Philosophy

The part of philosophy known as the philosophy of language, which includes and is sometimes identified with the part known as semantics, is as diverse in its problems and viewpoints as any part of philosophy. As in the case of other philosophical fields, many of the problems of current interest have long histories. Also, as elsewhere, problems in the philosophy of language are interrelated with problems in other areas notably epistemplogy, logic, and metaphysics, but also for example the philosophy of mind and the philosophy of science. I will not try in the space of the present overview to present any serious account of the history of the subject or a general introduction to its problems.¹ I will try to give only a brief account of its recent history of and some introduction to the problems that are dealt with in the papers that are here included. First, some preliminary remarks need to be made.

The time is ripe for a collection of papers from the various disciplines that deal in their several ways with one or another aspect of meaning. I think inquirers in all these fields - in philosophy, psychology, linguistics, etc. - would each agree that there were significant questions about meaning with which they themselves didn’t deal and which perhaps properly belonged to another discipline. In this overview, I will try only to give some impression of the recent history of philosophical investigations concerning meaning and reference and my personal assessment of the present situation. I must emphasize the fact that this will be a personal assessment: in the, area of philosophy of language perhaps as much as in any other philosophical field, chaos reigns, little is agreed on, new methods are to be eagerly seized upon, and we have hardly begun. Philosophy has historically both taken inspiration from and made contributions to other disciplines, ones not perhaps at the time in question thought of as ‘philosophical ’: I think that this is very liable in the future to prove to have been true of the present time in the philosophy of language.

A preliminary statement about the provenance of the articles included should also probably be made. They all belong to that segment of philosophical opinion known as analytic philosophy.² Thus there are no selections included from the writings of phenomenologists or existentialists, for example, though of course philosophers of these and other viewpoints have written in recent times about language or philosophical problems concerning language.³ Besides obvious considerations of space, in justification of leaving out certain well-known writers in the philosophy of language, it might I think be said that philosophical writings in the analytic tradition are more accessible to laymen, whether or not initially more interesting to them. And my own opinion is that, in this area at least, the writings of analytic philosophers are as liable to be of interest to workers in other fields as those written from any other current philosophical viewpoint. This is, of course, not a surprising situation, in view of the fact that the hallmark of analytic philosophy is precisely to approach philosophical problems and theories in terms of the language used to formulate them.⁴

As a final preliminary, it should be said that it has been decided that the selections to be included should fall within semantics in the narrow sense, i.e. problems of meaning and reference. Thus a number of topics in the philosophy of language which have recently been discussed extensively in the philosophical literature do not appear at all or do so only in passing. Among these topics are presupposition, contextual implication, truth, and the analytic-synthetic distinction. Especially, a large and important literature not represented here has to do with the analysis and classification of ‘speech acts things one does by or in saying something, including how such acts relate to meaning and reference.⁵ But one must stop somewhere.

The division of semantics in the narrow sense into problems to do with meaning and problems to do with reference (or referring) might almost be said to be traditional in contemporary philosophy. This does not mean, of course, that there is much agreement over anything that comes under either heading or even over the distinction itself. It is better to try to convey the latter by citing examples of problems on either side, rather than by explicit distinction. Grouped under the topic of meaning are such questions as what it is for a linguistic expression to mean a particular thing or to have a particular meaning, what it is for one expression to be synonymous with or a paraphrase of another, and what it is for a sentence (or statement) to be analytic, i.e. true in virtue only of its meaning (or the meaning of a sentence which expresses it). Grouped under the topic of reference are such questions as what it is for an expression to refer to something, what it is for a person to do so, and how these two are related, what sort of meaning certain referring expressions have, and whether complete sentences have references at all (candidates: facts, propositions, truth-values), whether some expressions have only references and not also or instead meanings, etc. Also, naturally, there are questions and views which involve both meaning and reference and the relations between the two: for example, one historically important type of semantical theory attempts to provide a basis for understanding language and its use entirely in terms of reference (naming, denoting, etc.), i.e. without using a distinct concept of meaning as well.⁶

Many of these inquiries might be regarded as simply the pursuit of the venerable philosophical task of clarifying a matter which is puzzling or problematical and which no one else is concerned to clarify, i.e. (if you will) as preliminary, quasi-scientific speculation concerning an essentially scientific question which is not at the moment capable of being treated as such. Philosophy has traditionally been the repository of such problems. But, from the history of semantics, it is apparent that other philosophical motives have been present from early times. For example, certain epistemological and metaphysical problems have been regarded as involving the question of what meaning is. The fact that the answers to some of these questions are not clear even now would itself be sufficient to explain why philosophers are still concerned with analyzing the concept of meaning. Consider just the ancient ‘problem of universals’: one aspect or version of this problem is (roughly and briefly) the question whether from the obvious fact that certain types of linguistic expressions, so-called ‘general terms’ (common nouns and most verbs and adjectives) occur, or occur in certain ways, in sentences used to express true or at least coherent statements about different things, it somehow follows that there are certain entities, perhaps abstract, which correspond not to the things the statements are about but rather to what is said about them. This is red, that is red, so is there redness as well? Essentially this same question has been discussed by both Plato and Quine and by both because both were pursuing ontology: a special case of what there is is what there has to be if language is to work in the way it does, so of course how it works and what there has to be, given that it works that way, become questions of interest to ontologists.⁷ This kind of situation also arises in epistemology, logic, and other fields.

I will now sketch briefly the backgrouhd in the recent history of analytic philosophy which seems to me necessary or desirable in approaching the essays in philosophical semantics included this collection. There would probably be general agreement among philosophers that the late Ludwig Wittgenstein (died 1951) and John L. Austin (died i960) are at least among the most important analytic philosophers in their influence on thinking about language, if indeed they aren’t the two most important. I will begin by discussing their conceptions of meaning, which in the case of both involves also discussing referring (naming, etc.) in that both were opponents of what has been called the denotative or entitative theory of meaning - though it is primarily Wittgenstein whose influence was the greatest along this line.

The concept of meaning in recent philosophy is an interesting story for two reasons: first, what separates several of the recent movements in philosophy is primarily differences that have to do with the concept of meaning and, second, one or two quite novel accounts of meaning have been put forward and widely accepted recently. Briefly, as to the first, the recent movements of pragmatism, logical positivism, and ‘ ordinary language * philosophy are separated, among other things, by different conceptions of meaning.⁸ As to the second, I refer to ‘ one or two ’ conceptions of meaning because, although it is Wittgenstein’s remarks on meaning that I have chiefly in mind, his writings are difficult and have been differently understood.⁹ In Gilbert Ryle’s well-known paper ‘The Theory of Meaning'¹⁰ one gets a pretty good idea of the background in preceding philosophy of Wittgenstein’s work, what led up to his work, and some of the prevailing sorts of view - along with some indication of what he, Wittgenstein, wanted to say. In Ryle’s ‘ Ordinary Language’,¹¹ one gets more the form his views have taken w'hen others elaborated them - I cite Ryle because he is a leading figure here - and I think the form in which they tend to be held by others. The latter, the popular form of Wittgenstein’s ideas, can (I think not too unfairly) be regarded as a sort of rigidified, simplified form of Wittgenstein’s views. I will treat of the genuine article first and then turn briefly to the more popular version.¹²

Wittgenstein’s handling of the concept of meaning in the Philosophical Investigations and the Blue and Brown Books has in common with the popular version of it two things: (ɪ) that it is essentially negative, in that it denies what is alleged by its adherents to have been a widespread, indeed practically universal tendency, in philosophical theories of meaning, and (2) that it proposes to substitute for this what it represents as the ordinary conception of meaning, the everyday concept in daily use, which is regarded as not suffering from the traditional sort of mistake. (It is over (2) of Wittgenstein’s ideas that the genuine and popular versions primarily differ, i.e. over what the ordinary conception is regarded as being.)

The traditional mistake is described as thinking of the meaning of a word or phrase as an object or thing or entity.¹³ Two questions arise here: (i) What is it to conceive of the meaning of a word or phrase as an object (thing, entity)? and (ii) What’s wrong with thinking of the meaning of a word or phrase in this way? (i) This question arises because it is not clear exactly what Wittgenstein, Ryle, Austin, et al., other philosophers, or, for that matter, the ordinary man would count as an ‘ object ’ or ‘thing’ or ‘entity’, i.e. where the dividing line is between ‘objects’ and non-objects, etc.; and it is clear that these words are pretty comprehensive ones. I think, actually, that this way of putting (i), though common, is not very perspicuous. What the point seems to come to is this: that there isn’t anything X or any such thing as the ∅ such that, for a given word or phrase W, the meaning of W is X or the ∅. This reading seems to accord both with what Wittgenstein and the others say and with what they are attacking. For, as to the latter, it does seem to be a feature of many theories of meaning that had been put forward by philosophers and others before (say) 1935, that they did purport to provide an X or such a thing as the (f) which allegedly answered to the description that the meaning of a given word W was X or the ∅. And it was apparently thought to be part of being a theory of meaning to provide this thing X or the ∅ in a general way, so that the theory entailed or consisted in saying what kinds of things meanings were, there being perhaps several kinds according to the theory.

The alleged mistake is suggested to be that traditional theories always provided such an answer to ‘What is the meaning of W?’ (or ‘What does W mean?’) that the answer could, in principle, be put in the form ‘X’ or ‘the ∅’, i.e. be put in the form of naming or identifying the thing X or the ∅ which was (the very same thing as) the meaning of W.¹⁴ This assertion seems to me as a matter of history to be correct.

The allegation is thus that this traditional feature is an error, and in fact an error due to a (quite natural) misconception of ordinary talk about meaning. This sort of error might be called ‘ using a misleading model ’ (i.e. to understand meaning) or ‘ being misled by a grammatical similarity ’ (viz. of locutions in ‘ meaning ’, ‘ means ’, etc., to other locutions) - with the result that the concept, depth-grammar, or logic of ‘meaning’ or of the concept of meaning is misunderstood. That is (the suggestion is), philosophers think there must be an answer ‘X’ or ‘the ∅’ to the question ‘ What is the meaning of W? ’ or ‘What does W mean? ’, because they take it (perhaps unconsciously) to be like ‘Who is Johnny’s teacher?’ or ‘Who teaches Johnny?’, which may have the answer ‘Mary’ or ‘the girl in the corner’.

Thus, in belief, the error alleged is that of thinking that one can name and/or identify (the thing that is) the meaning of W, or that one can explain the meaning of W by naming or identifying the meaning of W as X or as the ∅ .It is, for example, the error of thinking that ‘W means X' is like ‘Mary teaches Johnny’.

The question remains: (ii) why, according to these philosophers, is it an error to suppose that one can name (as X) or identify (as the f) the meaning of a word or phrase W? I think they would answer simply that there is no such nameable or identifiable thing (or things or sets of things) which is the meaning. This may seem just a denial of the assumption, but the support for it consists in (a) an account or sketch or set of reminders of how we ordinarily handle the concept of meaning (e.g. of what is involved in explaining the meaning of a word, how we explain the meaning of a word), together with (b) an account or sketch of what these ways of handling the concept presuppose. Thus, as to (a), Wittgenstein begins the Blue Book by immediately changing the subject from ‘What is the meaning of a word?’ to ‘What is an explanation of the meaning of a word? ’ And similarly Austin in ‘The Meaning of a Word ’, at the beginning of Part II. Both philosophers begin by distinguishing verbal explanantions from what Wittgenstein calls ostensive ones or what Austin describes as ones involving imagining or experiencing situations related to the meaning of the word in question. Wittgenstein examines ostensive definition at length, because verbal definitions presuppose that the pupil can already understand some language (viz. that in terms of which the word being explained is being explained).

Other relations besides the one allegedly denoted by ‘means’ are no help to traditional theories: I mean such things as naming, standing for, referring, denoting, connoting, etc.¹⁵ These are no help because the ordinary concepts expressed by these words won’t bear the weight of a general semantical theory and because philosophers have used them as they have used the notion of meaning, i.e. in such a way that Wittgenstein’s and others’ considerations apply. As I have said, most or all traditional theories have taken one or more such relations as basic for semantics.

So much for the traditional error which is alleged to characterize most or all earlier theories of meaning in philosophy.

What it is proposed to substitute for the conception of meaning as some sort of entity (in the sense explained above) is what has been called the Doctrine of Meaning as Use, this being taken to state the ordinary conception of meaning. In the form in which it is presented in the popular version of Wittgenstein’s views on this subject, this doctrine says (as in Ryle’s ‘ Ordinary Language ’) that the meaning of a word or phrase is its ordinary use, i.e. its standard use (not necessarily its everyday use), in the sense of the way it is used in the language. I don’t believe that anything so specific is to be found anywhere in Wittgenstein’s published works, though he does repeatedly recommend, as a point of philosophical method, ‘looking to’ or ‘asking for’ the use rather than the meaning; and it is clear that the reason he does so is the tendency philosophers have to fall into the traditional error already described.¹⁶ Also, some of Wittgenstein’s remarks seem to sanction such a theory of meaning, e.g. (the one most cited):

§43. For a large class of cases - though not for all - in which we employ the word ‘ meaning ’ [Bedeutung] it can be defined [or, explained] thus: the meaning of a word is its use [Gebrauch] in the language.¹⁷

But I think probably that Wittgenstein, like Austin in ‘The Meaning of a Word ’, was just intending to remind us of how (as he thought) one used ‘meaning’ (including explaining meanings), i.e. what ‘meaning’ and ‘means’ mean, just as Austin was ‘explaining the syntactics’ and ‘demonstrating the semantics’ of ‘meaning’, in accordance with his general account of what it is to explain the meaning of a word.

As we have just seen, denotative or entitative theories of meaning, i.e. those which attempted to provide, by a general formula, some entity or thing as the meaning of a linguistic expression (or, taking this as basic, to reduce all other varieties of meaning to it), have tended to dominate recent thinking in philosophical semantics and its history as well. Hence concepts like referring, denoting, naming, etc., on the referring side of the fence, have come in for considerable study. There is a series of articles of great philosophical interest, stretching from Frege’s ‘ On Sense and Reference’ of 1892¹⁸ and Russell’s ‘On Denoting’ of 1905, which is directed at Frege’s views (as well as others’), to the selections in the present volume.

This literature has been expounded and evaluated by Leonard Linsky in his book Referring,^a which takes one up to the present articles. These earlier papers - not included here because of their easy availability elsewhere and because they can, though still well worth studying closely, reasonably be regarded as part of the background of contemporary research, rather than themselves part of it - together with Strawson’s earlier ‘On Referring’ (1950)^b have as their essential points the following, which I try to place in some perspective with respect to the author’s own views. Frege’s semantical theory was based on two relations, one of the meaning sort and one of the referring sort. He held that expressions of a natural language had the former relation, which he called ‘ expressing (a sense) ’  unless they were nonsense or had no ‘ meaning ’, but even so might not ‘ stand for ’ or ‘ refer ’ (bedeuten, bezeichnen) any object or entity, and that it was a desideratum of a scientific language that it should guarantee a reference, as well as a sense, to all of its expressions - including sentences, the reference of which (if any) he held to be a truth-value, the True or the False. Russell, though holding a similar view earlier, came in 1905, in ‘On Denoting’, to regard Frege’s semantics as incoherent and, apart from this, involving an unnecessary postulation of types of entities, viz. the abstract entities called ‘senses’ which Frege held to be expressed by the meaningful expressions of any language. Russell advanced instead a semantics in which, allegedly, only a naming relation (a ‘referring ’ concept, then) was fundamental; expressions which could not be regarded as names (e.g. because there wasn’t any such thing as what they might have otherwise been regarded as naming) were analyzed, i.e. sentences containing them were explained to really meaiff the same as other sentences not containing them but rather containing only Russellian names. An important or crucial tool in carrying out the analyses required to make this account plausible was his Theory of Definite Descriptions, which in his best known example required analyzing ‘The present king of France is bald’ as really meaning ‘There is one and only one present king of France and he is bald ’, so that the former statement can be regarded as false (rather than, botheringly, meaningless) because there is no pre¬ sent king of France, as the latter version clearly states there to be. Strawson’s article ‘ On Referring’ sided with Frege in holding that meaning was not referring and that ‘ The present king of France is bald ’ (or rather the statement one can make with it, which he regarded as distinct from this sentence) was without a truth-value, neither true nor false, since there was no king of France, as it (he said) presupposed, rather than stated or entailed (as Russell had said).

Note that both Russell and Frege supplied entities that were the meanings of linguistic expressions in their semantical theories, though Strawson did not: Frege, certainly, since he assumed a special type of entity, his ‘senses’, as the meanings of expressions, as well as the ‘ references ’ of those expressions that had them; but Russell too, at least in the sense that the only semantical properties and relations he was at this period prepared to recognize involved an expression as one relatum and what was certainly an entity as the other: the nominata of Russellian names are the only candidates for ‘ meanings ’ that he supplies in this theory. It is theories of this rather general sort, which includes both Frege’s and Russell’s, that Wittgenstein, Austin, Ryle, et al. attacked.

Writings of Wittgenstein and Austin themselves are not included, since unfortunately their work ended some years ago. ¹⁸ This is not by any means to say that everything that can be got out of their writings has been got out already. But it is true that their work forms the background for much current research. Significantly, in connection with the present collection, the remaining major influence on current research in the philosophy of language is linguistics - specifically and primarily, transformational or generative grammar.¹⁹ For this reason, several of the selections in the linguistics and psychology of this collection are of direct interest to the concerns of some contemporary philosophers of language.²⁰

It should also be mentioned that one particular recent development in the philosophy of language - or perhaps rather in analytic philosophy itself - viz. the theories of J. J. Katz, is in its very conception integrated with and based on generative grammar. ²¹ A somewhat different approach, which seems to rely on specific syntactical results and conjectures about words important to the problem under consideration, more than on speculations as to the form semantic theory should take, is practiced by Zeno Vendler, e.g. in his essay ‘Singular Terms’ reprinted here²².

Having tried to give some indication of their background in the recent philosophy of language and some introduction to the problems which are dealt with in the selections here included, I will now say something more particularly about each of the latter. Those included under the rubric Reference constitute the leading articles from the philosophical periodical²³ literature on referring since interest was renewed in the subject by Strawson and Geach in the 1950s.²⁴

Linsky’s ‘ Reference and Referents *, a chapter from his book Referring, makes a number of important negative points while also advancing some interesting positive theses. Among the former is his emphasis (already present in Strawson’s ‘On Referring ’) on the fact that it is primarily speakers, rather than linguistic expressions, that refer to things; and among the latter, positive theses is the thesis that the only ordinary sense in which expressions refer is derivative from the more basic sense in which speakers or groups ofspeakers refer to things. Strawson’s 1964 Theoria article, ‘Identifying Reference and Truth-Values’, presents his view of the dialectic be¬ tween his early type of view and its opponents and a new account of the existential presuppositions of a speaker’s using uniquely referring phrases, which is more liberal than his earlier 1950-4 view, which however is a special case of the new account. Briefly, the new account makes these presuppositions relative to the conversational context: depending on what referring expressions would occur in stating the question at issue or topic of the conversation, a statement formulated with a sentence involving a uniquely referring expression suffering from ‘ radical reference-failure ’ (i.e. denoting a non-existent thing) may or may not be without a truth-value; it is not necessarily without one, as on the earlier account.

Donnellan’s ‘ Reference and Definite Descriptions ’ pursues still further, and more positively, certain themes already broached or mooted by Strawson and Linsky, while emphasizing the difference between two uses of expressions which sometimes have a uniquely referring use, viz. between what he calls the attributive and the referential use of such expressions, e.g. between the use of ‘ the murderer ’ in ‘ The murderer, whoever he is, must be insane’ and ‘The murderer, i.e. Jones, must be insane ’, respectively. He attempts to sort out the wheat from the chaff in the previous literature and to distribute the wheat properly between the two uses he distinguishes. Vendler’s ‘Singular Terms ’, to which reference has already been made in connection with new methods, investigates in some detail the syntax of the type of expression Russell called definite descriptions, finding that there are certain constraints on their uniquely referring use statable in terms of the overall syntax of the conversational context or this together with certain presuppositions concerning the context of utterance. John R. Searle’s ‘The Problem of Proper Names’, a chapter from his recent book Speech Acts, gives his account of the use of proper names as referring expressions, which is related to that of Wittgenstein in the Philosophical Investigations, and criticizes the earlier accounts of Frege and Russell. It seems to me that the cumulative effect of these papers is to confirm some contentions in the previous literature, often with a difference, and to discredit others - at least when it is the use of referring expressions in ordinary language that is in question. It is pretty clear, even after these papers, that further investigation of what is involved in referring, ordinarily so-called, or of what useful technical analogues of it might be developed, is to be desired.

The Quine selection is an excerpt from his John Dewey lectures, ‘Ontological Relativity’, ²⁵ in which he continues to pursue certain topics emphasized in his book Word and Object and elsewhere.²⁶

In our selection, he contends that, besides the indeterminacy in principle of the answer to the question whether an expression of another language has been translated correctly, there is a similar indeterminacy in principle concerning the answer to the question what an expression of another language refers to - hence the ‘ inscrutability of reference Quine’s penetrating examination of the epistemological foundations of our understanding another’s language is pressed even farther here.

The papers by Grice and Ziff form a group. Grice’s ‘ Meaning ’, originally published in 1957, distinguished a ‘ natural ’ from a ‘ nonnatural ’ sense of ‘ meaning ’ and announced a program of explaining the latter, including the meaning of a linguistic expression, in terms of the notion of the intentions of the utterer of the expression, viz. in terms of what beliefs the utterer intends to induce by the utterance in his hearer. Ziff’s paper, while discussing Grice explicitly, actually offers a number of general criticisms of the kind of analysis of meaning put forward by Grice.²⁷ In using intentions to explicate meaning, Grice’s approach is of course quite different from that of Quine. Gilbert Harman’s paper, ‘ Three Levels of Meaning ’, distinguishes different sorts of theories among those that have been offered as ‘ theories of meaning ’. Harman refers to three ‘ levels ’ of meaning:

the first would offer an account of the use of language in thinking; the second, an account of the use of language in communication; the third, an account of the use of language in certain institutions, rituals, or practices of a group of speakers.

He holds that there are certain relations of priority or presupposition among theories of these different levels and that failure to distinguish them has introduced confusion into philosophical semantics. In detailing the last, he discusses among others the views of Quine, Grice, Fodor and Katz, Alston, Davidson, and Chomsky. The Wiggins-Alston symposium, here published for the first time, is devoted to our ordinary, everyday methods of distinguishing different senses or meanings of a linguistic expression, what exactly they are, and what philosophical presuppositions and consequences they have.

I feel I really should comment on one particular feature of the current situation in philosophical research, one not unrelated to some lines of research in other disciplines. This is the apparent disregard of the Meaning-as-Use Doctrine of Wittgenstein, Austin, Ryle, et al. For a decade or more, this doctrine in one or another of its versions tended to dominate philosophical thinking about meaning. Yet the later, most recent developments of Grice, Ziff, Katz-Fodor, and others seem, at least, to ignore it and to offer accounts of meaning which would yield a statement of necessary and sufficient conditions for the application of a given term from any correct statement of what it meant. (It is, however, true that they do not purport to supply some entity which is the meaning of a term.) The relevance of this observation seems to me to be that it is, in effect, simply assumed without discussion that the rather radical positions of Wittgenstein and Austin are incorrect. And, perhaps more importantly in the present state of our knowledge, the question of what sort of formal representation of meanings would be appropriate if Wittengenstein and Austin were right is not pursued.

Despite this last worry of my own, it is, I think, quite clear from the present selections from the recent philosophical literature and the original essays published here for the first time that interdisciplinary communication between philosophy and other fields dealing with meaning, especially the ones represented in this collection, is already bearing fruit and promises even richer yields to come.

¹ The history of philosophy of language is comprehensively treated in Norman Kretzmann’s article ‘Semantics, History of ’ in The Encyclopedia of Philosophy, ed. Paul Edwards. A recent introduction is W. P. Alston’s Philosophy of Language (Englewood Cliffs, N.J.: Prentice-Hall, 1964); see also the more treatise-like book of J. J. Katz, The Philosophy of Language (New York: Harper Row, 1966). There are several anthologies: Leonard Linsky’s Semantics and the Philosophy of Language (Urbana: University of Illinois Press, 1952-a new, expanded edition is announced), my Philosophy and Ordinary Language (idem, 1963), Vere Chappell’s Ordinary Language (Englewood Cliffs, N.J.: Prentice-Hall, 1964), Katz and J. A. Fodor’s The Structure of Language (idem, 1964), G. H. R. Parkinson’s The Theory of Meaning (London: Oxford University Press, 1968).

² So does most of the literature mentioned in the preceding footnote.

³ See the last section of Kretzmann’s article mentioned in the first footnote for a general survey of recent philosophy of language.

⁴ Among the accounts of the history of analytic philosophy are J. O. Urmson’s Philosophical Analysis, its development between the two world wars (London: Oxford University Press, i956), G. J. Warnock’s English Philosophy since 1900 (London: Oxford University Press, 1958), J. A. Passmore’s A Hundred Years of Philosophy (London: Duckworth, 1957), and A. M. Quinton’s ‘Contemporary British Philosophy’ in D. J. O’Connor’s A Critical History of Western Philosophy (London: Collier-Macmillan, 1964).

⁵ J. L. Austin is the prime-mover in this connection; see his How to Do Things with Words (Cambridge, Mass.: Harvard University Press, 1962), for example. John R. Searle has a book, Speech Acts (Cambridge: Cambridge University Press, 1968), as well as several important papers in this area, at least one of which, ‘ Austin on Locutionary and Illocutionary Acts’ (The Philosophical Review, lxxvII (1968), pp. 405-24), is later than his book. See also W. P. Alston’s ‘ Meaning and Use’, Philosophical Quarterly, xIII (1963), pp. 107-24, revised in Parkinson’s anthology The Theory of Meaning. See also Alston’s Philosophy of Language, especially ch. 2.

⁶ Russell’s semantics of around 1905-12, in e.g. ‘On Denoting’ (Mind, 1905, often reprinted) and The Problems of Philosophy (London: Oxford University Press, originally 1912), is a view of this type, with the avowed aim of doing without meanings in addition to references.

⁷ See Alston, Philosophy of Language, Introduction, for other examples of the philosophical relevance of questions about language.

⁸ Of course there are differences on this matter, like others, even within a given movement; and perhaps sometimes these will be as great as differences between adherents of different schools. I am speaking broadly and of what is typical or characteristic.

⁹ It is the later Wittgenstein I have in mind here, after 1930 or so. The relevant writings are his posthumously published Philosophical Investigations (Oxford: Blackwell, 1953) and the earlier versions of parts of that work known as The Blue and Brown Books (Oxford: Blackwell, 1958), originally privately circulated in the late 1930s.

¹⁰ In British Philosophy in the Mid-Century, ed. C. A. Mace (London: George Allen & Unwin,. 1957) originally, and several times reprinted.

¹¹ The Philosophical Review, lxII (1953), pp. 167-86. This and the preceding article are reprinted in my anthology and elsewhere.

¹² Austin’s ‘The Meaning of a Word’, in his Philosophical Papers (Oxford: Oxford University Press, 1961), especially Part 1, is rather like the genuine Wittgenstein article - apparently through historical coincidence.

¹³ Cf. Ryle’s ‘The Theory of Meaning’ and Austin’s ‘The Meaning of a Word’, Part 11. See also Alston’s ‘The Quest for Meanings’, Mind, lxII (1963), pp. 79-87.

¹⁴ Ryle epitomizes this feature of these theories in his name for them, ‘ “ Fido ’’-Fido theories’. The word ‘Fido’ means Fido, the dog, which is its meaning.

¹⁵ London: Routledge & Kegan Paul, 1967. See also his article ‘Referring’ in Edwards’ Encyclopedia.

¹⁶ Mind, LIX (1950), pp. 320-44.

¹⁷ It is clear here, though not often insisted upon, that a relation of synonymy between sentence (-forms) is also involved in Russell’s semantics of this period, i.e. besides his naming relation. If the latter relation is really the only one, as Russell apparently thought, then he would have to hold that this synonymy relation was definable in terms of naming; but he nowhere even indicates an awareness of the problem, much less offers a solution to it.

¹⁸ In the case of Wittgenstein, parts of his literary remains and notes by others of his lectures and conversations are still appearing. In the case of both men, the bulk of their published works have appeared posthumously.

¹⁹ I am still speaking broadly. At an earlier date, behaviorist psychology was an important influence on philosophical semantics, e.g. in Charles Morris’ Signs, Language, and Behavior (Englewood Cliff's, N.J.: Prentice-Hall, 1946). This kind of affiliation is explicit also in Quine’s work, e.g. in Word and Object (New York: Wiley, 1960). Formal semantics, i.e. the mathematical work deriving from Tarski, has also been important. And there has also been recent work along Fregean lines.

²⁰ I have in mind, for example, the paper of Katz in the linguistics section.

²¹ The first publication on the linguistic basis for this viewpoint was a contribution to linguisitics, Katz and Fodor’s ‘ The Structure of a Semantic Theory’, Language, xxxIx (1963), pp. 170-210, reprinted in their anthology The Structure of Language. The chief proponent of this conception of philosophical conceptual analysis is Katz; see his The Philosophy of Language, especially chapters 1, 2 and 5.

²² For his methodological thinking, see the first chapter of his collection of articles, Linguistics in Philosophy (Ithaca, N.Y.: Cornell University Press, 1967), especially § 1.5. Some earlier thoughts on meaning occur in his comment on John R. Searle’s ‘ Meaning and Speech Acts ’ in Knowledge and Experience, ed. C. D. Rollins (Pittsburgh: University of Pittsburgh Press, n.d.). Other philosophers seem to be using this same kind of approach, e.g. Dennis W. Stampe,in his paper ‘ Toward a Grammar of Meaning’, The Philosophical Review, lxxvii (1968), pp. 137-74.

²³ Linsky’s book Referring, mentioned above, must be remembered. There are now some indications that linguists will be treating reference in a serious way in the near future (if they aren’t already), the bridge between the two discussions being the syntactical and semantical problems surrounding pronominal reference, especially in connection with what philosophers know as ‘ intensional’ contexts (i.e. ones like ‘ X believes that. . ‘ It is necessary that. . .’, etc.). See, e.g. George Lakoff’s forthcoming ‘ Counterparts, or the Problem of Reference in Transformational Grammar’, presented at the summer meeting of the Linguistic Society of America, July 1968, in Urbana, Illinois.

²⁴ The relevant writings are Strawson’s ‘On Referring’, already discussed, his Introduction to Logical Theory (London: Methuen, 1952), and his reply to Wilfred Sellars’ ‘ Presupposing’, The Philosophical Review, lxiii (1954), pp. 197-215, ibid. pp. 216-31. The Geach article is his ‘Russell’s Theory of Descriptions’, Analysis, x (1950), pp. 84-8, in which a line much like Strawson’s is advanced. Geach’s later book Reference and Generality (Ithaca, N.Y.: Cornell University Press, 1962, rev. ed. 1968) should also be mentioned. In the statement above I am omitting literature on presupposition and contextual implication, both of which have been involved in discussions of referring in the literature.

²⁵ The Journal of Philosophy, vol. lxv, no. 7 (4 April 1968), pp. 185-212.

²⁶ Many of the themes of his later work are foreshadowed or explicitly stated in earlier articles of his, a number of which are collected in his From a Logical Point of View (Cam¬ bridge, Mass.: Harvard University Press, 1953; 2nd ed., 1961) and in his The Ways of Paradox (New York: Random House, 1966).

²⁷ An important book in recent philosophical semantics is Ziff’s Semantic Analysis (Ithaca, N.Y.: Cornell University Press, 1960). A number of other essays of his also deal with various questions about language. See his collection of essays, Philosophic Turnings (Ithaca, N.Y.: Cornell University Press, 1966).