Yet, the introduction of a phenomenological language as the language of the immediately given in order to provide a semantical account of the possibility of understanding ordinary language has the consequence that Wittgenstein seems now to be unable to give an ontological account on the world as it 'really' is (cf. section 3.1.i above). For a language of the immediately given describes only the world as it appears to us. Wittgenstein's philosophy is now on the edge of becoming idealistic or even solipsistic. 3.3.3. But Wittgenstein is facing a problem here: "The world we live in is the world of sense-data, but the world we talk about is the world of physical objects" (WITTGENSTEIN'S LECTURES 1930-1932, p.82, quoted by Stern p.12). That is to say: our everyday language, which is shaped for pragmatic concerns, i.e. for the need to successfully manipulate our environment (pp.12,79), appears inadequate for the task of describing or fitting to immediate experience (p.12). In 1929, Wittgenstein therefore tried to combine these two languages into one account. Stern shows (pp.80-81) that Wittgenstein believed in a veritative symmetry of these two languages, that is: as propositions of the two languages are supposed to share the same truth conditions, anything one can say about one's own experience in the phenomenological language can also be said in the physicalistic language (pp.80-81). It is striking that Wittgenstein never considered ordinary language to be inadequate, whereas he often considered it to be misleading. In the TRACTATUS, ordinary language can mislead because it hides the actual logical form of propositions, whereas the phenomenological account considers ordinary language to be misleading in not representing the world correctly, since it talks about tables and watches instead of sensations or sense-data (pp.130,132). 3.3.4. This caused Wittgenstein to think hard about a problem he had treated as irrelevant in the TRACTATUS: the 'logic of colours.' He realized that colour propositions are not captured by the truth- functional logic of the TRACTATUS and its independent elementary propositions: "If, for instance, I say such and such a point in the visual field is BLUE, I not only know that, I also know that the point isn't green, isn't red, isn't yellow, etc. I have simultaneously applied the whole colour scale. [...] It's such a whole system which is compared with reality, not a single proposition" (WITTGENSTEIN AND THE VIENNA CIRCLE, p. 64, quoted pp.99-100). Stern says that Wittgenstein was now forced to acknowledge the consequences of recognizing the limitations of truth-functional logic (p.100), but Stern does not make entirely clear that this required at least two further assumptions. For an apologist of the TRACTATUS may be quick to object that the colour problem shows only that a proposition like "THIS [point] is blue" is not an elementary proposition, and that sense-data are therefore not Tractarian simple objects. Wittgenstein, on the other hand, had in the meantime (i) decided to adopt a phenomenological language as the language of the elementary propositions, and (ii) he had accepted one of the leading theories of colour of this time, which takes four colours as the primary ones, implying that all other colours can be reduced to them, while the four primary colours are not reducible to one another (cf.pp.130,132-133). Only if these two assumptions are made does the retreat from a logical analysis in the spirit of the TRACTATUS become necessary. 3.3.5. Both the conception of language and the task of philosophy change dramatically with the introduction of logically interrelated elementary propositions of a phenomenological language. Philosophy becomes primarily phenomenology, the logical study of the phenomena of experience (p.139), it consists in "the logical investigation of the phenomena themselves" (SOME REMARKS ON LOGICAL FORM, p.163, quoted by Stern pp.129,135). The undertaking is nevertheless still 'logical,' since in order to get access to 'the phenomena themselves,' ordinary language has yet to be analyzed. But the goal of analysis consists now in uncovering elementary, phenomenological propositions depicting immediate experience somehow 'directly.' Having given up the TRACTATUS conception, Wittgenstein began to look at language as a system of rules: "If, then, I understand what the specification of a length means, I also know that a man is 1.6 m tall, he is not 2 m tall. I know that a measurement determines only ONE value on a scale and not several values. If you ask me, 'how do I know that?' I shall simply answer, 'because I understand the sense of the statement.' It is impossible to understand the sense of such a statement without knowing the rule" (WITTGENSTEIN AND THE VIENNA CIRCLE, pp.77-78; quoted pp.100-101). Wittgenstein here conceives the grasp of a rule as unproblematic - it is presupposed by the language we are already mastering. Philosophy therefore becomes a matter of clarifying the rules we have been following and knowing all along (p.101). Stern appropriately labels this stage of Wittgenstein's philosophy "logical holism" (p.99). He however makes clear that this account does not consist of a full-blown holism, since it merely aims at analysing or clarifying the systematic structure of logically related concept-'bundles' such as colour or number concepts (pp.99- 100). 3.3.6. One may characterize the underlying motives for the shifts mentioned in this section (3.3.) by saying that Wittgenstein is now taking familiar phenomena like colours or measurement more seriously than the philosophical theory of the TRACTATUS had allowed him to do. This is, of course, not a compelling argument given the competition between different philosophical accounts (so that the advocate of the TRACTATUS may remain unconvinced), but it is an attitude I find highly appealing and very respectable. -- Whereas the TRACTATUS was concerned with clarifying the logical form of language, discovering the substance of the world, and showing psychology its way, Wittgenstein's new conception of philosophy is entirely restricted to clarifying language. Reality is limited to something immediately given, but it remains obscure how it 'comes in' into us (presumably, Wittgenstein still believed in a 'mental projection,' but see section 3.4.2.). 3.4. By October, 1929, however, Wittgenstein had already begun to realize that a phenomenological language was useless (cf.p.137): "For what would this language express? Presumably, what we now express in our ordinary language" (PR 1). In other words: By speaking a phenomenological language, we were not able to gather more information about the world than we have already speaking ordinary language, and there was no change in our behaviour. 3.4.1. Yet, this way of looking at phenomenological language would not impress anyone who wants to detect 'the phenomena themselves' for theoretical or philosophical reasons. Wittgenstein does, in fact, provide suggestions why the attempt to develop a phenomenological language seems to be a hopeless undertaking or must be even "absurd" (MS 107, p.176; quoted by Stern, p.137). For in order to be understood, a phenomenological language has to be interpreted by means of ordinary or physicalistic language, and Wittgenstein points out that this is hardly possible: The immediately given, or what is just present, is something which is not in time. It is timeless (p.146), since it refers to something which is there 'in less than an instant' and cannot exhibit any indication of past or future events. But the use of ordinary language, considered as a system of signs spoken or written down, is a physical process, which "unwinds in physical time" (PR 69). Hence, it is impossible to speak a time- and tenseless phenomenological language - it is a 'never never language' (*6*) or an "inarticulate sound" (PR 67, quoted p.147). As similar objections can be raised with regard to space, Wittgenstein sums up: "We are talking about the spatial and temporal phenomenon of language, not about some non-spatial, non-temporal phantasm" (PI 108). 3.4.2. The argument also becomes important in a related issue. "No physical representation [such as ordinary language] could ever be equivalent to a phenomenological language, for the signs would have to be applied and brought into a projective relationship with the world" (p.148). Like any other process in the physical world (including the psychological world; cf. sect. 3.1.iii above), 'mental projection' once again involves time. Stern adds: "[N]o description of the phenomena will describe 'that which goes on in the reading of the proposition,' the mental act that transforms the dead signs into meaningful symbols. That relationship is a precondition for understanding any language whatsoever; it is not something that can be stated IN language" (p.149, including a quote from Wittgenstein's). -- This is a very interesting move, because it points out that both the Tractarian and the phenomenological accounts try to go beyond the limits of (ordinary) language. Wittgenstein now admits that the projective relationship between a word and an object, the reference, is not constituted by something mental and therefore is not an object of investigation of empirical psychology. Rather, reference becomes inscrutable. Consequently, abandoning the idea of an isomorphic projection between language and the world and therefore giving up the conception of language as a representation of the world, not only the idea of something 'immediately given' vanishes, but the entire possibility of providing an ontological account of how the world 'really' is disappears. The best thing we can now do is to say merely that our world consists of those things we are talking about. This is obviously the origin of Wittgenstein's linguistic idealism (*7*). 3.4.3. These considerations certainly do not oblige an advocate of phenomenology (in the Wittgensteinian sense) to abolish it, since he may acknowledge this line of reasoning and still remain unconvinced that his phenomenological convictions or intuitions have been refuted. He may hope to overcome the difficulties in one philosophical way or another, e.g. in a transcendental way, taking time to belong to the 'form' of experience (*8*), while assuming a special philosophical or transcendental way of speaking. -- In a similar vein, the objection that the phenomenological account consists of extreme verificationism, such that a word has a different meaning every time we associate different immediate experiences verifying the propositions containing the word (p.152), is not compelling either. For this difficulty may be resolved by introducing some kind of 'power of abstraction,' a type- creating process of which the human mind is supposed to be capable. 3.5. It should be obvious that Wittgenstein's phenomenological account in 1929 contained the possibility of a (linguistic) solipsism in which the human mind is able to convert dead signs into meaningful ones by extraordinary mental processes. As is well known, this picture becomes the target of Wittgenstein's criticism in the BLUE BOOK and in later writings. His 'knockdown argument' is there supposed to be the line of reasoning of the so-called private language argument. 3.5.1. To begin with, the advocate of the possibility of linguistic solipsism, stuck with his own phenomenological language referring to his own private sense-data only, can hardly explain how it is possible that two 'phenomenological' speakers understand each other. For they may not have anything in common -- neither a language nor objects to which the expressions of their languages refer. Understanding or communication among speakers of a linguistic community depends on the idea that at least some of their words refer to things perceptually present at least in principle to all of them (e.g. the term "cat" refers to cats). 3.5.2. In addition, linguistic solipsism can hardly give us an account of how language is acquired, whereas Wittgenstein's later approach uses the fact of a community's training infants and children to make certain responses under certain circumstances to explain why they will finally exhibit almost the same (linguistic) behaviour as the other members of the community (cf.pp.127,187- 192). 3.5.3. Furthermore, Wittgenstein argues that any appeal to mental states or processes or whatever is going on 'in the head' is useless in deciding whether or not a person has understood a word. For any kind of representation 'in the head,' considered to be an interpretation of a linguistic sign, needs to be interpreted anew, and this merely opens an infinite regress (cf.PI 138- 155). Rather, the application of a sign indicates one's understanding, and it is obviously only its application that matters in a conversation. 3.5.4. Finally, a solipsistic speaker, having access to his private sense-data only, would be unable to decide whether his statements were actually true or whether he only believed them to be true (PI 202, 258). That is to say: the well-known phenomenon of making a mistake, of holding fast to false statements, cannot be explained on the basis of phenomenology. The "superprivacy" of phenomenological language has cut itself off from our community's everyday language at the price of robbing the phenomenological language of any sense at all (cf.pp.180- 181). 3.5.5. In other words: language is a social phenomenon. In order to be meaningful, a sign has to be used under particular circumstances, in a context or in a community's practice (cf.p.188-189; PI 198-202). And as Wittgenstein also argues that the meaning of a sign depends on the remainder of the language it belongs to (PI 27-35), Stern rightly characterizes this philosophy as "practical holism" (p.120). 3.6. Wittgenstein's new account of language explains the possibility of understanding or communication by saying that two speakers are following the same rules. But in retrospect (cf. section 3.1. above), we see that he is now unable to give an ontological account on how the world 'really' is (since two speakers can understand each other if they agree on taking a certain object to be called "cat", whether or not this object actually is a cat). Reference, being the relationship between a word and an object, remains elusive, and this is also true for the epistemological problem of how we make contact with the world around us in order to make judgments about it. Consequently, the question "How is it possible that a speaker understands his or her own utterances?" cannot be settled either. Finally, no account concerning 'the limits of language and thought' is given. On the other hand, further questions emerge: What is involved in following a rule? How do I know in which language game I am participating? How do I know which rule I should follow? Wittgenstein's notes do not give us any answers to them (I insist until further notice on my view that these questions make sense even if 'I obey the rule blindly' (PI 219)). 4. FINAL REMARKS Wittgenstein once said to Rush Rhees: "In my book I say that I am able to leave off with a problem in philosophy when I want to. But that's a lie; I can't" (*9*). This confession makes sense to me. For as long as a serious philosopher feels captured by or entangled in philosophical problems, he or she cannot stop thinking about them until they have been solved or dissolved in some manner or other. Otherwise, the philosopher remains tormented - even though he or she may take a break for a while. Admittedly, Wittgenstein aimed to bring philosophy to an end, but being a sincere thinker, he continued to struggle with philosophical problems throughout his entire life (cf.p.20). No wonder his philosophy was always in transition. Until I read Stern's book, my own practice had more or less neglected the material of 1929 and the early 1930s, since I wanted to understand the PHILOSOPHICAL INVESTIGATIONS and especially ON CERTAINTY. I had difficulties enough reading the PHILOSOPHICAL REMARKS or the PHILOSOPHICAL GRAMMAR, since they are neither based on the TRACTATUS philosophy nor do they really develop the approach of the PHILOSOPHICAL INVESTIGATIONS yet. Generally, I looked at these and other 'intermediary' writings only if I could find no supporting remarks on issues connected with or dependent on those I was working on. I believe that I am not the only one to have followed such a procedure. Stern has convinced me that I had overlooked something important: Not only do I now understand the transition from the TRACTATUS to the PHILOSOPHICAL INVESTIGATIONS better, I even understand the arguments of the PHILOSOPHICAL INVESTIGATIONS much better (*10*). ______________ Endnotes: *1* Stern's book confirms my impression: Although he quotes many unpublished remarks, their number is relatively small in comparison to the number of his quotes from already published material. *2* "Ich bin in einem Wirrwarr" in MS 132, p.161 is quoted with references as regards its context along with an example showing Wittgenstein at a loss in: Michael Kober, GEWISSHEIT ALS NORM, Wittgensteins erkenntnistheoretische Untersuchungen in 'Ueber Gewissheit', Berlin: de Gruyter 1993, pp.232-233. *3* In this case, however, Stern suggests a search of other sources in the Nachlaß (p.176), alluding to Wittgenstein's "Notes for the Philosophical Lecture'", ed. D. Stern, in: L. Wittgenstein, PHILOSOPHICAL OCCASIONS 1912- 1951, ed. J. Klagge and A. Nordmann, Indianapolis, In.: Hackett 1993, pp.445- 458. *4* I had to realize in conversation that the anecdote with Piero Sraffa, who showed Wittgenstein an insulting Neapolitan gesture and asked him to specify its logical form (cf.p.107) - which Wittgenstein was unable to give -, is often taken for Wittgenstein's reason to abandon the whole TRACTATUS. I should mention that this is certainly not a 'knockdown argument,' but rather contributes a motive. For either Tractarian language was supposed to be concentrating on scientific language only (and Neapolitan insulting gestures presumably do not belong to it), or these gestures should have been treated in a way one is apt to treat questions, commands, etc. from the Tractarian point (whether or not this is possible I do not know). Only if one is convinced that Neapolitan gestures do not have a Tractarian logical form, and if the TRACTATUS is taken to handle all kinds of human acting (cf.PI 23), then Sraffa's challenge amounts to an argument. *5* Later on, the Vienna Circle would call ordinary language talking about 'physical' objects like watches, tables, and chairs a "physicalistic language" (cf. Carnap's and Neurath's papers of that period). That is: physicalistic language, understood in this way, is not the technical language of physics. *6* I owe this expression to Heikki Nyman (in conversation). -- The impossibility of a language referring to something 'immediately given' in a timeless manner reminds me of Zeno's paradox of the flying arrow, in which no possibility is left to give an adequate account of movements and changes either. *7* Stern, it should be mentioned, does not think that Wittgenstein's later philosophy consists of a linguistic relativism (p.127), whereas I do. However, I conceive it to be an EXTERNAL kind of linguistic relativism, that is: referring to the customs of a community acting within the world and therefore not immune to criticism; cf. Michael Kober, GEWISSHEIT ALS NORM, loc.cit., chapters 3.9.-3.10. *8* Even Wittgenstein played with this idea (cf.PR 52); cf.p.147. *9* From Garth Hallett, A COMPANION TO WITTGENSTEIN'S "PHILOSOPHICAL INVESTIGATIONS," Ithaca, NY 1977, p.230, quoted by Stern, p.20. -- It should be emphasized that the wording of PI 133, to which Wittgenstein seems to refer, can, but NEED NOT be read as if he actually believes in the possibility of stopping doing philosophy (Hans Sluga pointed this out to me). *10* I am grateful to Hans Sluga and Toby Stewart, who ensured that my English is ordinary philosophical English.