The linguistic turn and other misconceptions about analytic philosophy

Analytic philosophy has a complex history of more than one hundred years and this movement is so variegated that it can hardly be characterized by a single feature. Most of those who have tried to do so either were not aware of its diversity or considered only some part of its history. For example, it is sometimes believed that analytic philosophy is committed to a thoroughly anti-metaphysical stance. Such a belief may be rooted in some of the famous pronouncements of the logical empiricists,1 in the philosophical method put forward by Wittgenstein in his Tractatus Logico-philosophicus,2 or in the fact that some of the works of early analytic philosophy due to Russell and Moore – two of the founding fathers of the movement – have usually been interpreted as reactions against Bradley’s metaphysics and other versions of the British idealism of the time. Other facts, however, which support a completely different view, should not be overlooked. For one thing, Russell’s theory of the proposition and his logical atomism, as well as his philosophy of logic, clearly had metaphysical implications.3 For another, the logical empiricists’ anti-metaphysical crusade, which had been forceful in the twenties and the thirties, began to run out of steam in the sixties. At that time, other prominent figures of analytic philosophy were much less prone to reject any form of metaphysics as fundamentally unclear or unscientific: Quine’s famous criterion of ontological commitment had already been formulated in a paper which appeared in the Review of Metaphysics,4 Strawson had published his Individuals. An Essay in Descriptive Metaphysics,5 and Kripke’s semantics for modal logic would soon arouse a wave of metaphysical thinking about the existence of possible worlds. Today, metaphysics is a well established and respected important part of analytic philosophy – indeed, one of its main divisions – although the style of the authors who take part in it is, to be sure, not really akin to the one Hegel or Bergson used in their writings.6

Another example of a frequent misconception is the once commonly shared belief according to which analytic philosophers systematically disregard questions related to the history of philosophy because they insist on making a sharp distinction between the conceptual issues of philosophy itself and the historical problem of grasping what philosophers of the past have really said or thought, whereas philosophers with a leaning towards history, by contrast, often maintain that a study of the history of philosophy is indispensible for a good understanding of the deepest problems they try to cope with. Again, it is not difficult to cite famous analytic philosophers who fall squarely into the first category, who do not have a deep knowledge of the history of philosophy since Antiquity, or who have never been convinced that history of philosophy should occupy a large part of the syllabi in philosophy departments. But quoting famous examples does not make a rule, and against a hasty generalization, it can be remarked that Russell, who wrote a voluminous History of Western Philosophy,7 can certainly not be suspected of despising history of philosophy. More importantly, while it is true that, for a long time, most analytic philosophers did not attach the greatest importance to history (because they thought philosophy consisted mainly of debates and discussions in which clearly stated arguments were logically analysed), an important historical turn has been taken in the eighties and nineties, when more and more analytic philosophers became interested in the origins of the movement, or in the precise context which made it possible, and in the parting of the ways with other main traditions in the Twentieth Century. Coffa’s book on the semantic tradition8 and Hylton’s work on Russell and the emergence of analytic philosophy9 are good examples of such interest.10 It is also well known that Burton Dreben, although he did not publish very much, had a deep influence on many philosophers in Harvard, who later integrated an essential historical dimension in their philosophical work. Today, history of philosophy is the subject of many publications within the analytic tradition. In particular, early analytic philosophy has been the focus of a vast literature,11 but the interest of analytic philosophers for history actually extends far beyond this particular subject.

Another wrong – though widely shared – idea about analytic philosophy is its alleged allegiance to science and the scientific method: philosophia ancilla scientiarum. It is sometimes even believed that the general project of analytic philosophers is to reduce philosophical problems either to logical or to scientific ones, or to deal with them in a scientific way, and, as a consequence, to reject all genuine philosophical questions. Again, it is not difficult to find textual evidence for trying to justify such a (wrong) view. Did not Russell famously claim that logic was the essence of philosophy?12 Did not Carnap write that

we [the members of the Vienna Circle] give no answer to philosophical questions, and instead reject all philosophical questions, whether of Metaphysics, Ethics or Epistemology. For our concern is with Logical Analysis. If that pursuit is to be called Philosophy, let it be so; but it involves excluding from consideration all the traditional problems of Philosophy?13

More generally, the logical empiricists denied that philosophy could grasp any knowledge of a special kind or unveil truths about any special domain of objects just because it is supposed to use a properly philosophical, non scientific, method. For example, they denied any specifically philosophical knowledge of space, time, and causality, as opposed to the knowledge physicists can reach about these concepts. On this point, Philipp Frank once wrote: “Nowhere is there a point where the physicist must say: ‘Here ends my task, and from here on it is the task of the philosopher'”.14 As for Quine, his naturalization of epistemology implies that the philosophy of knowledge is nothing but a scientific activity under the control of other sciences, and no first philosophy: it is “a chapter of psychology and hence of natural science”.15 For him, philosophy is continuous with science. Such a series of quotations and references, however, gives but a partial view of the matter, and it is not difficult to find other citations which would bring the light on analytic philosophers with completely different orientations on this issue. To give just a single famous example, in the Tractatus, Wittgenstein insisted that philosophy is not at all on a par with science: “Philosophy is not one of the natural sciences. (The word ‘philosophy’ must mean something whose place is above or below the natural sciences, not beside them.)”16 And it would not be difficult at all to find many contemporary analytic philosophers who would strongly deny that the results of science are of any deep consequence for philosophy, and still others who would completely reject the idea that logical methods should be the basis of philosophical work.

As we can see, the three prejudices I have just mentioned about analytic philosophy in general – its rejection of metaphysics, its lack of interest in historical issues, and its allegiance to science – are not difficult to set aside as pure misconceptions, and, as a matter of fact, although there was a time when these prejudices were quite common, only people who are really not acquainted with the tradition of analytic philosophy would still hold them today. By contrast, it is not so easy to refute another widely held opinion which I shall now consider, i.e. the view according to which analytic philosophers are essentially linguistic philosophers and according to which analytic philosophy was born when philosophers introduced a fundamental change of method by taking a linguistic turn. Compared with the ideas I have very quickly examined above, this one is not so easily rejected, if only because it has been held by Michael Dummett, one of the most prominent figures of the analytic tradition. One of the reasons why it is not so easy to simply dismiss this view as erroneous is that “linguistic philosopher” is not a precisely defined phrase, and the term “linguistic turn” is far from having a single meaning. Since it was first coined – apparently in the middle of the Twentieth Century –, it has been used in many different contexts, even outside philosophy.17

On the other hand, when Dummett contended that “analytic philosophy was born when the ‘linguistic turn’ was taken”,18 he had a precise, if disputed, story to tell about this founding event and about what the linguistic turn actually is. The first part of this story is a characterization of analytic philosophy. On this point, according to Dummett,

what distinguishes analytical philosophy, in its diverse manifestations, from other schools is the belief, first, that a philosophical account of thought can be attained through a philosophical account of language, and, secondly, that a comprehensive account can only be so attained.19

In the second part of the story, we learn that the first clear manifestation of this belief is to be found in § 62 of Frege’s Die Grundlagen der Arithmetik of 1884, when Frege answers an epistemological Kantian question (“How are numbers given to us, granted that we have no idea or intuition of them?”) by a linguistic investigation. Indeed, just after raising the question, instead of trying to devise a solution of this puzzle by constructing a theory of knowledge in the traditional way, Frege gives an unexpected linguistic form to his answer: “Since it is only in the context of a proposition that words have a meaning, our problem becomes this: To define the sense of a proposition in which a number word occurs. […] In our present case, we have to define the sense of the proposition ‘the number which belongs to the concept F is the same as that which belongs to the concept G’.”20 This is not to say that Frege intended to introduce a dramatic change of method in philosophy, even though his famous context principle was clearly formulated (in the introduction of the book) as applying to an inquiry about the meaning of words.21 As Dummett rightly remarks, Frege did not find it necessary to insist on the particular and original form of the answer he gave there to the traditional question he had raised. “No justification for the linguistic turn is offered in Grundlagen: it is simply taken, as being the most natural way of going about the philosophical inquiry”.22 According to Dummett, however, the linguistic turn which was taken there for the first time, was to become “the methodological strategy of an entire philosophical school”.23

No doubt such a condensed version of the story Dummett tells us needs to be unpacked. The first part of it is what Dummett calls “the fundamental axiom of analytical philosophy – that the only route to the analysis of thought goes through the analysis of language.”24 Commentators have not failed to notice that this so called “axiom” was predicated on an assumption about the general task of philosophy as seen by Frege. This is made explicit in chapter 13 of Dummett’s book where he writes:

The philosophy of thought is that part of philosophy to which, apart from the philosophy of mathematics, Frege devoted his principal attention: he himself referred to it as ‘logic’ […] The philosophy of thought concerns itself with the question what it is to have a thought, and with the structure of thoughts and their components.25

Thus, according to Dummett, “philosophy of thought” is at the core of Frege’s philosophical project. It is well known that what Frege calls “thought” (Gedanke) is nothing such as a psychological entity; it is something objective which may be grasped by the thinker and by no means depends on his thinking activity. In Frege’s philosophy, the sense (Sinn) of a sentence (Satz) is a thought and, for example, the sense of the sentence which spells out Pythagoras’ theorem is one and the same for anyone who can grasp it: it is something objective which does not depend on the psychological representation (Vorstellung) each individual thinker may attach to it but belongs to what Frege famously called a “third realm” (drittes Reich), the two other ones consisting, first, of the things of the external world and, second, of the psychological realm of representations.26 Now this de-psychologization of thought or, to put it in Dummett’s wording, this “extrusion of thoughts from the mind”, is a key element of his account of the origins of analytic philosophy. According to Dummett, it is precisely because Frege’s doctrine of the objective thought came along with a thoroughly unsatisfactory ontological mythology that Frege had to take one further step and to give an account of our – we human beings’ – capacity to be able to grasp these entities of the third realm. And the key link with a philosophy of language is to be found precisely here: in the idea that the embodiment of thoughts is the institution of a common language. Here is how Dummett states this idea of his:

The accessibility of thoughts will then reside in their capacity for linguistic expression, and their objectivity and independence from inner mental processes in the common practice of speaking the language, governed by agreement among the linguistic community on standards of correct use and on criteria for the truth of statements.27

On Dummett’s account, if we want to understand the origins of analytic philosophy, it is not enough to point at a change of method happening in § 62 of Die Grundlagen der Arithmetik; we have to dig further and realize that the linguistic turn is itself rooted in a deeper layer, namely in Frege’s antipsychologism and his programme of a philosophy of thought. In his book, Dummett also undertakes to explain why Husserl’s anti-psychologism did not lead to the linguistic turn whereas Frege’s version of it did. But this is an issue which I shall leave aside here.

Nor is this the appropriate place to give a systematic examination of all the objections which have been raised against Dummett’s understanding of Frege’s philosophy or against his interpretation of the origins of analytic philosophy. Here, I shall consider only some of these objections, which have direct relationships to my topic. An obvious objection – which Dummett himself considers – is based on what Frege writes about philosophy and the deficiencies of natural languages, namely that “a great part of the work of the philosopher consists in […] a struggle with language.28 There are many texts in which Frege points out that the grammatical structure of the languages we use is not a good mirror of the structure of thought. If this is the case, how can our accessibility to thoughts be based on our use of language? An obvious answer to this objection is that the language which functions as a medium for our grasping of thoughts is not natural language but a conceptual notation which results from a critical logical analysis of natural languages. But is such an answer still valid after Russell has shown the fundamental flaws of Frege’s conceptual notation? Dummett gives textual evidence that even after the discovery of Russell’s paradox, Frege kept on thinking that language was the main access we have to the realm of thoughts. Oddly enough, however, Dummett does not insist on the fact that the basis of our access to thoughts through language is the result of a logical analysis, and that logic is a key factor for this access.29

In his review of Origins of Analytical Philosophy, Peter Hylton rightly points out the remarkable fact that in Dummett’s account of the origin of analytic philosophy, Frege’s interest in logic and mathematics is underemphasized. It is even “almost completely ignored”, so that “the implicit claim here is […] that [logic] is unimportant to his role as precursor of analytic philosophy”.30 This is one of the main objections Hylton raises against Dummett: what actually makes Frege’s extrusion of the thoughts from the mind crucial for a correct understanding of the origins of analytic philosophy is not so much that language is the medium through which we have access to the realm of thoughts but, rather, that these thoughts are submitted to a logical analysis, in which logic plays a crucial role, and this is a point that Dummett seems to miss. If we compare – as Hylton does – Frege with Russell on this point, we are quickly led to the following point which can be made against Dummett’s thesis. Russell also defended a form of “extrusion of the thoughts from the mind” in some of his early writings (Principles of Mathematics, 1903) at the time he was a realist about propositions. But he later gave up the idea in Principia Mathematica (1910-1913) and in The Philosophy of Logical Atomism (1918), although logical analysis evidently still is of the utmost importance in these writings. Therefore, on Dummett’s account, these later works should not be considered as having played any crucial role in the early development of analytic philosophy. Such a view is surely, if not absurd, at least unacceptable today in the light of what we know about this period in the history of philosophy, as Hylton rightly remarks.

Implicit in Dummett’s account is that both logical analysis in general and Russell’s works in the pre-war period in particular may be neglected in an inquiry on the origins of analytic philosophy, and this is what makes it really implausible today. It might be said in Dummett’s defence that it was only in the 1990s that scholars really began to investigate the historical background in logic, the foundations of mathematics, and physics in the development of early analytic philosophy.31 At the time Dummett gave the lectures which were the basis of his book (spring 1987), the crucial role of this kind of background had not yet been made as clear as it has become today through a huge amount of studies on the subject. It is doubtful, however, that Dummett himself would endorse such a line of defence, even today. He might prefer recalling that in the first chapter of his book, he explicitly denied that he intended to proceed to “a genuine historical inquiry”, and that he had made perfectly clear he would “ignore the contributions to the birth of analytical philosophy of the British philosophers Russell and Moore, and concentrate on those of philosophers writing in the German language”.32 This defence being granted, the trouble is that the precautions taken in the first chapter of his book have usually been overlooked, since most discussions about the characterization of analytic philosophy by the linguistic turn refer precisely to Dummett’s essay. This leaves the question open: can we nonetheless characterize analytic philosophy by some kind of linguistic turn, and in this case, what should we mean by that?

The phrase “the linguistic turn” began to be widely used after Richard Rorty chose it as the title of his famous anthology.33 This was not a coinage of Rorty’s: he explicitly took it up from Gustav Bergmann who had used it at least as early as 1953 in “Logical Positivism, Language, and the Reconstruction of Metaphysics”.34 Bergmann’s point, in this paper, is to defend the idea of a reconstruction of metaphysical questions in an ideal language, an idea which is at the core of what he considers to be one branch of the multifarious movement of logical positivism. From our present viewpoint, however, what matters is not so much to understand the exact meaning of Bergmann’s reconstructionism but, rather, to pay attention to the way he characterizes the whole movement of logical positivism, because this is precisely where the idea of a linguistic turn enters the scene. Here is what he writes about it:

[The logical positivists] unmistakably share a philosophical style. They all accept the linguistic turn Wittgenstein initiated in the Tractatus. To be sure, they interpret and develop it in their several ways, hence the disagreements; yet they are all under its spell, hence the common style. Thus, if names in themselves were important, it might be better to choose linguistic philosophy or philosophy of language.35

A few lines below, Bergmann adds that this by no means implies that language is the main, let alone the only, topic of their investigations; what Bergmann means is that they are “philosophers through language; they philosophize by means of it. […] Their characteristic contribution is a method.”36 I shall refrain from making comments on Bergmann’s contention that the logical positivists were all “under the spell” of the Tractatus and on other questionable judgments he passes about the logical positivists. What I want to underline, here, is that Bergmann is clearly at odds with Dummett on the interpretation of the linguistic turn. First, because according to Bergmann, the linguistic turn was not initiated by Frege but by Wittgenstein’s Tractatus37 and, secondly, because on Bergmann’s view, the linguistic turn does not characterize the whole tradition of analytic philosophy but only logical positivism, a movement to which Moore, Russell, and Frege do not belong. As a matter of fact, it makes perfect sense to see the Tractatus as explicitly and deliberately initiating a linguistic turn in philosophy, especially if we do not underemphasize the importance of logic in the early development of analytic philosophy, and if we compare Frege’s notion of logic with the one Wittgenstein expounds in his book. There is no space here to elaborate on Frege’s notion of logic and to make a thorough comparison with Wittgenstein’s so that in what follows, I shall have to restrict myself to a few hints.

The first point I would like to underline is that both Frege and Wittgenstein have a universalist understanding of logic although this has completely different implications for them as regard the relation between logic and language. That logic is universal, in this context, means 1) that there is just one logic and that it is absurd to put the word “logic” in the plural form or to consider a plurality of logical languages ; 2) that it is no less absurd to restrict logic to some particular domain of objects or universe of discourse; 3) finally, that there is no meaningful way to speak about logic without assuming logic itself, so that if the laws of logic are to be formulated in a language, there is no meta-linguistic meaningful way of talking about them. But whereas for Frege, logic is about the laws of truth and applies to thoughts (in the objective, non psychological, meaning that Frege gives to this term), for Wittgenstein, the laws of logic do not say anything, they are not about anything, and they are not related to anything outside language: “It is the peculiar mark of logical propositions that one can recognize that they are true from the symbol alone” (6.113).38 Wittgenstein makes it perfectly clear that for him, the propositions of logic essentially show the structure of language and that they describe neither the thoughts of Frege’s “third realm” nor anything else: “The propositions of logic demonstrate the logical properties of propositions by combining them so as to form propositions that say nothing.” (6.121) And also: “It follows from this that we can actually do without logical propositions ; for in a suitable notation we can in fact recognize the formal properties of propositions by mere inspection of the propositions themselves.” (6.122).

Actually, a more thorough examination of Frege’s and Wittgenstein’s conceptions of logic would confirm what has just been hinted at through these few quotations and remarks: whereas for Frege, the analysis of language may well be (as Dummett puts it) what gives us accessibility to an analysis of thoughts – these being the real target of logical analysis –, the linguistic turn that Wittgenstein takes as regard logic is much more radical since for him, the laws of logic simply do not point at anything outside language: “They have no ‘subject-matter'” (6.124). This being granted, the second and more important point is that this linguistic turn in logic comes along with a linguistic turn in Wittgenstein’s understanding of philosophy. Dummett has shown that the analysis of language is of the utmost importance for Frege’s programme in the philosophy of thought, and it may be argued that it is also most important for Frege’s philosophy of logic and of mathematics. But again, we may ask whether this is a good and sufficient reason to locate the linguistic turn in Frege, since we can see that Wittgenstein’s approach to philosophy is a linguistic one in a much more radical sense. In the Tractatus, Wittgenstein explicitly articulates his view about philosophy, and he characterizes it as an activity of clarification which does not go beyond language: “Philosophy does not result in ‘philosophical propositions’, but rather in the clarification of propositions” (4.112), and this is because “Most of the propositions and questions of philosophers arise from our failure to understand the logic of our language.” (4.003). The consequence is that “All philosophy is a ‘critique of language’ ” (4.0031). In view of what Wittgenstein says here, it is no wonder that when Bergmann spoke of a “linguistic turn” in philosophy, he referred to the author of the Tractatus, not to the author of Die Grundlagen der Arithmetik, as the initiator of this revolution in philosophy.39

If we now look at Rorty’s anthology, the picture we get of the linguistic turn is much more akin to the one Bergmann gives than to Dummett’s interpretation. In Part 1 of his book, Rorty compiles a series of papers under the title: “Classic statements of the thesis that philosophical questions are questions of language”, and the authors of the texts we can find there are Schlick, Carnap, Bergmann, and a few representatives of the so called “ordinary-language” branch of philosophy such as Ryle, Wisdom and Malcolm: not a single paper written before the thirties is to be found in the book, although Rorty would probably agree with Bergmann that the author of the Tractatus falls squarely in the category of the “linguistic philosophers”. In the introduction, Rorty makes it clear that the purpose of the anthology is to think about what he calls “the most recent philosophical revolution, that of linguistic philosophy”.40 He then gives the following definition:

I shall mean by ‘linguistic philosophy’ the view that philosophical problems are problems which may be solved (or dissolved) either by reforming language, or by understanding more about the language we presently use.41

If this is what is meant by “the linguistic turn”, then it is pretty clear that Frege should not be counted among the linguistic philosophers (and, as a consequence, that analytic philosophy can hardly be characterized by the linguistic revolution in philosophy). What is perhaps more interesting to note, in Rorty’s quotation, as well as in what Bergmann says about the logical positivists, is that there are kinds of linguistic philosophers, so that the generic term “linguistic turn” actually subsumes a number of philosophical methods, which, from Wittgenstein’s to Carnap’s, or from Schlick’s to Ryle’s and to the method of other ordinary-language philosophy from Oxford, are not only different but also often strongly opposed to one another. Carnap is a good example of a philosopher who has unmistakably taken a linguistic turn at some point of his career and who has nevertheless clearly distanced himself from Wittgenstein, although he has also acknowledged the strong influence of the Tractatus on his own thought. The turn was taken in the early thirties when Carnap introduced the syntactical method in philosophy, based on a distinction between an object language and a metalanguage. In the foreword of The Logical Syntax of Language, both his linguistic turn and his distance from Wittgenstein were made plain when he wrote: “Philosophy is to be replaced by the logic of science – that is to say by the logical analysis of the concepts and sentences of the sciences, for the logic of science is nothing other than the logical syntax of the language of science“.42 According to Carnap, the problems of traditional philosophy were formulated in a misleading way, as if their solution required the knowledge of some kind of object, whereas a correct reformulation of them makes it clear that what is really required for their solution is a logical analysis of the language of science, not a part of science itself, and even less a philosophical, non scientific, method allowing special access to objects, nor any access to special kinds of objects which would be foreign to a scientific investigation.

It is well-known that Wittgenstein himself took his distance from the Tractatus in his later philosophy. But he would not depart from the idea that the problems of philosophy are basically problems we have with the way we (mis)use language. He kept on thinking that philosophers usually do not realize how their questions and reflections deviate from what he calls the “grammar” of some of the words which occur in their sentences. In Wittgenstein’s Philosophical Investigations, we thus find still another understanding of what it means to take the linguistic turn in philosophy:

[The philosophical problems] are, of course, not empirical problems; they are solved, rather, by looking into the workings of our language, and that in such a way as to make us recognize those workings: in despite of an urge to misunderstand them. The problems are solved, not by giving new information, but by arranging what we have always known. Philosophy is a battle against the bewitchment of our intelligence by means of language.43

What should we conclude from this very sketchy survey of different ways in which some analytic philosophers have understood the analysis of language as the best means to cope with philosophical problems? When Rorty published his anthology, his concern was not with analytic philosophy as such but with a set of philosophers most of whom had been influenced by the Tractatus – though in different ways – although most of them, including Wittgenstein himself, wanted to distance themselves from this fascinating book. Among them, some were labelled “Ideal-Language Philosophers”, others “Ordinary-Language Philosophers”, depending on whether their philosophical method relied on the construction of artificial logical languages or on a descriptive analysis of natural languages. In the sixties, however, these two branches of linguistic philosophy had to face ever stronger criticism and Rorty was not the only one who thought it was time to take stocks of more than thirty years of exploring the consequences of a linguistic turn in philosophy. Neither Bergmann nor Rorty meant to capture the essence or any fundamental axiom of analytic philosophy. More recently, several authors have deemed, against Dummett’s view, that the linguistic turn is best suited to characterize some particular period in the development of analytic philosophy and not its whole history. Charles Parsons, for example, considers that the linguistic philosophy covers no more than thirty years of this history: “the idea that a certain kind of reflection on language is fundamental to much of philosophy does in my view characterize quite well one important period in the history of analytical philosophy, that of its rise to dominance in the English-speaking world, roughly from the early 1930s to the early 1960s.”44 As for Peter Hacker, he agrees with Bergmann that the linguistic turn was initiated by Wittgenstein’s Tractatus, and considers that linguistic philosophy developed through both “the logical empiricism of the Vienna Circle and its affi liates” and through “Oxford postwar philosophy”, before declining in the 1970s.45

When Dummett’s book on the origins of analytic philosophy was published in 1988, most readers paid less attention to the author’s warning to the effect that his essay would “not be a genuine historical investigation” than to his less cautious statement of what he called “the fundamental axiom of analytic philosophy”, namely “that the only route to the analysis of thought goes through the analysis of language”46 and to his questionable assertion that the fi rst occurrence of the idea is to be found in Frege’s Die Grundlagen der Arithmetik. Probably because Origins of Analytic Philosophy is a quite misleading book in many ways, few commentators have paid enough attention to the fact that even according to Dummett, the crucial step in the development of analytic philosophy was actually taken not by Frege but by Wittgenstein. Here is what Dummett writes in the last chapter:

If we identify the linguistic turn as the starting-point of analytical philosophy proper, there can be no doubt that, to however great an extent Frege, Moore and Russell prepared the ground, the crucial step was taken by Wittgenstein in the Tractatus Logico-philosophicus of 1922.47

The trouble is that it is hard to see how this is compatible with “the fundamental axiom of analytic philosophy” and with what Dummett says elsewhere in his book, if only because Wittgenstein was not at all committed to what Dummett calls a “philosophy of thought” when he comments on Frege’s philosophy. Consequently, it is no wonder that Dummett’s essay was to arouse so many criticisms. Ray Monk, for example, once remarked that if we take Dummett’s characterization of analytic philosophy seriously, we shall have to conclude that Lacan was an analytic philosopher, and that Russell was not.48

While an effort of clarification would have really been needed, “the linguistic turn” seems to have been a very attractive phrase to many philosophers who have thought it wise to endow it with all kinds of different meanings. Some have argued that we should talk of “linguistic turns” – in the plural form – because there has been many, insisting that the first one had been taken by Locke in his Essay Concerning Human Understanding, more precisely in Book III of the Essay, entitled, as we know, “Of Words”. On this basis, we are told that “the one that can be credited with making the first linguistic turn not only in the modern period but in the history of philosophy is John Locke (1632-1704)”.49 Others have found reasons to talk of a “linguistic turn” taken by Thomas Kuhn, because the author of The Structure of Scientific Revolutions once regarded the key feature of the incommensurability thesis as being its semantic-linguistic aspect.50 Still others have deemed that the linguistic turn should not be restricted to the English-speaking world and have recognized one such turn in the works of Hamann, Herder, and Humboldt, which paved the way for the philosophy of Heidegger and Gadamer. If Heidegger liked to quote the poem in which Stefan George writes “Where word breaks off no thing may be”, isn’t it because he too was defending some version of the linguistic turn?51 Even Jürgen Habermas has cherished the idea of a hermeneutic German counterpart of the linguistic turn in analytic philosophy.52 Finally, I have already remarked that historians too have a debate about a methodological revolution which they like to tag “the linguistic turn”.53 As we can see, at the point where the idea of a linguistic turn was in urgent need of clarification, we end up with a phrase which is used in so many different ways that it does not have any precise meaning anymore.

We may want to try to set aside all the secondary uses of the term and stick to a core meaning which would help clarifying the matter. This, however, would probably not allow us to reach a better characterization of analytic philosophy anyway. There is little doubt that the idea of dispelling confusions which results from our use of language played a fundamental role in the history of analytic philosophy, but it is very doubtful that such a vague idea can usefully characterize the whole movement, first because the term “linguistic turn” subsumes too great a variety of methods, and secondly because many philosophers who are clearly recognized as belonging to the analytic tradition would not consider this idea as an essential feature of their philosophical activity. In contemporary analytic philosophy, the idea of a linguistic turn is more rightly seen as an important phase of the past history which has long been digested, integrated, and aufgehoben, so that the analysis of language is but one of the many tools which are available to analytic philosophers. The many branches and ramifications of analytic philosophy in its historical development from its early phase to the present day can be described and the problem of its unity investigated, but analytic philosophy as such is definitively too variegated a movement to be characterized by a single feature.

I thank Todor Polimenov for helpful comments on a first draft of this paper.