Philosophical Books Vol. 51 No. 2 April 2010 pp. 102–116

“HE DOESN’T WANT TO PROVE THIS OR THAT”— ON THE VERY YOUNG WITTGENSTEIN phib_504

102..116

CATARINA DUTILH NOVAES

University of Amsterdam Michael Potter’s Wittgenstein’s Notes on Logic (Oxford, OUP, 2009) is an impressive piece of scholarship.1 It can be described as a work of conceptual archeology, attempting to retrace the development of Wittgenstein’s ideas at the very early stages of his intellectual life, in particular the period in which the Notes on Logic were composed. Their final composition took place in the fall of 1913, but (as so many of his writings) the Notes are a compilation of extracts from Wittgenstein’s notebooks, which in turn span through the whole period of his early collaboration with Russell in the prewar period. The main thread of the book is the development of Wittgenstein’s thought against the background of the two major intellectual influences at this early stage, Frege and Russell.2 As such, the enterprise clearly requires profound knowledge not only of Wittgenstein’s ideas but also of Frege’s and Russell’s, and luckily for the reader, Potter fulfills this requirement with flair. One of the tangible results of Potter’s analyses is a new proposed edition of Wittgenstein’s Notes on Logic, to be found in Appendix B of the book, which differs significantly from the so-called Costello edition (Costello 1957) under which the text is mostly known.3 In this notice, I address four topics that emerge from the book: its methodological choices; its discussion of the interplay between Wittgenstein, Russell and Frege; aspects of the relation of the Notes to the Tractatus; and the disappearance of the epistemic subject from the realm of logic in the hands of the young Wittgenstein. Given the richness of analysis in the book, there is a range of other topics equally deserving of attention, so the choice of topics here reflects my personal interests and is not intended to provide a complete overview of the book. For that, the reader is simply urged to read the book itself; it is extremely well written, almost a page-turner, which is a rather astonishing feature in a work of philosophical scholarship. Potter has a talent for explaining complex ideas and concepts in ways that make them sound almost uncomplicated, while avoiding oversimplification. 1. I will not comment any further on the several practical features of the book, but let me say that it contains a wide range of useful devices such as an index listing all citations to passages from the Notes in the book, among others. 2. Some may object that Potter does not pay sufficient attention to the crucial influence of Boltzman and Hertz over Wittgenstein, but for reasons of space I shall not pursue this matter any further. 3. There is another, less well-known edition of the Notes: Michael A.R. Biggs “Editing Wittgenstein’s ‘Notes on Logic,’ ” Workings Papers from the Wittgenstein Archives at the University of Bergen n.11 (1996, 336 pp £40, hardcover), Accessed February 26, 2010

102 Philosophical Books Vol. 51 No. 2 © 2010 The Author. Journal compilation © 2010 Wiley Periodicals, Inc.

1. Methodology At the outset, in the Introduction, Potter points out that the methodology he is about to deploy in the book is not strict, conventional textual analysis. Indeed, another possible description of his enterprise is as a biography of ideas, not of lives. But he assumes (correctly, to my mind) that to understand the development of ideas it is inevitable to take a closer look at the carriers of these ideas than is usually the case in conventional philosophical analysis. In practice, this entails the need for closer inspection of slightly unconventional sources such as correspondences, diaries, and other personal sources. A very positive feature of the book is Potter’s meticulous attention to letters written by the main characters of the plot; Russell’s letters to his lover Lady Ottoline Morrell, in particular, offer significant insight into the developments of that period. When developing his ideas at this early stage, Wittgenstein was of course profoundly engaged in debates with Frege and Russell: Russell as a constant physical presence, Frege by means of sporadic visits and letters. Potter is deeply interested in the process of this development, not only in the end results: the assumption is clearly that there are good reasons to be interested not only in the end product, that is, in the texts actually produced (the polished Tractatus, but even the Notes on Logic), but also in the very process leading to the composition of these texts. In a similar vein, Potter is not only interested in the arguments that Wittgenstein or others may have put forward to defend certain views, but also in their reasons for defending these views. While arguments aspire to objective validity/cogency, reasons may be, and often are, deeply subjective. Reasons pertain to the genealogy of a given idea or theory, and are typically related to external assumptions and commitments which may themselves not be fully justified, and yet may play a fundamental role in the intellectual choices and preferences of a given person. Arguments, by contrast, must be self-contained or in any case must make all underlying assumptions explicit. Arguments must be compelling: upon accepting the premises, one cannot but accept the conclusion; but reasons are typically not compelling in this way, they leave room for different responses. It is most probably for these reasons that traditional philosophical methodology focuses on arguments and eschews reasons subjectively construed; prima facie, there can be no objective judgments of propriety as to whether a given reason is indeed a legitimate reason for a given position. The dichotomy also echoes Popper’s once influential distinction between the context of discovery and the context of justification, and his contention that only the context of justification is of (philosophical, scientific) interest. Potter, by contrast, is interested in the “context of discovery” of Wittgenstein’s views, and in particular in their progressive development. One might expect that, for this reason, Potter might be too soft on his characters, prioritizing the subjective level of reasons over the objective level of the correctness and cogency of arguments. Not so, and this is in fact a very 103 © 2010 The Author. Journal compilation © 2010 Wiley Periodicals, Inc.

refreshing characteristic of the book.4 Wittgenstein scholars tend to be overly respectful toward their object of study, often going to considerable lengths to find interpretations that make him come out as holding perfectly reasonable positions grounded in perfectly legitimate arguments (as noticed by Potter himself in his recent review of M. McGinn’s Elucidating the Tractatus [Potter 2010, 192]). Potter will have none of that; when discussing not only Wittgenstein’s but also Frege’s and Russell’s ideas and doctrines, he will often say that a given argument is not a very good argument for the position in question, or even that it is not an argument at all but in fact no more than a reformulation of the thesis being defended. (Indeed, why should we think that these great minds would always formulate rock-hard, impregnable arguments? Even they must have had their off-days.) Still, the question remains: why should we depart from traditional philosophical textual analysis, with its focus on end products and contexts of justification? A first obvious reason is that it would be quite a stretch to characterize the Notes as an end product; indeed, it is far from clear what its status was for Wittgenstein himself and others. Reports of discussions? Preliminary work? A proto-BA dissertation to be submitted in Cambridge? Moreover, the connections between the different parts of the Notes is also a topic open to speculation, and Potter’s painstaking work of textual and historical analysis gives us a very plausible account of the matter (Appendix A), leading to a reconstruction of the text (Appendix B) that differs significantly from the Costello edition. But perhaps more importantly, even in his polished texts (e.g., the Tractatus), Wittgenstein rarely offers clear-cut arguments for the views he presents and the claims he makes. So in this case, perhaps more than in other cases, the methodology used by Potter is required in order to shed at least a dim light on what exactly Wittgenstein was up to. As Potter puts it: There are several benefits to be derived from this approach. One is that it gives us a better chance of understanding Wittgenstein’s own reasons for some of the views he held. This is probably a good reason to study the early works of almost any major philosopher, but it is especially so with Wittgenstein, whose own arguments for his views are so often too compressed to be comprehensible without understanding the context in which he formulated them. (p. 1, emphasis added) If we are to gain the maximum insight from his work, we need to understand, certainly, what motivated him to address the problems he did in the way he did. (p. 2) 4. In fact, it is the combination of these two features that makes Potter’s approach somewhat unusual. Many interpreters have defended the idea that, in order to understand Wittgenstein’s writings, it is essential to know a great deal about Wittgenstein himself (as claimed by R. Monk in the introduction of his Wittgenstein biography and by the proponents of the so-called “Chicago reading” of the Tractatus). (I owe this point to Aidan McGlynn.) What is novel here is to see Potter coming from a whole different tradition and putting forward an interesting combination of biographical analysis and critical approach.

104 © 2010 The Author. Journal compilation © 2010 Wiley Periodicals, Inc.

Is Potter’s proposed methodology justified and fruitful? I think it is, but only insofar as the biographical, “subjective” level of reasons does not supplant the critical, “objective” level of analysis of the validity of arguments and of the internal coherence of the doctrines presented. And given that Potter does combine both levels in his investigation, the potential risks of his quasibiographical approach are dissipated; interest in the reasons why someone does or says certain things does not entail blindness toward the cogency of the arguments put forward to defend a certain claim. Ultimately, Potter seems to be making a plea for methodological pluralism in philosophical analysis: his quasi-biographical perspective is intended to clarify aspects that remain otherwise obscure and impenetrable within the end product focus of traditional textual analysis, but it is not intended to replace it. The claim seems to be that, particularly in the case of the very young Wittgenstein, the peculiarities and idiosyncrasies of this young philosopher seem to require methodological pluralism. One example of the fruitfulness of Potter’s quasi-biographical approach is his account of what Wittgenstein himself considered his “fundamental thought” (section 5.7). This fundamental thought was the idea that there are no such things as logical constants. The puzzling thing is that, in the Tractatus, this claim is given a rather unimportant location (4.0312), so it is not immediately obvious in which sense it is fundamental. Potter claims that it is fundamental from a biographical point of view: it constituted a turning point for Wittgenstein in the process of distancing himself from Russell’s conception of logic toward his own conception, fully formulated in the Tractatus. Throughout the book, this hypothesis is further tested and, to my mind, convincingly established as correct. Nonetheless, it seems to me that the book would have benefited from a more extensive analysis of its methodological principles and their implications. Potter seems to suggest that the results achieved in the book should speak for themselves, and therefore that no extensive justification of his somewhat unconventional methodology is required. Still, his methodological choices might have given him the opportunity to offer interesting reflections on philosophical methodology in general. This would have been a welcome addition to current discussions on philosophical methodology as well as to the book as a whole. 2. The Dynamics of the Interplay Between Wittgenstein, Russell, and Frege One of the most engaging aspects of Potter’s book is his rendering of the interactions between Wittgenstein, Russell, and Frege. Of course, we all know that Wittgenstein’s early exposure to philosophy was almost exclusively limited to the works of Frege and Russell; somehow, he had gotten hold of Russell’s Principles of Mathematics when studying in Manchester, and via this book became familiar with the work of Frege. Moreover, we are all familiar with the Vorwort of the Tractatus, where Russell and Frege are both mentioned as those to whom 105 © 2010 The Author. Journal compilation © 2010 Wiley Periodicals, Inc.

Wittgenstein owed “in large measure the stimulation of [his] thoughts.” But what Potter narrates very vividly is how exactly Wittgenstein was influenced by Frege and Russell on a number of topics and issues, that is, the details of how Wittgenstein’s ideas were shaped against the background of the influence of both authors. In Potter’s book, this complex interaction of people and ideas, oscillating from uncritical acceptance to outright rejection or even to dissolving completely a certain number of issues, becomes a lively, dynamic process, as it surely was when actually taking place.5 (One of its most dramatic and interesting episodes consisted in Frege sending Wittgenstein to Russell under the argument that he was too old to be Wittgenstein’s mentor. The anecdote is widely circulated, but as Potter points out [p. 19], its exact circumstances are still unclear.) Of course, Wittgenstein almost invariably added his own twist to what he borrowed from Russell and Frege, not being one to simply take over ideas and opinions from others. Nevertheless, at times he seemed to absorb almost uncritically some of Russell’s views, such as the idea that names refer directly to objects (p. 13). Indeed, up to a certain point, Wittgenstein’s relationship to Russell was clearly that of a pupil toward a mentor, and narrating the transition from the status of Russell’s pupil to the status of his equal is one of Potter’s topics in the book. As part of the same process, Frege’s influence over Wittgenstein becomes increasingly felt.6 In effect, while Russell has undoubtedly shaped Wittgenstein’s intellectual development in these early years like no other, in many senses Frege was more of a kindred spirit to Wittgenstein than Russell. Russell was essentially interested in using logic as a tool to discuss philosophical questions, but was much less interested in a philosophical discussion of logic itself. Wittgenstein, like Frege, was adamant in wanting to understand the very nature of logic. As Potter reports, in a letter to Lady Ottoline Morrell, Russell formulates a stunningly accurate description of Wittgenstein and his general approach to logic: “He doesn’t want to prove this or that, but to find out how things really are” (p. 50). In modern terminology, one could say that Russell was essentially interested in philosophical logic, not in the philosophy of logic, and Wittgenstein’s focus on the latter rather than the former is certainly at least partially to be credited to Frege’s influence (although Potter follows British terminology, using “philosophical logic” also to refer to “philosophy motivated by logic,” which I prefer to refer to as “philosophy of logic”).7 Potter also identifies the influence of Frege in Wittgenstein’s most fundamental “guiding principles” (presumably, methodological principles as well as principles determining the very questions to be asked), which would guide him in the Tractatus and possibly even beyond.8 5. Potter also elaborates on the influence that Wittgenstein may have had upon Russell and Frege (29.6 and 29.7), but I will not comment on this here for reasons of space. 6. “In previous chapters of this book it has almost exclusively been Russell’s influence on Wittgenstein that we have been discussing. From now on, however, it will be increasingly necessary to take into account the influence, in writing and in person, of Frege.” (p. 58) 7. On the confusion concerning this terminology, see the discussion at http://people.ucalgary.ca/ ~rzach/logblog/2008/06/philosophical-logic-and-mathematical.html 8. “These guiding principles Wittgenstein owed to Frege, not to Russell.” (p. 262).

106 © 2010 The Author. Journal compilation © 2010 Wiley Periodicals, Inc.

Moreover, he claims that “what is of lasting worth” in the Tractatus, given that it is “riddled with implausibilities” (p. 262), namely insight into the central themes pertaining to the philosophical analysis of logic, Wittgenstein owes fundamentally to Frege. The remarkable thing about Frege is that he not only introduced a whole new approach to logic; he also held sophisticated philosophical views on its very nature (which are in fact at times more “conservative” than his logic properly speaking), and deemed it essential to raise these fundamental questions which are much too often overlooked (in Russell’s words, Frege wanted both to “prove this or that” and to “find out how things really are”). Wittgenstein gave radically different answers to these questions, but he was certainly of the same mind as Frege concerning their importance. Of course, throughout the book the relevance of Russell’s influence is made abundantly clear; Potter surely does not side with, for example, Dummett in maintaining that Frege’s influence over Wittgenstein is overwhelmingly more significant than Russell’s. Russell’s influence was significant in many important ways, including on a personal level. But ultimately, the picture painted by Potter is that of Frege exercising a lasting and profound (even if at a distance) influence over Wittgenstein, concerning very fundamental aspects of his philosophical enterprise as a whole (see in particular section 29.5).9 To illustrate at once the richness of patterns of Wittgenstein’s interaction with Frege and Russell, and Potter’s vivid description of these events, I now quote a fairly long passage from the book, where both aspects are particularly well represented. Potter is here discussing Frege’s threefold theory of reference, involving not only terms and objects but also mediating senses, and Russell’s rejection thereof (in fact, there is a fourth element in Frege’s account, namely subjective ideas of objects present in each person’s minds). A fundamental difference between Frege’s theory of reference and Russell’s therefore is that Frege aimed to explain how communication is possible, whereas Russell treated communication only as an observed fact from which it might be possible to draw conclusions (for instance, about the existence of other minds). Wittgenstein, on the other hand, was quite happy to circumvent the issue entirely by tolerating solipsism. “He admits,” Russell reported, “that if there is no matter then no one exists but himself.” And if we embrace solipsism, the problem which led Frege to distinguish ideas from senses, and Russell to make communication a kind of miracle, simply dissolves. (¶) Wittgenstein’s response to this disagreement between Russell and Frege is characteristic in several aspects: it is characteristic that he should have been unwaveringly determined to accept the consequences of his views, however unpalatable or frankly implausible they might have been; characteristic that he did not feel the draw of Russell’s robust (if somewhat selective) common sense; characteristic too, that he should adopt a view one of whose effects 9. There is a forthcoming translation into English of Frege’s letters to Wittgenstein in De Pellegrin (forthcoming); the volume will also contain a critical study of the correspondence by J. Floyd (forthcoming), where, among others, Geach’s hypothesis of the “delayed” impact of Frege’s criticism over Wittgenstein is discussed.

107 © 2010 The Author. Journal compilation © 2010 Wiley Periodicals, Inc.

was not so much to argue against an opposing view but rather to undermine it by reconfiguring the project to which it was intended to contribute. (p. 74) Wittgenstein (in particular in his early period) is certainly among the most courageous bullet biters in the history of philosophy.10 The bullet to be bitten here was the bullet of solipsism (one he continued to bite in the Tractatus, but of course under his own sui generis version of the doctrine), and the effect it had was to expose one of Frege’s very goals when formulating a theory of reference, namely to ensure the possibility of intersubjective communication, as a nonissue. In Frege’s theory, senses are postulated to a large extent in order to avoid the “intolerable” conclusion that, ideas being essentially subjective, and if words were related only to subjective ideas, there would be no real communication during a linguistic exchange: people would be constantly, and literally, talking past each other. For Wittgenstein, however, people are not talking past each other because they are not even talking to each other to start with. Solipsism in the Tractatus and possible reasons compelling Wittgenstein to embrace such a radical position will be briefly discussed in section 4 below,11 but the point here was to illustrate the complexity of Wittgenstein’s reactions to the views of his two main interlocutors: no straightforward rejection, acceptance, or even fusion of the often conflicting poles of influence, but rather more often than not what could be described as turning Frege’s and Russell’s views inside out. Potter’s uncharacteristic method of “biography of ideas” is able to expose these developments in a particularly vivid manner.

3. The Notes and the Tractatus At several occasions in the book, Potter reiterates his goal of studying the Notes and the period of their conception in their own right, not only insofar as they contribute to a better understanding of the Tractatus.12 This seems fair enough, if nothing else because it differs from the typical focus that has been placed on the Notes by Wittgenstein scholars so far. Nevertheless, Potter does discuss the 10. Incidentally, this is one of the features that make the Tractatus particularly suitable for undergraduate teaching: it teaches students that philosophy is primarily about drawing conclusions from premises, no matter how unpalatable they are. 11. But notice that I make no attempt of actually making sense of Wittgenstein’s (alleged) solipsism in the Tractatus. This has been a topic of much discussion among Wittgenstein scholars over the last decade, but I must admit that, to me, it is still the most baffling aspect of the Tractatus. It is a realistic form of solipsism, which seems to imply that the subject does have access to the world, but not to other subjects. 12. “Although I hope in this book to contribute to the same project of Tractarian exegesis, I am to do so by a somewhat different method. Instead of studying the Tractatus, and drawing on Wittgenstein’s earlier writings only when they contribute to understanding it, I shall here be focusing on the 1913 Notes on Logic, treating them if not quite as a terminus in Wittgenstein’s work then at least as worthy of study in their own right.” (p. 1)

108 © 2010 The Author. Journal compilation © 2010 Wiley Periodicals, Inc.

general question of what, in the Notes, announces what was to come in the Tractatus as well as the ways in which the latter differs fundamentally from the former. We could summarize Potter’s findings and hypotheses concerning the relations between the Notes and the Tractatus in the following way: Potter believes that the Birmingham Notes, which Wittgenstein dictated in German to a stenographer in Birmingham as extracts from his notebooks, formed an important basis for what is known as the Prototractatus. But he also stresses that the philosophical transformation undergone by Wittgenstein in 1916 (visible in his notebook entries) corresponds to a real change in his views, especially with respect to the goals and scope of his enterprise as a whole (now including matters such as ethics and mysticism). This transformation led him to make significant changes in the Tractatus, which was by then already largely written. Therefore, there is only so much that the Notes can clarify with respect to the Tractatus: many of Wittgenstein’s views concerning logic as presented in the Tractatus are indeed to be found in a germinal state in the Notes, but there was also a fundamental shift of focus along the way. Potter does dedicate a whole chapter (chapter 27) to the discussion of “Tractarian objects,” apparently esteeming that the background provided by his discussion of the Notes and the period of their composition could shed some light on the Tractarian issue of the analysis of elementary propositions. But significantly, while in his pre-1916 period Wittgenstein had very little or even no interest in what could be described as a philosophy of value (see section 28.4), the Tractatus notoriously makes rather mysterious claims concerning ethics, mysticism, and other topics on which, in theory, nothing can be said. None of this is in any way anticipated in the Notes, but Potter resists the idea of some sort of mystical conversion in 1916; he points out that, even before the war, Wittgenstein was an avid reader of Tolstoy and Dostoevsky, in whose works one finds a similar form of “fatalistic humanism” (p. 247). All this means that an analysis of the Notes may not be able to add fundamental new insights to the debate between so-called resolute versus so-called irresolute readings of the Tractatus. In Potter’s own terms (but elsewhere), these two positions can be characterized as follows:13 According to the irresolute reader, there are metaphysical truths about the nature of the world toward which the book gestures, even though it cannot quite express them; according to the resolute reader, the book panders to the project of constructing a substantial metaphysics only long enough to explode it as wholly misconceived. (Potter 2010, 192) Still, it is somewhat surprising that, in the book itself, Potter does not say a single word on the debate between irresolute and resolute interpretations of the Tractatus. In fact, some of the major players in this debate (Diamond, Conant, 13. See also S. Hamilton’s review of M. McGinn’s Elucidating the Tractatus in this journal (Hamilton 2009).

109 © 2010 The Author. Journal compilation © 2010 Wiley Periodicals, Inc.

McGinn) are not even listed in his list of secondary sources.14 This does appear to be a conscious choice on Potter’s part, but one wonders what reasons he may have had not to refer at all to this influential debate in Wittgenstein scholarship. Is it because he thinks that an analysis of the Notes and of the period of their composition has nothing to add to the debate? Or is it because he does not consider the debate to be sufficiently interesting so as to deserve attention in his book? As much as I appreciate the fact that, given the considerations just presented here, there is only so much that the Notes can add to this debate, it would have been interesting to hear what Potter might have to say on it against the background of the Notes (even if only to conclude that the Notes have nothing to add in this respect). Potter does seem to suggest that, at least at the time of the composition of the Notes, Wittgenstein’s attitude was “irresolute” rather than “resolute”: there is nothing to indicate that Wittgenstein rejected the idea that there are (or might be) metaphysical truths about the world. Even though the Notes do not contain much in terms of metaphysical considerations, Potter claims (and correctly, to my mind), that “[w]hat we see in operation in the notes [. . .] is a developing method of approaching metaphysics via the symbolism” (p. 241). Now, this is perfectly compatible with the irresolute interpretation of the Tractatus. But again, it might well be that the transition from an “irresolute” to a “resolute” attitude toward metaphysics is part of the important and well-attested philosophical changes that Wittgenstein underwent in 1916. In that case, the Notes cannot help us much indeed. While the irresolute reading of the Tractatus has its problems, there is certainly something frustrating about the resolute reading. How plausible is it that Wittgenstein would have gone through the trouble of thinking seriously about issues such as the structure of facts, the nature of relations, etc. while all along not taking any of it seriously? What emerges from Potter’s book is that he was taking these issues seriously in the prewar period, given that they clearly underpinned his investigations into the nature of the proposition. One could speculate on the possibility of different temporal layers present in the Tractatus; it might explain the apparent tension between the meticulous presentation of the Tractarian ontology at the beginning (belonging to his “irresolute” phase) and the “resolute” conclusion of the Tractatus as we know it, telling us to throw away the ladder and “admonishing us to silence” (p. 247). Potter makes in any case the valuable contribution of showing that, even if he had indeed embraced a total rejection of the idea of metaphysical truths about the world at the time of the final composition of the Tractatus (something that is at any rate far from being definitively established), it all seems to indicate that this was not his position at the time of the composition of the Notes. 14. Generally speaking, Potter (as he recognizes himself in the Introduction) is rather parsimonious in his references to the secondary literature on Wittgenstein. Of course, this literature is massive, and giving a comprehensive overview of it has never been one of Potter’s goals. Still, it would seem that Potter has deliberately chosen not to mention some fairly obvious secondary sources.

110 © 2010 The Author. Journal compilation © 2010 Wiley Periodicals, Inc.

4. The Exile of the Epistemic Subject from the Realm of Logic One crucial aspect of the development of Wittgenstein’s ideas which, to my mind, Potter fails to illuminate adequately, is what we could describe as the exile of the epistemic subject from the realm of logic. This is a very general feature of the young Wittgenstein’s account of the nature of logic, but it is acutely felt in particular in his critique of Frege’s judgment stroke. Wittgenstein famously said in the Tractatus (4.442) that “Frege’s ‘judgment stroke’ ‘|—’ is logically quite meaningless.” At first sight, this may seem a point of detail, but it embodies what is one of the most fundamental differences between the Tractarian views on the nature of logic and Frege’s, namely that in the former there is no place for the epistemic subject which occupies a central place in the latter.15 My claim here is that Potter does not offer a fully satisfactory account of these developments, and in particular of why Wittgenstein’s realm of logic is entirely devoid of (human) actions, be they speech-acts, epistemic acts, or others. Indeed, the only subject still present in the Tracatus is the mysterious metaphysical subject appearing in section 5.6, but significantly, this subject represents the very limits, the outer border, of the world: it is not in the world, and it is not in any way the object of logical analysis. This issue is important also against the background of the broader picture of the development of logic and its philosophy in that period. As argued by B.G. Sundholm in several of his writings (1998, 2001, 2002), and more recently by N.J.J. Smith (2009), Frege’s very conception of logic as the investigation of inference is entirely at odds with how most contemporary philosophers and logicians conceive of it. Indeed, according to currently pervasive views on the nature of logic, human acts such as judgments have no place whatsoever in logical discussions. Logic as conceived nowadays, by contrast, is not centrally concerned with subjects’ judgments at all: it is concerned with eternal relations amongst propositions; that these propositions are possible contents of judgments is, at most, of secondary concern, related only to the possible applications of logic to reasoning. (Smith 2009, p. 7) Thus, it is crucial to understand why exactly (both his reasons and his arguments) Wittgenstein so forcefully rejected Frege’s conception of logic as primarily occupied with the epistemic subject’s acts of drawing inferences and making judgments. (Such a rejection is particularly significant against the background of the marked influence of Frege over Wittgenstein, described by Potter himself throughout the book.) Not only would this give us further insight into these key developments (in particular, the replacement of the notion of 15. As documented in the Frege-Wittgenstein correspondence (unfortunately, Wittgenstein’s letters to Frege no longer exist), Frege reacted very critically to the Tractatus. While there were attempts on both sides to establish sufficient common ground so as to discuss their disagreements fruitfully, eventually the debate came to a halt; at least Frege seems to have given up the hope that they would ever be able to understand each other on the content of the Tractatus. See (Floyd forthcoming).

111 © 2010 The Author. Journal compilation © 2010 Wiley Periodicals, Inc.

judgment by that of proposition at the central stage of logical analysis), it could potentially also provide elements that might contribute to resolving the dispute between the two conceptions. Fregean nostalgics (e.g., Sundolm 2001) often argue that we should return to a Fregean, epistemological conception of logic, but as far as I can tell, they have not been able to provide definitive arguments on why the current “ontic” focus of logic is necessarily wrongheaded, besides claims based on historical authority. Potter gives an overview of what might have been Wittgenstein’s reasons for rejecting Frege’s views, and in particular Frege’s notion of judgment, but he does not offer what might have been Wittgenstein’s arguments to substantiate the view that Frege’s conception of logic centered around the epistemic subject is fundamentally misguided. Now, it seems to me that the main reason why Potter fails to provide an answer to this question is that he himself appears to take Frege’s position to be somewhat misguided (as is particularly clear in chapter 10), failing thus to appreciate that Wittgenstein would need arguments so as to be truly entitled to this rejection. True enough, in section 29.2 Potter elaborates on what he describes as Wittgenstein’s “refusal to argue for his ideas in anything like the conventional manner of philosophers” (p. 251), in particular his rejection of the importance of arguments to philosophy. So it might be thought that here again, Wittgenstein is characteristically not providing arguments for his views, and thus that what I am asking of him (and of Potter) is out of place. But while Wittgenstein notoriously did not always give arguments to support the views he held, he was very keen on rejecting other people’s views with poignant counter-arguments. As Potter narrates himself, Wittgenstein’s counter-arguments often had a paralyzing effect over Russell, leading him even to abandon the project of a book he intended to write. Thus, against this background, the question appears to be perfectly legitimate: what were Wittgenstein’s arguments to reject Frege’s conception of logic as fundamentally concerned with the judging and inferential acts of the epistemic subject? In effect, contra other scholars such as W. Goldfarb, Potter thinks that Frege was confused on the matter of whether assertion (considered as the public manifestation of a judgment) is or is not “merely psychological” (a question explicitly posed to Frege by Jourdain in correspondence). Indeed, it all boils down to what many have seen as a tension between Frege’s own antipsychologism and his insistence on the importance of judgment for logic (so as to require a special symbol for it in the notation, the judgment stroke). But the heart of the matter is that there isn’t necessarily a real tension here, because Frege’s notion of judgment and assertion is essentially normative, whereas his antipsychologism with respect to logic is directed against descriptive accounts of reasoning mechanisms as providing foundations for logic. (In this respect, Smith [2009] presents a very convincing account of Frege’s doctrines, one that dissipates possible worries concerning the alleged contradiction between antipsychologism and focus on judgments.) Presumably, Wittgenstein’s main reason to reject Frege’s conception of logic centered around the epistemic subject and his epistemic judgments is an exacerbation of Frege’s own antipsychologism; Wittgenstein’s notion of 112 © 2010 The Author. Journal compilation © 2010 Wiley Periodicals, Inc.

psychology is made so broad that it now includes the acts of Frege’s epistemic subject (it includes a whole chunk of linguistics as well, as noted by Potter in section 10.4). Wittgenstein crucially claims in the Notes (C40, p. 288 in the book) that “[t]here are only unasserted propositions. Assertion is merely psychological.” If assertion is indeed merely psychological, and if logic should not concern itself with psychological phenomena, then it follows straightforwardly that logic should not concern itself with assertion or judgment (or with any other human acts, all of which presumably pertain to the realm of psychology). And from this it follows straightforwardly that Frege’s judgment stroke must be logically meaningless: it represents something that simply does not belong to the realm of logic.16 Potter elaborates quite a bit on what he calls Wittgenstein’s “symbolic turn” (in chapter 6 in particular), which rests on the assumption “that there is some correspondence, even if not a perfect one, between the structure of language and the structure of what it represents” (p. 66). Now, given the assumption of symbolic correspondence, the claim to the effect that the judgment stroke is meaningless seems to imply that it is a symbol without a symbolized, that it represents nothing. Indeed, this is what human (and in particular epistemic) acts seem to be according to the ontology of the Tractatus: a whole lot of nothing. They simply do not belong to the Tractarian world. So the exile of the epistemic subject and his acts from the realm of logic, particularly conspicuous in Wittgenstein’s rejection of Frege’s judgment stroke, seems to rest crucially on the claim that the acts of this epistemic subject are “merely psychological,” a claim already made (at least concerning assertion) in the Notes. Potter recognizes that Wittgenstein is here applying his own, extremely wide notion of psychology, not Frege’s, but this is exactly where the crux of the matter lies: why did Wittgenstein feel compelled to adopt such a wide conception of psychology, and accordingly such a narrow conception of logic? The only suggestion that Potter seems to offer by way of an answer to this question is the accusation that Frege “was confused” on the matter, which may have been Wittgenstein’s motivation to re-delineate Frege’s conceptions of logic and of psychology. But what if Wittgenstein himself, not Frege, was confused on the matter?17 What if he failed to fully grasp Frege’s sophisticated notion of an epistemic subject which is clearly not “psychological” in Frege’s own sense of the term? Why is it that logic should not in any way concern itself with human actions? Here, Wittgenstein appears to be begging the question (by redefining the notion of psychology) rather than offering arguments—at least, this is what seems to emerge from Potter’s discussion of the matter. 16. This view is already stated in the Notes, almost verbatim: “The assertion-sign is logically quite without significance” (B32, p. 280). 17. In B32, he continues: “It only shows, in Frege and Whitehead and Russell, that these authors hold the proposition so indicated to be true. ‘|—’ therefore belongs as little to the proposition as (say) the number of the proposition. A proposition cannot possibly assert of itself that it is true.” Although this passage could certainly be interpreted in different ways, to me it suggests that Wittgenstein did not quite understand Frege’s use of the judgment stroke: the point for Frege was never to suggest that a proposition would assert of itself that it is true: it is the epistemic subject that judges a proposition to be true. My hypothesis is that Wittgenstein absorbed somewhat uncritically Russell’s not entirely accurate interpretation of the judgment stroke.

113 © 2010 The Author. Journal compilation © 2010 Wiley Periodicals, Inc.

Further elements to understand the expulsion of the epistemic subject from the realm of logic undertaken by Wittgenstein can be gathered from Potter’s discussion of Russell’s and Wittgenstein’s respective theories of judgment, in chapters 13 and 25. In chapter 13 Potter analyzes the different versions of Russell’s multiple relation theory of judgment, and reports on the several stages at which, according to Russell himself (in correspondence), Wittgenstein put forward serious objections to his theory (and determining what exactly Wittgenstein’s objections amounted to has been the object of much scholarly speculation—see fn. 5 on p. 118). The core of these objections seems to be related to the difficulties in accounting for false beliefs (given that they cannot relate to actually instantiated facts) and in formulating a theory of judgment that would exclude the possibility of judging nonsense: “Every right theory of judgment must make it impossible for me to judge that this table penholders the book” (Notes B33, p. 280). The conclusion of chapter 13 is that, even under its revised versions, Russell’s theory still falls prey to Wittgenstein’s objections. Then, in chapter 25, Potter narrates the possible course of events and ideas leading Wittgenstein to his own later theory of judgment in the Tractatus. The starting point is Russell’s formulation of a judgment as a multiple relation, in the following manner: agent A’s judgment that aRb is analyzed as

J ( A, § R , a , b ) It is a multiple relation theory of judgment because A is related not to the proposition aRb, but rather to each of its three constituents: a, b and the relation §R. But Wittgenstein objected that the form §R here occurs as a term, not as a form, which is wrong (or so he claims). A possible reformulation would then be

J§ R ( A, a, b ) But then, through a series of argumentative steps, Potter argues that this is still not right and concludes: “We conclude, then, that when the judgment relation is fully analyzed, the subject A will not occur as a term in it: so let us remove it” (p. 221). This yields J§R (a, b) as a representation of judgment, and similarly the act of doubting the same proposition would be represented as D§R (a, b), the act of believing as B§R (a, b), etc. So what has happened here? The judging subject has disappeared altogether! We now have acts of judgments without agents performing them, and the same for acts of doubting and believing: agent-less acts. This theory does satisfy Wittgenstein’s condition to the effect that a theory of judgment must exclude the possibility of judging nonsense; but while in his critique of Russell’s theory this desideratum seemed to be presented as a necessary but not sufficient condition, here it is presented as if it was a sufficient condition of adequacy for a theory of judgment. Potter’s conclusion in chapter 25 seems to hit the mark: “Because the judging subject A has disappeared, the kernel of Wittgenstein’s theory of judgment is now indistinguishable from his theory of meaning” (p. 223). (And as we know, Wittgenstein’s theory of meaning is itself completely agent-less: no speaking agents are required in order for relations of meaning to be 114 © 2010 The Author. Journal compilation © 2010 Wiley Periodicals, Inc.

established.) But again, no specific arguments appear to have led Wittgenstein to expel the judging subject from his theory of judgment: rather, the technical difficulties engendered by Russell’s multiple relation theory led Wittgenstein to get rid of an element that he was all too ready to jettison anyway. So even in what is arguably a domain that is rightfully hers, judgment, there is no longer place for the epistemic subject, but it is far from clear that this expulsion has been entirely legitimate; it does not appear to have been properly justified. Thus, the only subject still present in the Tractatus is the metaphysical subject, in the outskirts of the Tractarian world. This metaphysical subject is alone, as Wittgenstein is happy to embrace (a certain form of) solipsism; moreover, the metaphysical subject does not appear to do anything, it does not seem to act in any way.18 This is a hard pill to swallow, and it is not entirely clear from Potter’s book why he felt compelled to expel the much more engaging and dynamic epistemic subject present in Frege’s conception of logic from his own conception of logic.19 5. Conclusions All in all, Potter’s book is a meticulous, thought-provoking work; it represents most certainly a major contribution to Wittgenstein scholarship, in particular insofar as, with its historical and critical focus, the approach synthesizes two usually disjoint methodologies in Wittgenstein scholarship. Moreover, it proposes compelling answers to several open questions concerning the development of Wittgenstein’s thought in this early period. To my mind, there are, however, a few important issues that the book fails to clarify completely, as I have argued here, in particular the question of why exactly Wittgenstein so forcefully rejected Frege’s “epistemological” conception of logic in favor of an “ontological” approach. But of course, this only means that there is more work ahead, and the results obtained in the book have established beyond any doubt that Potter’s methodology is worth being pursued. We can look forward to the next installment.20 Bibliography H.T. Costello, “Notes on Logic,” Journal of Philosophy, 54 (1957), pp. 230–45. E. De Pellegrin (ed.), Interactive Wittgenstein: Essays in Memory of Georg Henrik von Wright. Synthese Library 349 (Springer, 2011). 18. Solipsism in the Tractatus is often discussed as a reaction to Kant’s notion of the synthetic a priori or to Russell’s notion of logical forms as entities (for example, in chapter 11 of McGinn (2006) and Sullivan (1996). But it seems to me that the contrast between Frege’s epistemic subject and the Tractarian metaphysical subject is equally worth looking into in more detail. Here, I have just introduced the issue, which for now shall remain a topic for future work. 19. To be sure, Potter does describe in detail Wittgenstein’s reasons and arguments to reject other Fregean tenets, such as the view that propositions are names of truth-values (section 9.2). 20. Thanks to Stephen Read, Juliet Floyd, Jaap van der Does and Edgar Andrade-Lotero for comments on an earlier draft.

115 © 2010 The Author. Journal compilation © 2010 Wiley Periodicals, Inc.

J. Floyd, “Interpretive Themes,” in E. De Pellegrin (ed.), Interactive Wittgenstein: Essays in Memory of Georg Henrik von Wright. Synthese Library 349 (Springer, 2011). A. Hamilton, “Review of M. McGinn’s Elucidating the Tractatus,” Philosophical Books, 49 (2009), pp. 266–9. M. McGinn, Elucidating the Tractatus—Wittgenstein’s Early Philosophy of Language and Logic (OUP, 2006). M. Potter, “Review of M. McGinn’s Elucidating the Tractatus,” Philosophical Quarterly, 60 (2010), pp. 192–4. N.J.J. Smith, “Frege’s Judgement Stroke and the Conception of Logic as the Study of Inference Not Consequence,” Philosophy Compass, 4 (2009), pp. 639–65. P. Sullivan, “The Truth in Solipsism, and Wittgenstein’s Rejection of the a Priori,” European Journal of Philosophy, 4 (1996), pp. 195–219. B.G. Sundholm, “Inference Versus Consequence,” in T. Childers (ed.), Logica Yearbook 1997 (Filosofia, 1998), pp. 26–35. B.G. Sundholm, “A Plea for Logical Atavism,” in O. Majer (ed.), Logica Yearbook 2000 (Filosofia, 2001), pp. 151–62. B.G. Sundholm, “A Century of Inference: 1837–1936,” in P. Gärdenfors, J. Wolénski and K. Kijania-Placek (eds.), The Scope of Logic, Methodology and Philosophy of Science Vol. II (Kluwer, 2002), pp. 565–80.

116 © 2010 The Author. Journal compilation © 2010 Wiley Periodicals, Inc.