While writing this book I have been generously assisted by institutions, friends and colleagues.

By electing me to a two‐year Research Readership, which relieved me of teaching, the British Academy made the initial research on the Wittgenstein manuscripts easier and more efficient than it would otherwise have been. I am grateful to my college, St John’s, for the many facilities it offers to its Fellows. Its support for research and the pursuit of scholarship is heart‐warming. I am indebted to the Bodleian Library, in particular to the staff of the Western Manuscript Department, for many services. The publishing team at Basil Blackwell Ltd, especially Mr S. Chambers and Mr A. McNeillie, have been most helpful in planning and executing this difficult publishing project. As in the past, so too now, it has been a pleasure to work with them in close cooperation. I am most grateful to Miss Jean van Altena for the excellence of her copy‐editing.

Professor N. Malcolm, Dr S. Mulhall, Professor H. Philipse, Professor J. Raz, Mr B. Rundle, Professor S. Shanker, Mr T. Spitzley and Professor T. Taylor kindly read and commented on various drafts of essays or exegesis. Their criticisms, queries and suggestions were of great assistance. I am especially indebted to Dr H. J. Glock and to Dr J. Hyman, whose comments on essays and exegesis alike were invaluable. Dr Glock and Mr Spitzley kindly checked my German transcriptions and translations. Participants in the university seminars which I have given together with Dr G. P. Baker over the past three years have contributed greatly to the clarification of my thoughts. Their questions were challenging and a stimulus to further efforts. I am most grateful to them all, but especially to Dr O. Hanfling, who both curbed some of my excesses and spurred me on to improve my arguments.

For various reasons it was not feasible for Dr Baker to join me in writing this third volume of Analytical Commentary.¹ However, despite occasional disagreement in interpretation and deeper disagreement over nuance (and it is the chiaroscuro that finally makes the sketch), he read the whole manuscript and joined me in giving the university seminars. His painstaking and helpful criticisms as well as his constructive suggestions saved me from error again and again.

Finally, I am, as before, indebted to the Wittgenstein executors for permission to quote from the unpublished Nachlass. Professor G. H. von Wright has, as always, been unstintingly generous in putting at my disposal the results of his extensive bibliographical research on the Nachlass.

Volume 1 of this Analytical Commentary on Wittgenstein’s Philosophical Investigations –– Wittgenstein: Understanding and Meaning (1st edition 1980, 2nd edition 2005) laid the groundwork for understanding the trajectory of Wittgenstein’s thought in his second great masterwork. It explained and gave due prominence to the Augustinian conception of language as informing a multitude of philosophical theories about the nature of language, all of which Wittgenstein aimed to undermine. Conceiving of the essential function of words as naming and of the essential function of sentences as describing stands in the way of an unprejudiced view of the manifold techniques of using words and of the diverse functions of sentences in the stream of human life. These misconceptions give rise to philosophical mythologies of the so‐called name‐relation, logically proper names, sentence‐radicals and semantic mood operators, of the idea of a truth‐conditional theory of meaning for a natural language, of the notions of determinacy of sense and the general propositional form. The Augustinian conception of language generates misguided pictures of the relation between language and reality, sentence and fact, proposition and truth‐maker, and of ostensive definition as forging a connection between word and world. This in turn contributes to the pervasive illusion that grammar is answerable to reality, or that it reflects, and must reflect, the essential structure of the world. Against this backdrop of Wittgenstein’s demythologizing, his radical conception of the nature, scope and limits of philosophy was displayed. He held philosophy to be therapeutic and descriptive, not theoretical and hypothetico‐deductive. It destroys idols, but does not replace them. It is a quest for an overview of a segment of logical grammar — of the conceptual network of our language — that gives rise to conceptual questions. The purpose of that quest is the solution or dissolution of the questions. Achievement in philosophy consists in the resolution of philosophical questions, not in the acquisition of new empirical or putatively metaphysical information that provides answers to them. Understanding is indeed attained; but it consists in arriving at a clear vision of what is known and familiar, namely the common grammatical, conceptual, connections of our language. Philosophy contains no theories on the model of the sciences. In its questions, methods, and results, philosophy is wholly distinct from science and no scientific discovery can resolve the a priori questions of philosophy, any more than a discovery in physics can resolve an a priori question in mathematics.

Once this had been explained and clarified, it was possible to put Wittgenstein’s discussion of understanding into the right perspective. Meaning, explanation of what a word or utterance mean, and understanding constitute a triad of key concepts in philosophical investigations into language and the nature of linguistic representation. Reversing the direction of fit between these concepts that was (and still is) presupposed by the prevailing philosophical tradition, Wittgenstein elaborated the consequences of the grammatical propositions that meaning is what is given by an explanation of meaning, and that it is what is understood when the meaning of an utterance is understood. Understanding (which is akin to an ability rather than being a mental state) and the criteria of understanding assume a dominant role in his description of the network of grammar in this domain. Clarification of the internal relations between meaning, understanding and explanation also illuminates their complex connections with truth, evidence, justification, definition, rules of use, grammatical propositions and so forth. This network of conceptual connections is put to use in elaborating an anthropological, ethnocentric conception of language as opposed to the calculus conception of language that Wittgenstein had embraced in the Tractatus.

Wittgenstein: Rules, Grammar and Necessity, Volume 2 of the Analytical Commentary (1st edition 1985, 2nd edition 2009) examined the complexities and problematic interpretation of Investigations §§185 – 242. Wittgenstein’s conception of a rule needed much clarification, especially against the backdrop of the ideas of syntactic and semantic rules advanced by logicians, theoretical linguists and philosophers throughout the twentieth century, and especially in the last thirty years of the century. His animadversion to hidden rules of depth‐grammar (prominent in the Tractatus and among contemporary linguistic theorists) needed explaining. His insistence that one cannot follow a rule without knowing and understanding the rule one is following, that a rule is a reason for doing something and not a mere regularity, that rules do not compel but justify, needed elucidation. His insight that in order for one to be following a rule, the rule and its consideration need not enter into one’s action, required clarification: all that is necessary is that one cite it, or would cite it if challenged, ex post actu, as one’s reason for acting in accord with the rule. Wittgenstein’s elucidation of following a rule and of the connection between rules and practices had to be made clear in order to explain how an internal relation between rule and what accords with it is forged. This led naturally to an examination of Wittgenstein investigations into the internal relations in general, into the nature of logical, mathematical and grammatical internal relations, as well as to his demystification of necessity and demolition of metaphysics.

Volume 3, Wittgenstein: Meaning and Mind (1st edition 1990) is here presented in an extensively revised 2nd edition, which brings it to the same level as that of the previous two revised volumes. It takes the Commentary forward from §243 to §427. These sections are no less controversial than the preceding ones. To be sure, the rocky ground already traversed should have taught one much. But as one plunges into the tropical undergrowth of the private language arguments, it is all too easy to lose one’s bearings. The path is overgrown with prevalent misinterpretations and misunderstandings, and dark distorting shadows are cast across it by the disposition to extract theories from Wittgenstein’s descriptions and grammatical statements. Following in his footsteps is not made easier by his ever increasing tendency to dart down a side‐path in order to show one a bog here and quicksands there.

The text of the Investigations §§243 – 427 is, as in the 1st edition, divided into five chapters. As in the previous volumes, there is nothing sacrosanct about this division. Other ways of segmenting the text can readily be imagined and justified. The divisions were chosen in part for ease of exposition and in part because they correspond, in my judgement, to fairly evident thematic changes in Wittgenstein’s book.

Each chapter of exegesis is preceded by an introduction, which surveys the argumentative development of the sequence of Wittgenstein’s numbered Bemerkungen (remarks) — referred to as ‘sections’ to avoid ambiguity — and explains the structure of the argument. This is supplemented by tree diagrams that represent the relationships between the various sections as I see them. Again, others may, with justification, prefer alternative ways of structuring the chains of argument. At the end of each such introduction there is a table of sources. The 1st edition of this volume of exegesis was reasonably thorough in its pursuit of the manuscript and typescript sources of Wittgenstein’s Bemerkungen in this part of the book. Nevertheless, it was done without the benefit of the Bergen electronic edition of Wittgenstein’s Nachlass which had not yet been published in the years in which I wrote this commentary (1987 – 1990). In this extensively revised edition I have been able to supplement my lists of relevant sources by making use of the search engine of the second Bergen edition. All Wittgenstein scholars are much indebted to Alois Pichler and his team of researchers for their labours in producing their excellent edition of the Nachlass and in devising the wonderful search engine. Of course, an element of judgement is involved in my selecting sources for each section in the text, and no doubt there are some passages that I decided not to be worth registering which another person might, with equal right, think to be pertinent.

Like the 2nd edition of Wittgenstein: Understanding and Meaning, this volume: Wittgenstein: Meaning and Mind is being published ab initio in two separate parts: Part I: Essays and Part II: Exegesis §§243 – 427. In this volume (Part II: Exegesis) I have inserted text‐boxes in the appropriate locus in the exegesis of Wittgenstein’s arguments to indicate the accompanying essay and its table of contents that occurs in Part I. This will indicate to anyone using the exegesis for the close study of Wittgenstein’s book that they might wish to pause in order to read the relevant essay that offers a survey of Wittgenstein’s treatment of the subject under discussion at this point in the exegesis.

In writing this revised edition of exegesis I have been able to correct what now seem to me to have been errors of judgement and to improve my explanations of Wittgenstein’s complex reasoning and often bewildering trains of thought. I have occasionally further elaborated the original exegesis, and have sometimes simply removed paragraphs or pages and redrafted them completely.

As has been the practice throughout this Commentary, in both 1st and 2nd editions, I have but rarely mentioned by name the authors of misunderstandings, but have discussed and attempted to rectify such misunderstandings as merit attention. Similarly, I have only in a few cases indicated the authors of misguided criticisms, but have tried to refute the criticisms themselves, presented, to the best of my abilities, in their most favourable light. The principle that I have followed in selecting misinterpretations and mistaken criticisms is that only those that are truly tempting and likely to recur in the future are worthy of attention. It is from these that students of Wittgenstein can learn something. Even though it is learning down which paths not to walk, that too is a valuable lesson. Most of the criticisms of Wittgenstein are my own; those that are not are attributed to their author.

As the revised 4th edition and translation of the Investigations by Joachim Schulte and me was published in 2009, I have been able to remove many suggestions for improved translation from the exegesis in the 1st edition of this volume of the Commentary. All quotations from the text of the Investigations are now taken from the 4th edition. All page references to what used to be called Philosophical Investigations, Part 2 are now to numbered sections of Philosophy of Psychology — A Fragment. I have also benefited from, and have occasionally quoted, secondary sources of conversations with Wittgenstein, students’ notes of Wittgenstein’s lectures, and memoirs of Wittgenstein that had not yet been published when I wrote the 1st edition.

1. Wittgenstein’s published works

The following abbreviations, listed in alphabetical order, are used to refer to Wittgenstein’s published works.

Reference style: all references to Philosophical Investigations are to sections (e.g. PI §1), except those to boxed remarks (previously Randbemerkungen (notes below the line)) on various pages. Reference to these is ‘boxed remark following §n.’ followed by page reference to the note in the first two editions. References to Philosophy of Psychology — A Fragment are to numbered remarks in PI, 4th edition, followed by page number in the first two editions in parentheses (e.g. PPF §174 (p. 202)). References to other printed works are either to numbered remarks (TLP) or to sections signified ‘§’(Z, RPP, LW); in all other cases references are to pages (e.g. LFM 21 = LFM, page 21) or to numbered letters (CL); references to The Big Typescript are to the original pagination of the typescript as given in the Bergen electronic edition of Wittgenstein’s Nachlass (Oxford University Press, Oxford, 2000), and in the margins of the Luckhardt/Aue edition.

2. Derivative primary sources

3. Nachlass

All references to other material cited in the von Wright catalogue (G. H. von Wright, Wittgenstein (Blackwell, Oxford, 1982), pp. 35 ff.) are by MS or TS number followed by page number (‘r’ indicating recto, ‘v’ indicating verso) or section number ‘§’, as it appears in the Bergen electronic edition of Wittgenstein’s Nachlass.

In the case of the first manuscript draft of the Investigations, MS 142 (the so‐called Urfassung), references are to Wittgenstein’s section number (‘§’), save in the case of references to pp. 77 f., which are redrafts of PI §§1 – 2 and to pp. 78 – 91, which Wittgenstein crossed out and redrafted on pp. 91 ff., subsequently assigning them section numbers in the redrafts alone.

Manuscripts

MS 105 – 22 are eighteen large manuscript volumes written between 2 February 1929 and 1944. They were numbered by Wittgenstein as Vols I–XVIII. In the first edition of this commentary they were referred to by volume number, followed by page number (e.g. ‘Vol. XII, 271’). Since then it has become customary to refer to them by von Wright number alone. Here I have followed this practice. The von Wright number is followed by the pagination in the Bergen edition of the Nachlass.

‘MS 114 Um.’ refers to Wittgenstein’s pagination of the Umarbeitung (reworking) of the Big Typescript in MS 114. The Umarbeitung begins on folio 31v of MS 114 and is paginated consecutively 1 – 228.

Typescripts

All other typescripts are referred to as ‘TS’, followed by the von Wright number and pagination as in the Bergen edition.

The successive drafts of the Investigations are referred to as follows:

TS 220 is the typescript of the Early Draft (Frühfassung (FF)) of the Investigations, referred to in the 1st edition of this Commentary as ‘PPI (‘Proto‐Philosophical Investigations’), dictated from MS 142 (the Urfassung (UF)).

TS 226_R is Rhees’s pre‐war translation of TS 220 §§1 – 116, referred to in the 1st edn of this Commentary as PPI(R).

TS 227a and 227b are the two surviving carbon copy typescripts of the Investigations (the top copy having been lost).

TS 238 is a reworking of TS 220, §§96 – 116, with renumberings, deletions, corrections and additions in Wittgenstein’s hand, referred to in the 1st edn of this Commentary as PPI (A).

TS 239 (the Bearbeitete Frühfassung (BFF)) is a reworking of TS 220.

ZF is the reconstructed Intermediate Draft (Zwischenfassung) of the Investigations, previously known as the Intermediate Version, and referred to in the 1st edn of this Commentary as PPI(I).

In transcriptions from the Nachlass I have followed Wittgenstein’s convention of enclosing alternative draftings within double slashes ‘//’.

4. Abbreviations for the other volumes of An Analytical Commentary on the Philosophical Investigations

All references to these are of the form ‘Volume’, followed by the volume number and the quoted title of an essay in the designated volume (and, in the case of split volumes, to part I or part II. References to the exegesis are flagged ‘Exg.’, followed by section number prefixed with ^‘§’ or ‘b.r.f. §’ (in the case of the boxed remark following a numbered section).

5. Abbreviations for works by Frege

6. Abbreviations for works by Russell