History of Analytic Philosophy of Language
- Béatrice Godart-WendlingBéatrice Godart-WendlingUniversity of Paris
The term “philosophy of language” is intrinsically paradoxical: it denominates the main philosophical current of the 20th century but is devoid of any univocal definition. While the emergence of this current was based on the idea that philosophical questions were only language problems that could be elucidated through a logico-linguistic analysis, the interest in this approach gave rise to philosophical theories that, although having points of convergence for some of them, developed very different philosophical conceptions. The only constant in all these theories is the recognition that this current of thought originated in the work of Gottlob Frege (b. 1848–d. 1925), thus marking what was to be called “the linguistic turn.” Despite the theoretical diversity within the philosophy of language, the history of this current can however be traced in four stages:
The first one began in 1892 with Frege’s paper “Über Sinn und Bedeutung” and aimed to clarify language by using the rules of logic. The Fregean principle underpinning this program was that we must banish psychological considerations from linguistic analysis in order to avoid associating the meaning of words with mental pictures or states. The work of Frege, Bertrand Russell (1872–1970), George Moore (1873–1958), Ludwig Wittgenstein (1921), Rudolf Carnap (1891–1970), and Willard Van Orman Quine (1908–2000) is representative of this period. In this logicist point of view, the questions raised mainly concerned syntax and semantics, since the goal was to define a formalism able to represent the structure of propositions and to explain how language can describe the world by mirroring it. The problem specific to this period was therefore the function of representing the world by language, thus placing at the heart of the philosophical debate the notions of reference, meaning, and truth.
The second phase of the philosophy of language was adumbrated in the 1930s with the courses given by Wittgenstein (1889–1951) in Cambridge (The Blue and Brown Books), but it did not really take off until 1950–1960 with the work of Peter Strawson (1919–2006), Wittgenstein (1953), John Austin (1911–1960), and John Searle (1932–). In spite of the very different approaches developed by these theorists, the two main ideas that characterized this period were: one, that only the examination of natural (also called “ordinary”) language can give access to an understanding of how language functions, and two, that the specificity of this language resides in its ability to perform actions. It was therefore no longer a question of analyzing language in logical terms, but rather of considering it in itself, by examining the meaning of statements as they are used in given contexts. In this perspective, the pivotal concepts explored by philosophers became those of (situated) meaning, felicity conditions, use, and context.
The beginning of the 1970s initiated the third phase of this movement by orienting research toward two quite distinct directions. The first, resulting from the work on proper names, natural-kind words, and indexicals undertaken by the logician philosophers Saul Kripke (1940–), David Lewis (1941–2001), Hilary Putnam (1926–2016), and David Kaplan (1933–), brought credibility to the semantics of possible worlds. The second, conducted by Paul Grice (1913–1988) on human communicational rationality, harked back to the psychologism dismissed by Frege and conceived of the functioning of language as highly dependent on a theory of mind. The focus was then put on the inferences that the different protagonists in a linguistic exchange construct from the recognition of hidden intentions in the discourse of others. In this perspective, the concepts of implicitness, relevance, and cognitive efficiency became central and required involving a greater number of contextual parameters to account for them. In the wake of this research, many theorists turned to the philosophy of mind as evidenced in the late 1980s by the work on relevance by Dan Sperber (1942–) and Deirdre Wilson (1941–).
The contemporary period, marked by the thinking of Robert Brandom (1950–) and Charles Travis (1943–), is illustrated by its orientation toward a radical contextualism and the return of inferentialism that draws strongly on Frege. Within these theoretical frameworks, the notions of truth and reference no longer fall within the field of semantics but rather of pragmatics. The emphasis is placed on the commitment that the speakers make when they speak, as well as on their responsibility with respect to their utterances.
1. The Early Philosophy of Language
A mathematician, but trained in philosophy during his university studies in Jena and Göttingen, Gottlob Frege became interested in philosophical analysis when he realized that mathematics was unable to define its most fundamental concepts. The Foundations of Arithmetic (Frege, 1884) attempted to characterize the concept of number objectively (i.e., according to a law), thus breaking with the predominant psychological approach in 19th-century philosophy. Frege’s relationship with language, however, was complex. In the Begriffsschrift (Frege, 1879), he stressed that one of the tasks of philosophy was “to break the domination of the word over the human spirit by laying bare the misconceptions that through the use of language often almost unavoidably arise” (in Heijenoort, 1967, p. 7). This distrust of language, which may take the form of a desire to reform it, defined the first role of philosophy: to carry out a critique of language. However Frege also considered that language has the advantage of being able to express what is outside of it and thus to be the vector—albeit an imperfect one—of knowledge. This is why, when inquiring into the nature of numbers, he brings into play the notion of “statement,” which alone makes it possible to speak of numbers: “What is something asserted by means of a number statement?” (1884, p. 55). The second function of philosophy will therefore be to seek the truth about the objects of the world in the way we speak of them. By defining thus the aims of philosophy, Frege brought language to the forefront “in a completely new sense” that the French philosopher Jocelyn Benoist explained in these terms: “By saying that what characterizes Frege’s attitude is the promotion of the philosophy of language to the rank of ‘first philosophy’, Dummett (1981) never meant that Frege’s philosophy could be reduced to a philosophy of language or is even fully realised within it, or that Frege’s philosophy constitutes the last word. Frege’s discovery is rather that the access to all sorts of questions traditionally dealt with by philosophy, including the supposedly fundamental questions dealt with by ontology, depends on some new kind of attention to language. The priority of the ‘philosophy of language’ is thus essentially epistemological” (Benoist, 2011, p. 57). It is in this very particular sense that Frege can be considered as the founding father of the philosophy of language, even if his contribution to this approach goes far beyond this theoretical position.
1.1 The Logical Clarification of Language
Although Frege’s first book, Begriffsschrift, is not considered to mark the beginnings of the philosophy of language, its verdict on natural languages and the interest of its formalism were at the foundation of the idea that the grammatical form of sentences was often misleading, so that they had to be analyzed through their logical form. More precisely, Frege’s argument began by denouncing that “logic has hitherto always been too closely related to language and grammar” and that we must henceforth think, no longer in terms of subject and predicate, but using the concepts of argument and function. Adding to this mathematical division of the sentence the idea that sentences are always composed of a judgement and a content, Frege defined a two-dimensional formalism that can express complex propositions containing connectors, quantifiers, as well as negations with different scopes (see Kneale & Kneale, 1962, pp. 478–493).1 This way of formalizing language was difficult to use, and remained unnoticed until 1903, the date at which Russell echoed it in his Principles of Mathematics. Working with the mathematician, logician, and philosopher Alfred North Whitehead (1861–1947), Russell consolidated and developed the foundations of Frege’s first-order logic in Principia Mathematica (Whitehead & Russell, 1910–1913) by using the logical notation of Giuseppe Peano (1858–1932). Conceived with the aim of reducing mathematics to logic and of solving the paradox of the set of all sets identified by Russell in Frege’s Grundgesetze der Arithmetik, Principia Mathematica’s formalism ushered in a hierarchical conception of language that Rudolf Carnap (1891–1970) was to put into play in Der logische Aufbau der Welt (Carnap, 1928). Russell’s contribution to the philosophy of language also consisted in exploiting Frege’s idea that the grammatical form of propositions may be misleading. Thus, questioning the truth-value to be attributed to a sentence such as “the king of France is bald,” which does not refer to any individual existing in the world considered, Russell highlights in “On Denoting” (1905) that the logical formulation of this proposition makes it possible to attribute a truth-value to it while preserving bivalent logic. Indeed, from a logical point of view, this sentence is rewritten: “there exists an x such that (a) x is king of France and (b) there is not a y such that y is different from x and y is king and (c) x is wise”. This goes to show that the sentence is obviously false, since there is no x satisfying the existential quantifier. In “On What There Is” (Quine, 1948), Quine synthesized the merits of this logical reformulation in these terms: “The virtue of this analysis is that the seeming name, a descriptive phrase, is paraphrased in context as a so-called incomplete symbol. No unified expression is offered as an analysis of the descriptive phrase, but the statement as a whole which was the context of that phrase still gets its full quota of meaning—whether true or false” (p. 6). Russell’s student, Wittgenstein, in the only book published during his lifetime, Tractatus Logico-Philosophicus (1921), pushed the theses of logical atomism developed by his master to the extreme, arguing that language is endowed with a structure reflecting the ontological hierarchy of the world (objects, states of affairs, facts), so that the proposition is the image (Bild) of a real or possible fact that has the same structure as itself. The proposition-picture therefore has the same logical form as the fragment of the world it describes, even if the logical form in itself is unsayable (aphorism 4.121). This interpretation that establishes an isomorphism between language and reality was defended by Peter Hacker (1975), but was rejected by James Conant and Cora Diamond (cf. Crary & Read, 2000). It is also in this series of aphorisms that constitutes the very particular form of this book that Wittgenstein put forward the idea—taken up by the Vienna Circle (Carnap, Hahn, & Neurath, 1929)—that the recourse to logical syntax (aphorism 3.325) would make it possible to avoid semantic ambiguities that are at the heart of many pseudoproblems in philosophy (aphorisms 3.323 and 3.324). In opposition to the approach of Russell, who in his theory of types mobilized the meaning of signs, Wittgenstein dissociated syntax and semantics by advocating that meaning should not be taken into account in the development of logical syntax (aphorisms 3.331 and 3.33). The ideas contained in the Tractatus were received enthusiastically by the members of the Vienna Circle and were implemented by Carnap in Die Logische Syntax der Sprache (1934). Though attracted by Principia Mathematica’s formalism, Carnap nevertheless devoted his first book, Der logische Aufbau der Welt (1928), to developing a construction system retracing the step-by-step derivation of all the concepts of science. Conceived in a phenomenological perspective, this reconstruction of the “genealogy of concepts” was accomplished by distributing the names of “objects” (i.e., of concepts) in different “spheres” (i.e., types) and by listing the different relationships they are likely to have with one another. Aufbau presents a hierarchical conception of language comprising, in ascending order, four main spheres—the autospychological, the physical, the heteropsychological, and the cultural—which make it possible to situate not only the different concepts but also the properties that characterize them by grouping them into “quality classes” or “circles of similarity.” In this system, the construction of a concept is achieved by means of a translation rule, called “constructional definition,” which consists in transforming all the statements in which a appears into statements that contain only the lower-level concepts b and c (it is said that a is reducible to b and c). On the other hand, the aim pursued in Die Logische Syntax der Sprache was to treat the syntax of formal and natural languages as “a calculus in the strict sense.” Thus, syntactic analysis operates only on words reduced to “characters” devoid of meaning and reference. Following Wittgenstein’s aphorism 3.33, this methodological choice is also presented by Carnap as an extension of the program of David Hilbert (1862–1943) to linguistics. Thus, just as Hilbert had formulated “a system of rules whose objects are mathematical formulas in their formal structure” (Carnap, 1934, p. 9), Carnap defined rules of formation and transformation whose function was, respectively, to account for the logical properties of the propositions (order and category of symbols) and to calculate the links of logical dependencies (consequence, incompatibility, etc.) that these propositions have with one another. This syntactic method was thus used by Carnap to construct two formal languages expressing arithmetic (Language I) and the propositions of classical physics and mathematics (Language II). The development of these syntaxes enabled him to argue that they constitute “a system of reference” with respect to which the syntactic properties of natural languages can be identified (1934, p. 8). Carnap’s idea—contrary to Frege, Russell, or Wittgenstein—is not that we must reform natural languages, because while he considered that languages resist formalization because they are governed by an internal logic that does not depend exclusively on their syntactic structure, he also emphasized that the syntactic adequacy of a natural language is based on historical and empirical criteria. In doing so, he emphasized the importance of the question: does this syntax agree with the way in which speakers of this language speak? And—although Carnap’s position has remained largely unknown—he directed reflection toward an examination of the use of language as was later advocated by the philosophy of ordinary language in the 1960s. Quine, who was a fervent admirer of Carnap, opposed his project of a logical construction of the world by putting forward a naturalistic point of view which claimed that logic, our logic, is only the result of a cultural accident (Two Dogmas of Empiricism, Quine, 1951). Considering, against his predecessors, that logic cannot be reduced to a simple calculus operating on propositions (“Ontological Remarks on the Propositional Calculus,” Quine, 1934, p. 474), Quine defended the universality of logic (as it enables scientific knowledge to be rewritten) while maintaining that there is no need to grant it a special place since any truth is revisable. The only legitimacy of logic is that it structures our “conceptual schema,” that is to say, our way of conceiving and speaking about the world (From a Logical Point of View, Quine, 1953). In this perspective, it is fallacious to think that we can change our point of view, although we can revise progressively, as in Neurath’s boat, our conceptual scheme (Word and Object, Quine, 1960b). This position, which was at the origin of the thesis of the indeterminacy of translation, initiated the transition from logical empiricism to naturalism, whose arguments are still at the heart of debates in the philosophy of language.
1.2 Meaning and Reference
Constituting the claimed or criticized basis for the philosophy of language, Frege’s “Über Sinn und Bedeutung” (1892) and “Der Gedanke” (1918–1819) introduced a real revolution in the field of philosophy by proposing an anti-mentalist theory of meaning allowing us to reformulate the question of the representation of the world within language. In his inquiry into the supply of knowledge that an equality of type a = b can offer with regard to a = a, Frege introduced a clear distinction between the meaning of signs (i.e., “any designation representing a proper name,” Frege, 1892 [Engl. Transl. 1970, p. 57]) and their reference by defining, on the one hand, the meaning as corresponding to the mode of donation of the reference and, on the other hand, the reference as denoting the object of the world designated by the meaning of the sign. In the case of a proposition, its meaning corresponds to the “thought” (i.e., to the “objective content”) that it expresses and its reference denotes its truth value, “true or false.” Thus, for Frege, the domain of meaning is extralinguistic, since it enables us to link linguistic signs to the (concrete, abstract, or fictional) referents available in the universe. The interest of Frege’s paper lies also in the fact that it anchors its reflection in the observation of precise linguistic facts. This feature, which also explains its impact in the philosophy of language, leads Frege to consider the grammatical form of sentences. This is why Frege examines the case of sentences containing reported speech (both direct speech with the use of quotation marks, and indirect speech) or even the case of subordinate clauses, which, because they only express part of a thought, cannot have a truth value. This fine-grained analysis leads him to modify his starting position by arguing that when “one talks about the sense of another person’s remarks [. . .], words do not have their customary reference but designate what is usually their sense” (Frege, 1892 [Engl. Transl. 1970, p. 59]). The first of the three Logical Investigations, “der Gedanke,” presents a Platonic conception of thought by defining it as the “owner” of truth or falsehood (Frege, 1977, p. 14). By being in no way related to human subjectivity—“a thought belongs neither to my inner world as an idea, nor yet to the external world, the world of things perceptible by the senses” (Frege, 1977, p. 26)—thought belongs to the realm of logic and epistemics, and its study depends not on the philosophy of language but on a depsychologized theory of mind. Frege also argued in this text that any affirmative proposition is distinguished from imperative, optative, and interrogative propositions by the fact that it contains “two things closely joined”: “the grasp of a thought (thinking) and the acknowledgement of the truth of a thought (the act of judgment)” (Frege, 1977, p. 7). Austin, Frege’s translator, was not indifferent to this kind of analysis, even though he replaced the problem in terms of truth values by that relating to the felicity conditions of utterances. By seeking to define what language can say to be true or false from a logical point of view, Frege clearly took an interest in the pragmatic aspects of what he calls “speaking” (das Sprechen). This led him to underline the emotive power of language, the modalities of speech (cognitive versus fictional) and to foreshadow the importance of the notion of illocutionary force (see “Thoughts” in Frege, 1977). But it was also according to Frege’s writings that the question of how to determine the meaning of propositions was posed. Thus, although in “Compound Thoughts” (1923–1926, in Frege, 1977, p. 55), Frege outlined a compositional approach to meaning such as “we look upon thoughts as composed of simple parts,” in Die Grundlagen der Arithmetik, he advocated on the other hand a contextualist principle in which “the meaning of a word must be inquired after in propositional context, not in isolation” (Frege, 1884, Engl. transl. p. 17). Wittgenstein took advantage of this Fregean principle, as shown by aphorism 3.3 in the Tractatus Logico-Philosophicus: “only the proposition has sense; only in the context of a proposition has a name meaning.”
Russell broke from the Fregean approach to meaning and reference by arguing that the meaning of a word corresponds to its reference (direct reference theory). In adopting this position as early as The Principles of Mathematics (1903), Russell borrowed from Moore the idea that the real constituents of a proposition are neither linguistic meanings nor intentional objects, but entities represented by words. Avoiding both idealism and psychologism, Russell initially maintained that there are two kinds of terms: things, indicated by proper names, and concepts, designated by all other words (adjectives correspond to predicates, verbs to relationships). This approach, which gives an existential status to a very large number of (concrete and abstract) entities, was however revised by Russell in “On Denoting” (1905). Using the “knowing by acquaintance / knowing about” distinction, Russell showed that many expressions (including proper nouns) actually correspond to incomplete symbols that can be eliminated by rewriting them in formal terms (cf. the above logical transcription of “The King of France is bald”). However this revision of Russell’s ontological system did not change his approach to meaning as corresponding to reference, since it was reaffirmed in An Inquiry Into Meaning and Truth (1940). Although Moore, Russell’s colleague at Cambridge, did not share his friend’s enthusiasm for a clarification of language in terms of logical analysis, the paradox he set out, during one of his lectures, caught the attention of Wittgenstein (On Certainty, Wittgenstein, 1969). Moore was interested in the propositions of ordinary language, and puzzled about the strangeness of the sentence “It is raining, but I do not believe that it is raining.” In doing so, he emphasized the singular logic of the verb “believe” and the dissymmetry between the first- and third-person pronouns (since the proposition “It is raining, but he does not believe that it is raining” is entirely mundane). Apart from this small incursion into the meanders of natural language, the philosophy of language from the period 1892–1930 remained focused on the idea that language is imperfect and that it must be reformed. But in order to do this, we must be able to distinguish between propositions that are meaningful and those that are nonsense. Having assimilated the meaning of a sentence to a thought, Frege considered that there could be no logically faulty thought, since it would have no meaning. This theoretical position was taken up by Wittgenstein in aphorism 3.03 of the Tractatus, in accordance with the aim of this book, which was to draw the limits of meaning and to exclude meaningless phrases and sentences from the language. The members of the Vienna Circle (Schlick, Carnap, Hahn, Neurath, and others) attached great importance to this question of delimiting meaning and developed a method to determine which words and propositions populate this area of nonsense. The aim of this circle was to promote logical empiricism by defending three theses: (a) the language of physics can unify the sciences, (b) philosophy is an elucidation of scientific, ethical, and aesthetic propositions, (c) metaphysical statements are meaningless (Unsinnig). In a famous article, “Überwindung der Metaphysik durch Logische Analyse der Sprache” (1932), Carnap stated the two conditions that must be met by a word in order to be able to say that it has a meaning. One must first know the syntax of the word, “that is to say, the mode of its occurrence in the simplest sentence form in which it is capable of occurring” (1932, p. 13; Engl transl. in Ayer, 1959, p. 62). This form, called “elementary sentence,” allowed Carnap to assign a “syntactic category” to the word, which, in turn—thanks to a chain of definitions—leads to the complete definition of the word, corresponding to the “immediate datum” to which this word refers. Thus each word must be connected to a verifiable observation sentence (i.e., a protocol sentence) and this verification requirement is the second condition that a word must fulfill in order to be considered meaningful. Unsurprisingly, the application of this method to metaphysical sentences shows that they are meaningless, so that it is necessary to prohibit their use in language. Quine’s From a Logical Point of View presented a systematic critique of the notion of meaning as expounded by Frege, Russell, and Carnap. Thus, in “Two Dogmas of Empiricism” (Quine, 1951), Quine undertook to refute Carnapian reductionism consisting in believing “that each meaningful statement is equivalent to some logical construct upon terms which refer to immediate experience” (p. 20). Countering “the supposition that each statement, taken in isolation from its fellows, can admit of confirmation or infirmation at all” (p. 41), Quine defended—while maintaining an empiricist position—a holistic point of view in which “our statements about the external world face the tribunal of sense experience not individually but only as a corporate body” (p. 41). The attack against Russell’s theory of direct reference is ironic in “The Problem of Meaning in Linguistics” (in Quine, 1953), which argued that “confusion of meaning with reference has encouraged a tendency to take the notion of meaning for granted. It is felt that the meaning of the word ‘man’ is as tangible as our neighbour” (p. 47). Quine relied on Russell’s theory of definite descriptions, however, to develop, from an ontological point of view, his Ockham’s razor. He thus argued in “On What There is” that “the only way we can involve ourselves in ontological commitments [is] by our use of bound variables” (in Quine, 1953, p. 12). This theoretical position, well known under the motto “To be is to be the value of a variable” was then used by Quine to put forward his thesis of the relativity of ontology. He held that: “we look to bound variables in connection with ontology not in order to know what there is, but in order to know what a given remark or doctrine, ours or someone else’s, says there is; and this much is quite properly a problem involving language” (pp. 15–16). The end of “The Problem of Meaning in Linguistics” also launched a criticism of the Fregean “Sinn” as presupposing the existence of a kernel of meaning common to various languages, by posing the first arguments that later led to supporting the thesis of the indeterminacy of translation. This thesis, which maintains that we always project our logic, our categories of thought, on the indigenous language was more fully developed in the lecture “Le mythe de la signification” that Quine delivered in French at the 1958 Royaumont conference (this lecture is unpublished in English, cf. Colloque philosophique de Royaumont, 1962). The theme of Word and Object (1960) was mainly devoted to this thesis, which claimed that we always translate from within our own language, so that the categories of the indigenous language are not discovered by the linguist, but superimposed onto it according to the conceptual scheme in which we have been brought up. This thesis, which is inseparable from the myth of meaning and the myth of objects, and argues that even the logical connectors “and,” “or,” and so forth are dependent on our own culture, introduced, in the words of the philosopher Sandra Laugier, “anthropology in logic, and indeterminacy in semantics” (Ambroise & Laugier, 2009, p. 152).
By giving logical analysis the role of determining whether a proposition has a meaning and therefore a truth value, the philosophy of language perceives a proposition as playing the role of mirroring reality. This approach, which defines truth and falsity in terms of correspondence or non-correspondence with the various states of the world, encountered from the start two related difficulties: how to evaluate propositions containing words without any reference (Pegasus, Odysseus, etc.) and how to preserve the principle of bivalence (any proposition is true or false). In this respect Frege’s position in “Über Sinn und Bedeutung” differed from that of his successors, since he considered that: “the thought remains the same whether ‘Odysseus’ has reference or not” (Frege, 1892, p. 63). But by claiming this, Frege opened the door to trivalent logic, such that sentences containing an empty reference will be evaluated as “neither true nor false” (on this subject, see Linsky 1967). Russell, with his theory of definite descriptions (1905), combated this kind of approach by showing that the logical transcription of a sentence containing an empty reference (“the king of France is bald”) shows that it is false. In doing so, he succeeded in maintaining the framework of bivalent logic. Against Hegel’s monistic logic, Russell developed during 1913 what he later called the Philosophy of Logical Atomism in order to determine the ontological elements that contribute to calculating the truth value of sentences. In The Philosophy of Logical Atomism (Russell, 1918–1919), he defined logical atoms (i.e., “the last residue in analysis”, p. 498) as corresponding to either “particulars” (“such things as little patches of color or sounds, momentary things,” p. 498) or universals (the predicates or relations that can be attributed to these particulars). In this perspective, any atomic proposition (i.e., containing only a particular and a predicate) can express a “fact” and is, depending on its adequacy to reality, true or false (bipolarity thesis). Thanks to logical connectors, several atomic propositions can be assembled to form a complex proposition whose truth value can be calculated. In doing so, Russell defined truth as testifying to an isomorphism between propositions and the facts that they describe, even though he admitted that this case can only be found in logically perfect languages. While retaining the framework of logical atomism and the idea that “propositions can be true or false only by being pictures of reality” (aphorism 4.06), Wittgenstein distanced himself from Russell’s position by arguing that it is not the “particulars,” but the facts that are the logical atoms of the real, because particulars cannot be apprehended independently of the facts in which they appear (“the world is the totality of facts, not of things,” aphorism 1.1). Moreover, Wittgenstein disagreed with Russell’s direct reference theory in which the meaning of a proposition is reduced to its reference, because it does not explain: why a proposition can be understood before knowing whether it is true or false; why a false proposition has a meaning just as much as a true proposition does; and why a proposition has the same meaning whether it is true or false. Wittgenstein proposed, therefore, to distinguish the meaning of a proposition from its reference by arguing that the reference corresponds to the fact described by the proposition, while its meaning lies in its truth conditions (“The proposition shows its sense. The proposition shows how things stand, if it is true. And it says, that they do so stand,” aphorism 4.022). Understanding a proposition therefore means being able to determine what must be the case for it to be true or false: “to understand a proposition means to know what is the case, if it is true. (One can therefore understand it without knowing whether it is true or not)” (aphorism 4.024). The members of the Vienna Circle were highly influenced by the Tractatus and focused their attention on some of its aphorisms, provoking Wittgenstein’s incomprehension and disinterest toward them. They placed aphorism 6.53 at the core of their approach—“The right method of philosophy would be this. To say nothing except what can be said, i.e. the propositions of natural science”—in order to develop a verificationist approach to meaning (see Waismann, 1945). This approach separates meaningful from meaningless statements and maintains that: “the meaning of a statement lies in the method of its verification. A statement asserts only so much as is verifiable with respect to it. Therefore a sentence can be used only to assert an empirical proposition, if indeed it is used to assert anything at all. If something were to lie, in principle, beyond possible experience, it could be neither said nor thought nor asked” (Carnap, 1932, p. 76). This method led the theorists of the Vienna Circle to argue that meaningful utterances can be divided into three classes: “First there are statements which are true solely by virtue of their form (“tautologies” according to Wittgenstein; they correspond approximately to Kant’s “analytic judgments”). They say nothing about reality. The formulae of logic and mathematics are of this kind. They are not factual statements themselves, but are used to transform such statements. Secondly there are the negations of such statements (“contradictions”). They are self-contradictory, hence false by virtue of their form. With respect to all other statements the decision about truth or falsehood lies in the protocol sentences. They are therefore (true or false) empirical statements and belong to the domain of empirical science” (Carnap, 1932, p. 76). This distinction between analytic and synthetic statements led Carnap to distinguish, in Die logische Syntax der Sprache, L-rules that are pure logical rules void of meaning (analytic truths), from physical rules, noted P-rules, which are meaningful because they describe the world by expressing either a law of nature or a particular proposition describing an empirical fact (synthetic truths). Carnap’s originality in the debate regarding analytical versus synthetic truth was to defend a Principle of Tolerance advocating that, depending on the scientific aim, sentences considered to be analytical can be modified by convention, thus making the adoption of a new language possible. For Carnap, “it is not our business to set up prohibitions, but to arrive at conventions,” for “in logic, there are no morals. Everyone is at liberty to build up his own logic, i.e. his own form of language, as he wishes. All that is required of him is that, if he wishes to discuss it, he must state his methods clearly, and give syntactical rules instead of philosophical arguments” (Carnap, 1934, §17). Quine’s critical contribution to his predecessors was, first, to call into question the notion that we can speak of “truth by convention” about logical truth (Quine, 1936) and second, to cast doubt on the epistemological relevance of the dichotomy opposing analytic truths (grounded in meanings independently of matters of fact) and synthetic truths (grounded in fact). Advocating a holistic conception of language, Quine argued—against the Principle of Tolerance that presupposes the analytic / synthetic distinction—that the evaluation of every synthetic utterance involves all the analytic and synthetic statements on which it is based. In other words, any statement whose truth value seems to depend only on experience makes use of a whole underlying theory that makes it unnecessary to distinguish analytic utterances from synthetic utterances. For Quine, the issue in this debate is to rid logic of its status of tautological or conventional truth and replace it with an approach that emphasizes that analyticity is related to human sociability by being a matter of learning: “a sentence is analytic for a given native speaker if he learned the truth of the sentence by learning to use one or more of its words” (“Two Dogmas in Retrospect,” Quine, 1991, p. 270). This naturalization of the notion of analyticity, which makes it a matter of degree, undermines the nature of scientific certainty defended by the Vienna Circle by making it clear that no statement can be excluded from doubt and from revision.
2. The Second Stage of the Philosophy of Language
Strawson’s article “On Referring” (Strawson, 1950) marked a real break with the logicist and truth-conditional approach to meaning developed by the analytical philosophy of the first half of the 20th century. By arguing against Russell’s theory of definite descriptions (“On Denoting” 1905), Strawson claimed that it is only by taking the use of utterances and of their contexts into account that we can determine their truth value and the reference to be assigned to their constituents. The defect in logical analysis that he points out is that it proposes solutions that are not consistent with the close dependence that statements in use have on their context to signify, to refer, and, if appropriate, to have a truth value. This emphasis on use was concomitantly defended by Wittgenstein who denounced “the preconception of crystalline purity” of logic (Philosophical Investigations, Wittgenstein, 1953, §108) and advocated an immanent examination of language by discovering its rules. Austin, in The William James Lectures (delivered at Harvard University, 1955; Austin, 1962), initiated the approach called “the philosophy of ordinary language,” which held that a scrupulous analysis of the uses of language in context showed that they have the function of carrying out different types of acts. By thus arguing that language serves to accomplish actions, Austin paved the way for what was to become pragmatics and laid the basis for linguistic analysis oriented toward the communicative dimension of language. The upheavals resulting from this new approach consisted, on the one hand, in reviving the debate on the relationship between statements, reality, and truth, and on the other hand, in introducing a completely different conception of meaning in terms of the felicity conditions of statements. This approach was underpinned by a reflection on the conventional and intentional dimensions of language. Austin argued that the success of a statement only depends on the respect of certain conventions, whereas Searle and Grice emphasized the intentional aspect of the meaning of statements. By thus establishing a dependency link between the psychology of the speakers and meaning, Grice backgrounded the notion of use and in doing so, created a break with the philosophy of ordinary language. The idea was no longer to respect the Austinian principle of linguistic immanence, but to base semantics on an intentional ontology. In this new perspective, the problem could then focus on the implicit aspects of communication, thus accounting for the fact that the interpretation of utterances is not limited to understanding their literal meaning.
2.1 Use and Context
Strawson’s main criticism of Russell’s theory of definite descriptions was that his analysis of the statement “Is the King of France bald?” is “mistaken” (Strawson, 1950, p. 324) since in the current context nobody would answer that this sentence is false: as the king of France does not exist, the question does not even arise. This common-sense objection led Strawson to distinguish between the sentence, the use of the sentence, and the utterance of the sentence. The sentence corresponds to the linguistic unit with an invariant meaning whatever its use. To say, “the king of France is bald,” in the 9th or the 17th centuries, always amounts to saying that the king of France is bald. On the other hand, they are different utterances of the same sentence that make it possible to put it in use. While the sentence in itself offers the possibility of designating virtually different individuals (Charles II, Henri IV, Louis XIV, etc.), only the use of the sentence can refer to an individual in particular, since it consists in uttering it in a particular context. Uttered in the 9th century, “the king of France is bald” will designate Charles II, while stated in the 17th century the sentence will refer to Louis XIV. It follows that a sentence can be true or false only when it is considered in use, since in the 9th century its use will be true and false in the 17th century. In other words, the truth of a sentence does not depend on its meaning alone, but on the use to which the sentence is put. Similarly, the meaning of an expression—such as “the King of France”—does not allow us to determine whether, when uttered in a particular circumstance, it asserts something true (or false) and refers to a specific entity. As Strawson explains, the sentence and the expression carry only “general directions” that are updated (or not) when they are put into use:
“Meaning (in at least one important sense) is a function of the sentence or expression; mentioning and referring and truth or falsity, are functions of the use of the sentence or expression. To give the meaning of an expression (in the sense in which I am using the word) is to give general directions for its use to refer to or mention particular objects or persons; to give the meaning of a sentence is to give general directions for its use in making true or false assertions. It is not to talk about any particular occasion of the use of the sentence or expression. The meaning of an expression cannot be identified with the object it is used, on a particular occasion, to refer to. The meaning of a sentence cannot be identified with the assertion it is used, on a particular occasion, to make.” (1950, p. 327)
Thus an expression is not, contrary to what Russell thought, referential in itself and a sentence is not true because it has a referent. To have a meaning, to be endowed with a truth value and to refer to something are three very different things, even if the attribution of a truth value to a sentence in use is dependent on its referential capacity. Strawson considered that the proper use of a sentence presupposes that Russell’s referential condition “there is a king of France” be satisfied. However, this condition does not belong to semantics, but to pragmatics, since it is thanks to the context of enunciation that the presence of the referent will be confirmed or invalidated. To say “the king of France is bald” in the 21st century amounts to misusing the underlying sentence. The theoretical upheaval introduced by Strawson was, first, to highlight that the question of reference is a problem of a pragmatic nature that is not part of logical analysis and second, to maintain that the semantic dimension does not take precedence over use and so plays a much lesser role than logicians thought.
The contribution of Keith Donnellan (1931–2015) to the debate on definite descriptions was to show that they can, depending on the intentions of the speaker, be used to achieve two types of use. In opposition to Russell’s and Strawson’s theories, which attributed only a referential use to definite descriptions, the famous article “Reference and Definite Descriptions” (Donnellan, 1966) argued that a definite description can also be used in an attributive manner. In doing so, Donnellan distinguished the case where the definite description is used to establish that an object has such-and-such properties (attributive usage) from one in which it has the function of producing an identifying reference (referential use): “A speaker who uses a definite description attributively in an assertion states something about whoever or whatever is the so-and-so. A speaker who uses a definite description referentially in an assertion, on the other hand, uses the description to enable his audience to pick out whom or what he is talking about and states something about that person or thing” (1966, p. 285). The need to distinguish between these two types of use can be seen in the different behavior they have toward presuppositional failure. Indeed, “when we hypothesize that the presupposition or implication [in the sentence ‘Smith’s murderer is insane’] is false, there are different results for the two uses. In both cases, we have used the predicate ‘is insane’, but in the first case, if there is no murderer, there is no person of whom it could be correctly said that we attributed insanity to him. Such a person could be identified (correctly), only in case someone fitted the description used. But in the second case, where the definite description is simply a means of identifying the person we want to talk about, it is quite possible for the correct identification to be made even though no one fits the description we used” (p. 286). As a result, in attributive usage, presuppositional failure leads to considering the sentence as neither true nor false, whereas in referential use the sentence can continue to be evaluated in terms of truth or falsity, since the use of a definite description, even if false or partially incorrect, may lead someone to select the person or object being spoken about. Thus, it is possible, at a party, to allow one’s interlocutor to correctly identify the man we want to talk about by using a false description “Look at the man who is drinking champagne” (though the man is in fact drinking water); the falsity of the presupposition “there is champagne in his glass” does not result in making this sentence neither true nor false. Thanks to this attributive / referential distinction, Donnellan succeeded in combining—contrary to Grice (1975 [in Grice, 1989])—the notions of intention and use.
It was also in reaction to Russell’s theory of definite descriptions that Wittgenstein drew attention to the importance of use. In opposition to the Russellian idea that the logico-semantic analysis of language would make it possible to identify the elementary components of reality, Wittgenstein argued that it is futile to try to determine and distinguish what is simple from what is not, because the verdict of simplicity is obtained only by considering the use made of the word or the sentence in its precise context of enunciation. Taking the armchair, the tree, and the chessboard as examples, paragraph 47 of the Philosophical Investigations developed the vague and relative character of the notions of “simple” and “compound” if they are not related to the use that we wish to make of them: “The question ‘Is what you see composite?’ makes good sense if it is already established what kind of compositeness—that is, which particular use of this word—is in question” (Wittgenstein, 1953, § 47). Thus speaking of the “simple constituent parts of a chair” does not mean anything, if one does not specify in what circumstances, purpose, or intention, this expression is used, because the referents that can then be associated will be as diverse as “the pieces of wood from which it is assembled? Or the molecules, or the atoms?” (Wittgenstein, 1953, § 47). Contrary to the position developed in the Tractatus where the use of a sign only displays the combinatorial possibilities of the object it stands for (§3.326 and §3.327), the Philosophical Investigations held that: “the meaning of a word is its use in the language” (§43). More precisely, the meaning of an expression corresponds to its use as it is governed by the grammatical rules of the language in question, that is to say by the type of explanation that could be provided to justify that this expression can be used meaningfully in this language. By thus overlapping the notions of meaning and rule-guided use, Wittgenstein places meaning in language games (Sprachspiele) that grammar can reveal in accordance with the life forms (Lebensformen) that specify how language is intertwined with our cultures and our worldview.
Although Austin disagreed with Wittgenstein’s main arguments, he contended—like Wittgenstein—that meaning is dependent on use. This priority given to use manifests itself in two forms: firstly, in paying particular attention to the precise uses of words by examining “what we should say when” (Austin, 1970, p. 182) and in using these attested uses to challenge the validity of certain philosophical issues. Resorting to usage thus constitutes a method of analysis that differs from logical analysis by considering that the reconstructions proposed by logicians do not agree with the ordinary language of individuals. The interest of this ordinary speech, which the observation of usage reveals, is that it is the fruit of a common agreement between men that has passed the test of history and that has thereby acquired the ability to be used properly. In opposition to the scholastic language of philosophy, Austin therefore argues that:
“our common stock of words embodies all the distinctions men have found worth drawing, and the connexions they have found worth marking, in the lifetimes of many generations: these surely are likely to be more numerous, more sound, since they have stood up to the long test of the survival of the fittest, and more subtle, at least in all ordinary and reasonably practical matters, than any that you or I are likely to think up in our arm-chairs of an afternoon—the most favoured alternative method.”(Austin, 1970, p. 182)
Stanley Cavell, in his essay “Must We Mean What We Say?” (Cavell, 1969) pursued this line of thought oriented to a scrupulous observation of use while maintaining, like Austin, that our statements perform actions.
2.2 Felicity Conditions Versus Truth Conditions
This attention to use had serious consequences for conceptions of meaning and truth. It called into question the appropriateness of using truth conditions to account for the meaning of utterances, and new parameters (the notions of act, of intention, of constitutive and regulative rule), considered as irrelevant during the first phase of the philosophy of language, emerged. The truth predicate also became an object of controversy and debates on its status (performative or not?) and its function (correspondence vs. coherence) arose.
2.2.1 Language as Action
Although Austin never rejected the term “philosophy of ordinary language” to describe his approach, he argued in “A Plea for Excuses” that attention to the use of words provides access to “realities we use the words to talk about.” In other words “we are using a sharpened awareness of words to sharpen our perception of, though not as the final arbiter of, the phenomena” (Austin, 1970, p. 182). The most accurate name for this current should therefore, Austin felt, be “linguistic phenomenology” because it reflects the link that the uses of words have with reality. The idea was not to postulate a representationalist thesis such that language is a copy of the world, but to argue that the use of words allows one to act on the world by modifying it. So to make a promise is to introduce something new in the world: a commitment, and therefore a responsibility, of the speaker that will have an effect on the world. But the action performed by the promise will be successful only if it respects a conventionally and socially established procedure that is not dependent on the notions of truth or falsehood. To make a promise without the intention of keeping it, or in knowing that its object is unachievable, are cases of failure due to the fact that certain felicity conditions of the statement are not respected. The functioning of our utterances is therefore not governed by truth-conditional considerations, but by pragmatic conditions. This approach, initially based on the analysis of performative statements that explicitly express the action they perform on the world (I give and bequeath my watch to my brother, I name this ship the Queen Elisabeth, etc.), was extended by Austin to constative statements on the basis that: “we see then that stating something is performing an act just as much as is giving an order or giving a warning; and we see, on the other hand, that, when we give an order or a warning or a piece of advice, there is a question about how this is related to fact which is not perhaps so very different from the kind of question that arises when we discuss how a statement is related to fact” (1970, p. 251). This extension to the other verbs of English (i.e., those that are not explicitly performative) led Austin to propose a typology of acts divided into five classes: verdictives, exercitives, commissives, behabitives, and expositives (Austin, 1962, pp. 150–163). More precisely, any enunciation in this approach may express three types of act: “the locutionary act (and within it the phonetic, the phatic, and the rhetic acts) which has a meaning; the illocutionary act which has a certain force in saying something; the perlocutionary act which is the achieving of certain effects by saying something” (1962, p. 120). While an illocutionary act, according to Austin, belongs to linguistic analysis, since its effect can be determined by taking into account the conventional meaning of the words (to inform, to order, to warn, to undertake, etc.), the success of a perlocutionary act is never guaranteed, because one can, for example, intend to flatter someone and succeed in practice only in annoying him. This is why Austin excluded the perlocutionary dimension from philosophical analysis by writing: “It is certain that the perlocutionary sense of ‘doing an action’ must somehow be ruled out as irrelevant to the sense in which an utterance, if the issuing of it is the ‘doing of an action’, is a performative, at least if that is to be distinct from a constative. For clearly any, or almost any, perlocutionary act is liable to be brought off, in sufficiently special circumstances, by the issuing, with or without calculation, of any utterance whatsoever, and in particular by a straightforward constative utterance” (1962, p. 109). One of the contributions of Stanley Cavell (1926–2018) was to rehabilitate the perlocutionary dimension in “Passionate and Performative Utterance” (2005) in order to account for the fact that all utterances necessarily have an “interpretation”.
The American philosopher John Searle, who was Strawson’s and Austin’s student at Oxford, profoundly modified Austin’s approach by incorporating the notion of intention. In this perspective, “the illocutionary act is the minimal unit of linguistic communication [. . .] produced by a being with certain intentions” (Searle, 1962, pp. 156–157). By thus introducing the intentional dimension, borrowed from Grice’s article “Meaning” (1957), Searle was able to account for the fact that every utterance involves the literal meaning of words (semantic level), as well as the speaker’s desire “to produce a certain effect by means of getting the hearer to recognize his intention to produce that effect” (pragmatic level). In order to account for this intertwining of the conventional and the intentional, Searle proposed a division of utterances as necessarily consisting of two ingredients: (a) a proposition that is assessable in terms of true or false and that corresponds to the common content of an utterance in the imperative, assertive, interrogative form, and so forth; (b) an illocutionary force marker (expressed by the verbal mode, the emphasis, the intonation, performative verbs, etc.) often revealed by the context. While Searle considered that the proposition, if envisaged in isolation from its illocutionary force, was unintelligible, he nevertheless broke with the Austinian conception by reintroducing truth-conditionality into his analysis. Moreover, Searle emphasized that every act is governed by constitutive rules which, unlike “regulative rules,” control “an activity the existence of which is logically dependent on the rules.” Thus, to express the illocutionary force, different semantic rules must be followed: (a) the preparatory rules that deal with the linguistic and contextual aspect of the utterance; (b) the rule of sincerity that reflects the intentions of the speaker; (c) the essential rule that testifies the speaker’s commitment to do what he said. The examination of the act of promising was the paradigmatic example from which Searle derived these rules, thus offering a basis of analysis for what was to become the theory of speech acts. As evidenced by his book Intentionality, from the 1980s on Searle turned to the philosophy of mind, although his approach to speech acts continued to feed his thinking in The Construction of Social Reality (Searle, 1995).
2.2.2 Competing Approaches to Truth
By considering a statement as realizing an act, the notion of truth conditions ceased to play the central role it had had in logico-analytic philosophy. But this evolution also resulted in a reflection on the status of the truth predicate. As early as 1949, in his article “Truth” that drew heavily on Austin’s paper “Other Minds” (1946), Strawson argued that the truth predicate corresponded to a performative allowing us to perform an act of approval on, agreement with, or confirmation of a statement already made (Strawson, 1949, p. 95). This theoretical position, which reconciled Ramsey’s redundancy theory of the truth predicate (“Caesar was murdered is true” is the same as “Caesar was murdered,” Ramsey, 1927, p. 4) and Austin’s approach, was contested by Austin. In a long friendly polemic that was to continue until Austin’s death (1960), Austin defended a contextualist conception of truth and refused to accept that truth can be eliminated. More precisely, he argued that no statement contains in itself the criteria of its truth, because these are provided by the context in which the statement is uttered. To situate the use of the statement in its historical context, Austin called for two types of conventions: descriptive (i.e., semantic) conventions, which give meaning to words as belonging to a given language, and demonstrative conventions, which refer to the historical reference that the statement acquires in a given use. Descriptive relationships thus refer to types, while demonstrative conventions refer to situations. However the application of these two types of convention is not enough to tell the truth: it is also necessary for the statement to be uttered in a context where it can say what it says and where what it says will be judged as corresponding, according to the relevant contextual properties, to the reality of which it speaks. Truth or falsity thus have, in Austinian theory, the status of relations of evaluation of the content of statements relative to their context, since it is a question of deciding whether what the statement describes, in its given context, is consistent with the situation to which it refers with the situation-type. Therefore, this Austinian approach to truth posits that truth is a relation of correspondence between words and facts. But this relation of correspondence must be understood in a conventional sense, since it corresponds to “a relation between a type conventionally attached to a sentence and a situation just as conventionally attached to a use of this sentence; a relationship that is evaluated by a contextual judgment as to the ‘sameness’ of the historical and contextually determined state of affairs, identified by the statement, and the situation-type, identified by the meaning of the sentence, as it is determined by its descriptive conventions” (Benoist, 2006, pp. 229–230).
A completely different conception of truth-correspondence was put forward by Donald Davidson (1917–2003) in the late sixties. Drawing on the founding work of Alfred Tarski (1901–1983), “The Concept of Truth in Formalized Languages” (first published in Polish in 1933 [Tarski, 1933] and in English translation in 1956 [Tarski, 1956]), Davidson argued that the formal definition of truth developed by this Polish logician could serve as the basis for a theory of meaning for natural languages. Tarski held that one must formulate a material condition of adequacy, which he called “Convention T,” such that the theory allows to derive as a consequence, for each statement of the language studied (i.e., the object-language), the T-equivalence of the form: X is true if and only if p (where X is replaced by a name of a statement of the object-language and the letter “p” by a statement proposing the translation of the statement X). Davidson’s thesis is that it is possible to construct truth theories for natural languages that derive T-sentences (“S” is a true sentence of L if and only if p) for each S-sentence of L that gives its truth conditions. Since such a theory displays the truth conditions of each sentence—taking into account the “meanings of all independently meaningful expressions on the basis of an analysis of their structure” (Davidson, 1970, p. 55)—this theory is considered a theory of meaning for L. However, as the philosopher François Rivenc pointed out: “It seems therefore that the T-Convention assumes, in order to function as a criterion of adequacy, that we understand the object-language prior to the elaboration of the theory. But Davidson proposes to construct a theory of truth subject to the same condition, for the interpretation of a language that we may not understand at all before the theory gives us the meanings” (Rivenc, 1998, p. 11). Saul Kripke strongly criticized the possibility that the T-Convention may apply if the object language is not already understood (in Evans & McDowell, 1976, p. 409). Many criticisms were levelled at Davidson’s theory (see Foster, 1976 and Davidson’s answer, Davidson, 1985, pp. 171–179; Lepore, 1986). After successive modifications of his theory, Davidson (1989) finally proposed a coherence theory of truth that implied correspondence: “if coherence is a test of truth, then coherence is a test for judging that objective truth conditions are satisfied, and we no longer need to explain meaning on the basis of possible confrontation. My slogan is: correspondence without confrontation” (1989, p. 307). Adopting Quine’s principle of charity, Davidson contended that a “presumption in favour of their truth” (Davidson, 1989, p. 319) must be given to beliefs because they are supported by a number of other beliefs and are interpreted “according to the events and objects in the outside world that cause the sentence to be held true” (Davidson, 1989, p. 317).
3. The Third Phase
From the 1970s on, theorists of the philosophy of language worked in two different directions. The first was the path of formalization and led to the development of modal semantics. In this new formal framework, determining the meaning of a sentence containing a modal expression was obtained by means of a function associating possible worlds to truth values: a proposition with “can” will be true in at least a possible world, whereas a proposition expressing a necessity will be true in all possible worlds. The second line of research explored the logic of communication in an approach that was initially nonformal (Grice), and then as part of the semantics of possible worlds through David Lewis’s work.
3.1 Possible Worlds and Theories of Direct Reference
Kripke’s early work on the semantics of modal systems (Kripke, 1963), combined with the research conducted by the Finnish logician Jaakko Hintikka (1929–2015), contributed to laying the foundation for the semantics of possible worlds. This type of semantics, on which Montague Grammar (1970; in Thomason, 1974) was to be grounded, also made it possible, within the framework of analytic philosophy, to rethink the fundamental notions of meaning and reference through the problem of the proper name, the terms of natural species and indexicals. Thus, contrary to Frege-Russell’s position that proper names are only abbreviated definite descriptions and declaring also his dissatisfaction with Searle’s conception (1958) in which proper names correspond to a bundle of descriptions, Kripke argued in Naming and Necessity (1972) that proper names (the name of a person, a city, or a country) are rigid designators in that they refer to the same object in all the possible worlds where this object exists (p. 36). Although the notion of “possible world” is used informally in this book, which is more philosophical than technical (Kripke, 1972, p. 15), it nonetheless proposes a clarification of this notion by stressing that “a possible world is given by the descriptive conditions we associate with it” (Kripke, 1972, p. 32), so that the “‘possible worlds’ are stipulated” (Kripke, 1972) by modifying only those “features of the world relevant to the problem at hand” (p. 14). In other words, the appropriate synonyms of “possible worlds” could be “possible state (or history) of the world” or “counterfactual situations” (p. 15), since the idea that supports the notion of “possible worlds” is that these differ from the real world only by certain properties and states of the world. The difficulty, noted by David Lewis (1968), of such a conception (called “transworld identity” by Kripke) is that it allows the creation of a chain of possible worlds in which, by successive modifications from world to world, an object could at the end of the chain be totally dissimilar to what it was at the beginning (Aristotle could thus become a cat). In order to overcome this difficulty, Lewis (1968 and 1986) developed an alternative theory, counterpart theory, which brings into play not the identity relation but the similarity relation that presupposes that the counterpart of a thing shares with it some essential properties. Lewis explained his position in the following terms:
“The counterpart relation is our substitute for identity between things in different worlds. Where some would say that you are in several worlds, in which you have somewhat different properties and somewhat different things happen to you, I prefer to say that you are in the actual world and no other, but you have counterparts in several other worlds. Your counterparts resemble you closely in content and context in important respects. They resemble you more closely than do the other things in their worlds. But they are not really you.”(Lewis, 1968, p. 114)
The further step that Lewis (1986) took was to defend a modal realism, proposing that our world is only one world among others that are also current from a certain point of view. The primacy of the world in which we live results from an indexical interpretation: the present world means “this world, our world” and its greater reality results only from a modal illusion rooted in our egocentrism.
The analysis of terms referring to natural species led Kripke to argue that the latter have “a greater kinship with proper names than is generally realized. This conclusion holds for certain for various species names, whether they are count nouns, such as ‘cat’, ‘tiger’, ‘chunk of gold’, or mass terms such as ‘gold’, ‘water’, ‘iron pyrites’. It also applies to certain terms for natural phenomena, such as ‘heat’, ‘light’, ‘sound’, ‘lightning’, and, presumably, suitably elaborated, to corresponding adjectives ‘hot’, ‘loud’, ‘red’” (Kripke, 1972, p. 134). In defending such a conception, Kripke differed from the position of John Stuart Mill (1806–1873), who argued that general names are endowed with a connotation, even though he agreed with Mill’s approach to proper names in considering that they only have denotation and no connotation (Mill, 1843). The scheme proposed by Kripke for proper names and the names of species (designating something) was then the following: “An initial ‘baptism’ takes place. Here the object may be named by ostension, or the reference of the name may be fixed by a description. When the name is ‘passed from link to link’, the receiver of the name must, I think, intend when he learns it to use it with the same reference as the man from whom he heard it” (Kripke, 1972, p. 96). In other words, Kripke adopted a different position from that of Strawson, emphasizing the “effective communication chain” and not resorting to the way the speaker thinks he has acquired the reference of the proper name. This reflection on the specificity of names was accompanied throughout the three lectures (delivered in Princeton in 1970) that make up Naming and Necessity by a critical analysis of the all-too-frequent amalgamation in philosophy of a number of dichotomies (for example, confusion between a priori and necessity).
Independently of Kripke’s thought, the American philosopher Hilary Putnam defended a rather similar position, because he attacked, on the one hand, the idea that reference is determined by meaning and supported, on the other hand, a causal theory of reference concerning proper names and natural-kind words. Considering in his famous article “The Meaning of ‘Meaning’” (Putnam, 1975), the common usage that posits that in one sense “meaning” means extension (the set of things for which the term is true) and in another sense intension (the psychological apprehension of the concept associated with the word), Putnam first highlighted the inadequacy of the meaning = extension equivalence, then demonstrated using the possible world of Twin Earth (where the word “water” means not H2O but XYZ) that meaning, taken in the sense of intension, does not make it possible to determine the reference of a name, because “it is possible for two speakers to be in exactly the same psychological state (in the narrow sense) even though the extension of the term A in the idiolect of one is different from the extension of the term A in the idiolect of the other. Extension is not determined by psychological state” (1975, p. 139). Instead of the conception, which he judged erroneous, in which the knowledge of meaning comes down to the psychological apprehension of a concept, Putnam advocated a sociological approach based on the concept of “division of linguistic labor” (p. 144). The idea is that there are “expert speakers” (p. 145) who possess a scientific method of identification for natural-kind words (gold, water, tiger, etc.). Other speakers rely on these experts to understand the meaning of these words without being forced to master the scientific techniques underlying the definition of words. It follows that “Whenever a term is subject to the division of linguistic labor, the ‘average’ speaker who acquires it does not acquire anything that fixes its extension. In particular, his individual psychological state certainly does not fix its extension; it is only the sociolinguistic state of the collective linguistic body to which the speaker belongs that fixes the extension” (p. 146). In practice, the meaning of a word can be learned either through an ostensive definition involving indexicals (“this is water”) or by using a description involving a stereotype (water is colorless, transparent, tasteless, thirst-quenching, etc.). Putnam’s contribution to indexicals was twofold: (1) he emphasized that they function as rigid designators that refer to the same object in all possible worlds; (2) he argued that natural-kind words also contain an indexical component (“‘water’ is stuff that bears a certain similarity relation to the water around here”, p. 152). Concerning the notion of stereotype, Putnam defines it as consisting of “central features” corresponding to criteria “which in normal situations constitute ways of recognizing if a thing belongs to the kind or, at least, necessary conditions (or probabilistic necessary conditions) for membership in the kind. Not all criteria used by the linguistic community as a collective body are included in the stereotype, and in some cases the stereotype may be quite weak” (p. 147). In this perspective, a speaker will be considered to master a word if he knows some stereotypes that characterize it. The general pattern that emerges from Putnam’s approach is that each name is correlated to a semantic representation containing four components: the syntactic marker (water: mass noun), the semantic markers (water: natural kind, liquid), a description of the additional features of the stereotype (colorless, transparent, tasteless, thirst-quenching, etc.), and extension (H2O), determined by the competence of expert speakers. This theory is known by the name of “semantic externalism,” since it determines the extension without resorting to psychological states while integrating the interactional and therefore sociological dimension of language. Putnam’s later work led him to adopt Dummett’s position that learning a language or understanding a sentence does not imply knowing its truth conditions, and he addressed issues specific to the philosophy of mind (the thesis of the identity between mental states and cerebral states).
The thesis according to which indexicals—like proper names—are rigid designators was at the heart of the work of David Kaplan. After having defined indexicals (a term borrowed from Charles Sanders Peirce) by the fact that their referent depends on the context of use and that their meaning consists of a rule that determines the referent in terms of certain aspects of the context, Kaplan (who was a student of Carnap) distinguished two types of indexicals: demonstratives (“that,” “he,” etc.), which require a visual presentation of a proximal object distinguished by a pointing gesture, and pure indexicals (“I,” “no,” “here” [in one sense], “tomorrow,” etc.) that do not require the intervention of an associated demonstration. Working like Kripke and Putnam in the context of the semantics of possible worlds, Kaplan developed a terminology capable of showing that “Indexicals, pure and demonstrative, are directly referential” (Kaplan, 1977, p. 492). Thus, he distinguished, on the one hand, the notion of context (the possible occasions of use) from the possible circumstances of evaluation (i.e., counterfactual situations) and defined the “content” of a sentence in context by a function that attributed an appropriate extension to the circumstances of evaluation (i.e., intension in the sense of Carnap). Thus, if the content is a proposition (i.e., the content of a sentence in a certain context), the result will be a truth value, whereas if it is a singular term in one circumstance, it will be an object. Finally, Kaplan used the term “character” to denominate the semantic rule that, by convention, determines the content of the indexical in each context. In this perspective, characters are represented as functions that correspond to the possible contexts of the contents. The specificity of the indexical “I” is thus: “In each of its utterances, ‘I’ refers to the person who utters it” (Kaplan, 1977, p. 520). According to Kaplan, these language rules formed the meaning of pure indexicals. Calling this approach the direct reference theory therefore falsely suggests, as Kaplan pointed out, that: “the reference is not mediated by a meaning, which it is. The meaning (character) is directly associated, by convention, with the word. The meaning determines the referent; and the referent determines the content” (1977, note 44, p. 520).
3.2 The Logic of Conversation
Trained in the practice of the philosophy of ordinary language, Paul Grice contributed to the pragmatic turn initiated by this approach by proposing in 1975 a theory of meaning accounting for the fact that “what words mean is a matter of what people mean by them” (Grice, 1989, p. 340). In doing so he posited—unlike Austin—that meaning is independent of use, since use, by corresponding to the conventional meaning of words, does not enable us to account for the intentions and hence for what speakers mean implicitly. In order to propose a theory that accounted for the difference that can exist between the explicit literal meaning (“what is said,” i.e., the natural meaning evaluable in terms of true or false) and the implicit meaning (“what is implicated,” i.e., the “nonnatural meaning” that does not contribute to truth conditions) of utterances, Grice defined the concept of “implicatures.” Specifically, he distinguished conventional implicatures that are triggered by the presence of a linguistic expression (the case of “therefore” in “He is an Englishman; he is, therefore, brave” that triggers the implicature “All Englishmen are brave,” 1989, p. 25) from conversational implicatures that stem from the application of conversational maxims (the case of dialogue A: “Smith doesn’t seem to have a girlfriend these days”; B: “He has been paying a lot of visits to New York lately” where B implicates that Smith has, or may have, a girlfriend in New York, 1989, p. 32). Conversational implicatures were in turn divided into two distinct cases: “generalized conversational implicatures” triggered by the presence of a linguistic form (the case of “a” in “X is meeting a woman this evening” that would normally implicate that the person to be met was someone other than X’s wife, mother, sister, or perhaps even close platonic friend, Grice, 1989, p. 37) and “particularized conversational implicatures” that do not appear in a systematic way because they are dependent on a particular context (A: “I am out of petrol”; B: “There is a garage round the corner,” 1989, p. 32). The discrimination between conventional and conversational implicatures is based on the following criteria: cancellable, calculable, detachable, and determined. Thus, contrary to conventional implicatures, conversational implicatures are characterized by the fact that: (a) they are cancellable without generating a contradiction, (b) computable on the basis of a principle of cooperation and conversational maxims, (c) not detachable (by synonymous substitution) and (d) indeterminate (they are fluctuating since they depend on the context). By conceiving of conversation as rational, Grice’s contribution was to highlight that conversational implicatures are not arbitrary, because their content can be predicted according to a single principle called the Cooperative Principle. This principle stated: “Make your conversational contribution such as is required, at the stage at which it occurs, by the accepted purpose or direction of the talk exchange in which you are engaged” (Grice, 1989, p. 26) and was accompanied by four maxims (Quantity, Quality, Relation, and Manner) that reflect the following aspects of this cooperation: the amount of information provided (“Make your contribution as informative as is required for the current purposes of the exchange”; “Do not make your contribution more informative than is required”); its truthfulness (“Do not say what you believe to be false”; “Do not say that for which you lack adequate evidence”); its relevance (“be relevant”); the way it is formulated (“Avoid obscurity of expression”; “Avoid ambiguity”; “Be brief”; “Be orderly”; 1989, pp. 26–27). Grice’s thought gave rise to many studies. Linguists and psycholinguists investigated whether conversational implicatures project themselves from simple sentences to complex sentences (Atlas & Levinson, 1981; Chemla & Spector, 2011; Chierchia, 2004; Chierchia, Fox, & Spector, 2012; Fox, 2007; Geurts & Pouscoulous, 2008; Sauerland, 2004), and van Rooij and de Jager (2012) proposed a modeling of conversational implicatures in terms of game theory.
Unlike Grice, who had proposed an informal theory of communication, David Lewis in his article “Scorekeeping in a Language-Game” (Lewis, 1979) advocated an approach to conversation rooted in the formal perspective of modal semantics. A student of Ryle, Strawson, Austin, and Grice, and who did his PhD under Quine’s supervision (at Harvard in 1967), Lewis proposed to consider the parameters of the context of enunciation relevant to the meaning of the statements under consideration as an index. Following the article by Richard Montague (1930–1971) “Pragmatics” (Montague, 1968), Lewis complicated the calculation of meaning performed in modal logic by defining it as a function associating a set of elements (a possible world, a time and place of enunciation, speakers, etc.) to a truth value. The determination of these contextual elements (which Lewis called indices) depends on their relevance to the state of the conversation at the moment of the utterance; they can therefore vary during the verbal exchange. Their evolution is, however, governed by rules of accommodation that specify, for example, “that any presuppositions that are required by what is said straightway come into existence, provided that nobody objects” (Lewis, 1979, p. 347). This approach allowed him to rethink the Russellian problem of defined descriptions by challenging from sentences—“The pig is grunting, but the pig with floppy ears is not grunting” and “The dog got in a fight with another dog”—that a definite description “the F” denotes x if and only if x is the one and only F in existence, because “the proper treatment of descriptions must be more like this: “the F” denotes x if and only if x is the most salient F in the domain of discourse, according to some contextually determined salience ranking. The first of our two sentences means that the most salient pig is grunting but the most salient pig with floppy ears is not. The second means that the most salient dog got in a fight with some less salient dog” (Lewis, 1979, p. 348). The language facts for which Lewis proposes a new analysis are not limited to presuppositions and definite descriptions, but also include movement verbs, vague statements, relative modalities, and performatives. By emphasizing the intricacy of meaning and context of enunciation, Lewis succeeded in showing that the application of formal semantics to the analysis of natural language is able to relativize the boundaries separating semantics from pragmatics.
A theme that was particularly worked upon during this second phase of the philosophy of language was the issue of the notion of presupposition. The status of presuppositional phenomena was first of all debated in order to determine whether they should be placed among the conditions of use (Austin, 1962; Fillmore, 1969; Frege, 1892; Russell, 1905; Strawson, 1950; Van Fraassen, 1968) or as a part of the content of statements (Ducrot, 1972; Karttunen, 1974; Kempson, 1975; see Godart-Wendling & Goubier, 2015). Since 1969 Charles Fillmore (1929–2014) had contended that presupposition was an implicit semantic phenomenon. It became the main focus of the work of Robert Stalnaker, who never ceased to argue that presuppositions belong to pragmatics since their function is to convey the speaker’s background beliefs, that is to say: “the propositions whose truth he takes for granted, or seems to take for granted, in making his statement” (Stalnaker, 1974, p. 48). In this perspective, presuppositions are therefore the “propositions implicitly supposed before the relevant linguistic business is transacted” (Stalnaker, 1970, p. 280) and therefore they may correspond to unspoken beliefs (Stalnaker, 1973, p. 447), which will, however, have to be taken for granted by the interlocutors for their communication to proceed rationally and for it to succeed. Seen from this angle, presupposition is therefore only a particular case of the exploitation of Gricean conversational rules since, by their use, “the speaker exploits the rules governing normal conversation in order to communicate something that is not exactly said” (Stalnaker, 1974, p. 52, footnote 1). The advantage of such an approach is that it takes into account the fact that communication requires certain information not to be explicitly expressed. The function assigned to presuppositions is to make communication more effective, since they avoid redundancy by making it possible not to state certain information already taken for granted. They allow the exchange to proceed by enabling the various protagonists of a conversation to identify the situations compatible with the assumptions and background convictions shared by all. In other words, presuppositions determine the linguistic context of the utterances and thereby become the pragmatic vectors of the constraints defining “what can reasonably or appropriately be said in that context” (Stalnaker, 1974, p. 53). But what distinguishes Stalnaker’s thought is his insistence that presupposition corresponds to a “propositional attitude of the speaker” (Stalnaker, 2002, p. 701) which consists in “[behaving] in one’s use of language as if one had certain beliefs, or were making certain assumptions” (1974, p. 52). It follows that presupposition no longer links, as with Fillmore, two linguistic entities, but a person and a proposition (1973, p. 447), since it is the speakers who are endowed with this “linguistic disposition” (1973, p. 202), inducing them to access the shared beliefs that underlie their conversation. Stalnaker therefore proposes a cognitive and social interpretation of presupposition, which has the effect of considerably increasing the number and diversity of language phenomena considered as presuppositional, since these are directly related to all the background beliefs strategically engaged in the conversation by the protagonists. In doing so, Stalnaker extended the program expounded by Grice in his William James Lectures (1989), as he outlined a description of discourse “as a sequence of intentional actions with a certain recognized purpose and direction” (Stalnaker, 2002, p. 702), a description that explains, without overloading semantics, “how rational agents choose the means to accomplish their ends” (Stalnaker, 2002, p. 702). But by situating presupposition on the side of cognition, Stalnaker also illustrates the very strong movement that was to lead many theoreticians of the philosophy of language to the philosophy of mind. As the heirs of Grice’s thought, Dan Sperber and Deirdre Wilson took the same path when they proposed to replace the Gricean principle of cooperation by the principle of relevance (Sperber & Wilson, 1986).
4. Toward Contemporary Thinking
Although all the lines of research defined since the 1960s are still being explored, two trends nevertheless characterize current work. In the wake of the work of Strawson, Kaplan, Putnam, and Lewis, thinking on the contextual parameters that influence the meaning of utterances is very active, leading to a sort of consensus in favor of contextualism. The second noteworthy orientation is the revival of the inferentialism advocated by Frege (1879) as a method of determining the meaning of sentences and their constituents. Concerning the formalization of languages, the notions of models and of possible worlds remain key concepts and constitute the theoretical framework used to represent semantic relations.
To consider that truth is not a semantic property of utterances, but that it belongs to pragmatic evaluation, is a displacement of the issue of truth proposed by the American philosopher Charles Travis (1943–). Taking the statement “The ball is round” in his article “Meaning’s Role in Truth” (Travis, 1996), Travis argues that: “what those words mean leaves it open for them to be used (in suitable circumstances) to say any of various things, each true under, and on, different conditions. There is no one set of conditions under which those English words, spoken of a given ball and time, would be, or say what is, true. Nor even one condition which is the condition for them to be true. If differences in truth condition make for different propositions, then what those words mean makes no one proposition the one (modulo referents) they express” (1996, pp. 454–455). Because the meaning of a sentence cannot, by itself, determine when it is true, Travis concludes that truth is the result of three factors: (a) the descriptions conveyed by words, (b) the state in which things are or would be; (c) the circumstances of use that determine how the statement could be satisfied (in relation to (a) and (b)). Thus, saying “The oven is hot” (with an oven at 140°C) is wrong if you want to cook a pizza, but true for the purpose of removing the rack bare-handed from the oven. Therefore the truth depends on the purpose or the expectations carried by the statement:
“it would be useless to be told that the oven is hot if we had no idea of the standard by which that was to be judged. Our perceptions of purposes to be served, and uses words would have in serving them, provide just such a standard. They thus allow descriptions of things to be the sort of good to us they are. These perceptions of occasions are perceptions of what it would be, on them, for a given description to describe truly, or for words which give it to state truth; to provide information which is correct. Their structure thus reveals some ingredients in truth, or what we are prepared to recognize about it. Part of the idea of truth is that a description (of something), to be true, must satisfy a general condition different in kind from conditions to the effect that what is described as thus must be as thus described: it must serve all the purposes that must be served (for truth) on that occasion, by having all the uses it ought in serving them.”(Travis, 1996, pp. 462–463)
By emphasizing that truth, depending on the occasion, calls for different standards, Travis defends a radical contextualism representative of the current movement.
4.2 The Inferential Approach to Meaning
Immanuel Kant (1724–1804), Gottlob Frege, and the later Wittgenstein of Philosophical Investigations are the main philosophers regarded by contemporary proponents of the inferentialist conception of meaning as their ancestors. From Kant, they keep the idea that semantic analysis does not concern terms, but rather the judgement expressed by the proposition. Moreover they agree with the Kantian view that judgement, which is a conceptually articulated activity, is distinguished by its normative character and by the fact that we are responsible for it. The interest of inferentialists in the later Wittgenstein lies in his contributions to these two theses. On the one hand, by developing a conception of meaning based on use, Wittgenstein made it possible to exploit the Kantian idea of judgement as being the only place where speech acts can be performed. On the other hand, his reflections on rules clarify the nature and the scope of the concept of normativity by showing that it is the result of social practices and that it intervenes in the attribution of intentional states. In order to represent semantically the articulation of judgement and the speaker’s responsibility, inferentialists use the notion of inference that Frege put forward in his Begriffsschrift (1879). Specifically, they keep his idea—at the propositional level—that a good inference never leads from a true claim(able) to one that is not true, and the use—at the subsentential level—of substitutional inference to determine the semantic role of concepts through their mutual articulations. The proposition “That’s a boche” allows the French speaker, who experienced the Second World War, to perform the substitutional inferences of “boche” with “German,” “barbarian,” or “person more prone to cruelty than other Europeans” (Brandom, 1994, pp. 69–72). These substitutions, that highlight the semantic content associated to the concept “boche,” by inferentially revealing its articulation with some other concepts, point out the function (here a depreciative one) of this word in a reasoning process. In doing this, inferentialists do not only hold that “to grasp or understand such a concept is to have practical mastery over the inferences it is involved in—to know, in the practical sense of being able to distinguish (a kind of know-how), what follows from the applicability of a concept, and what it follows from” (Brandom, 1994, p. 48), but also satisfy W. Sellars’s claim according to which “what was needed was a functional theory of concepts which would make their role in reasoning, rather than supposed origin in experience, their primary features” (Sellars, 1975, p. 285) as is the case in a referentialist approach. The responsibility argument leads Brandom to advocate an approach toward meaning in which enunciating a proposition (not only characterized by its content but also by its force) consists in “placing ourselves and each other in the space of reasons, by giving and asking for reasons for our attitudes and performances” (Brandom, 1994, p. 5). Brandom explains this theoretical position in these terms: “Saying or thinking that things are thus-and-so is to undertake a very distinctive kind of inferentially articulated commitment: putting it forward as a fit premise for further inferences, that is, authorizing its use as a premise, and undertaking responsibility to permit oneself that commitment, to defend one’s authority, under suitable circumstances, paradigmatically by exhibiting it as the conclusion of an inference from other such commitments to which one is or can become entitled” (Brandom, 1994, p. 11). Brandom’s inferential position has several important consequences. Firstly, we are obviously no longer in the realm of reference and therefore in the realm of the representation of an individual or a state of the world, but rather in the realm of expression and hence observation of use. In addition, we no longer proceed in a “bottom-up” way, starting from the meaning of words to obtain the meaning of the sentence, but instead we adopt a “top-down” approach, since it is considered—following Kant and Frege—that only a statement can act both as premise and conclusion of an inference and that “naming is not an action that makes one answerable to anything” (Brandom, 1994, p. 13). In other words, the act of speaking, such as it is actually performed in use, is first to determine the meaning of the words that compose it, by making explicit what the speaker is committed to when he chooses to use one word or another. This means that to understand the meaning of a word is the same as to master “its inferential use” (Brandom, 1994, p. 11). As a result, inferentialists are no longer in an atomistic conception of meaning but in one that is holistic, since in order to understand and to use any word, we must at least master some of its inferential relationships with some other words. Consequently, to master any single concept, one must master many. Thus, to know what “red” means is also to know that this term is a predicate, which designates not an object but a property, a property and not a relationship or a substance. It is also to know that from the statement “This is red,” we can infer “This is colored” or “This is extended,” but not “This is green” or “This is round.” Finally, it is to know that, depending on the circumstances, if the statement “This is red” is properly asserted, then some nonverbal behaviors are prescribed, while others are prohibited: if, in a car, my passenger says “The light is red,” I must for example stop the vehicle, and not accelerate. This example highlights the fact that the inferential perspective does not solely consider the logical inferences, but also and above all the inferences that are called “material” (i.e., whose validity depends on the semantic content of the concepts involved). The condition for the success of speech acts rests on the common stock of inferential relations that the speakers have learned to establish from their social interactions. In this perspective, the role of semantics is to elucidate the underlying commitments and entitlements of a speech act by revealing the inferential network linked to the propositional content expressed by the speaker. That is, as expressed by Brandom, “semantics must answer to pragmatics” by specifying all the propositions that can be deduced from the propositional content expressed, entitling a speaker and her listener in the social practice to give and ask for reasons. For the inferentialist perspective, reference is not a fundamental concept, but this does not imply that this approach “denies that there is an important representational dimension to word use” (1994, p. 28). Indeed, the role of this dimension is “to allow us to tell what the statements are true of, if they are true” (1994, p. 182). By thence taking part in the game of giving and asking for reasons, agreement on the existence of the same objects “plays an essentially social role” (Brandom, 1994, p. 183), and reference, which was conceived by the representational paradigm as having a pivotal role in semantics, henceforth concerns pragmatics by doing justice to the ontological relativity in languages, a thesis dear to Quine. Furthermore, to put inference at the heart of semantics has important consequences for how the role of logic is conceived. By assuming that meaning is constructed according to the inferences that we have implicitly learned to establish during our social interactions, the task of logic becomes to “help us to make explicit (and hence available for criticism and transformation) the inferential commitments that govern the use of our vocabulary, and hence articulate the contents of all our concepts” (1994, p. 30).
Links to Digital Materials
- A. J. Ayer. “Frege, Russell, & Modern Logic.”
- Daniel Bonevac’s lecture: “W.V.O. Quine, On What There Is,” The Analytic Tradition, Spring 2017.
- François Clementz’s lecture (Collège de France, Paris, 25/02/2009): “Russell, Wittgenstein et la genèse de l’atomisme logique.”
- Daniel Bonevac’s lecture: “Quine on Carnap on Logical Truth.”
P. F. Strawson and Gareth Evans on Truth:
- Robert Stalnaker MIT “The projection strategy.”
- Al-Saleh, C., & Bruno, A. (2006). Le débat Strawson/Austin sur la vérité: Théorie performative versus définition contextualiste. In J. Benoist (Ed.), Propositions et états de choses: Entre être et sens (pp. 119–230). Paris, France: Vrin.
- Alston, W. (1964). Philosophy of language. Englewood, NJ: Prentice Hall.
- Ambroise, B., & Laugier, S. (2009). Philosophie du langage: Signification, vérité et réalité. Paris, France: Vrin.
- Ambroise, B., & Laugier, S. (2011). Philosophie du langage: Sens, usage et contexte. Paris, France: Vrin.
- Anscombe, G. (1959). An introduction to Wittgenstein’s Tractatus. London, UK: Hutchinson University Library.
- Atlas, J., & Levinson, S. C. (1981). It-clefts, informativeness and logical form: radical pragmatics. In P. Cole (ed.), Radical pragmatics (pp. 1–61). New York: Academic Press.
- Austin, J. L. (1962). How to do things with words. Cambridge, MA: Harvard University Press.
- Austin, J. L. (1970). Philosophical papers. London, UK: Oxford University Press.
- Ayer, A. J. (1959). Logical positivism. Glencoe, IL: The Free Press.
- Bach, K., & Harnish, R. (1979). Linguistic communication and speech acts. Cambridge, MA: MIT Press.
- Baldwin, T. (2001). Contemporary philosophy: Philosophy in English since 1945. Oxford, UK: Oxford University Press.
- Baz, A. (2012). When words are called for. Cambridge, MA: Harvard University Press.
- Benoist, J. (Ed.). (2006). Propositions et états de choses: Entre être et sens. Paris, France: Vrin.
- Benoist, J. (2011). Frege philosophe de l’esprit. In S. Laugier & S. Plaud (Eds.), La philosophie analytique (pp. 55–68). Paris, France: Ellipses.
- Biletzki, A., & Matar, A. (Eds.). (1998). The story of analytic philosophy. London, UK: Routledge.
- Brandom, R. (1994). Making it explicit. Cambridge, MA: Harvard University Press.
- Brandom, R. (2000). Articulating reasons: An introduction to inferentialism. Cambridge, MA: Harvard University Press.
- Brandom, R. (2008). Between saying & doing: Towards an analytic pragmatism. Oxford, UK: Oxford University Press.
- Burkhardt, A. (1990). Speech acts, meaning and intentions: Critical approaches to the philosophy of John R. Searle. Berlin, Germany: Walter de Gruyter.
- Carnap, R. (1928). Der logische Aufbau der Welt. Berlin, Germany: Weltkreis-Verlag. English translation. (1967). The logical structure of the world and pseudoproblems in philosophy. London, UK: Routledge and Kegan Paul.
- Carnap, R. (1932). Überwindung der Metaphysik durch logische Analyse der Sprache, Erkenntnis, 2, 219–241. English translation, (1959). The elimination of metaphysics through logical analysis of language. In A. J. Ayer (Ed.), Logical positivism (pp. 60–81). Glencoe, IL: Free Press.
- Carnap, R. (1934). Die logische Syntax der Sprache. Vienna, Austria: Springer. Translation (1937). The logical syntax der sprache. London, UK: Kegan Paul.
- Carnap, R., Hahn, H., & Neurath, O. (1929). Wissenschaftliche Weltauffassung, der Wiener Kreis. Vienna, Austria: A. Wolf.
- Cavell, S. (1969). Must we mean what we say? A book of essays. Cambridge, UK: Cambridge University Press.
- Cavell, S. (1979). The claim of reason: Wittgenstein, skepticism, morality and tragedy. Oxford, UK: Oxford University Press.
- Cavell, S. (2005). Passionate and performative utterance. In S. Cavell & R. B. Goodman (Eds.), Contending with Stanley Cavell (pp. 177–198). Oxford University Press.
- Chapman, S. (2005). Paul Grice, philosopher and linguist. Basingstoke, UK: Palgrave Macmillan.
- Chapman, S. (2008). Language and empiricism after the Vienna Circle. Basingstoke, UK: Palgrave Macmillan.
- Chemla, E., & Spector, B. (2011). Experimental evidence for embedded scalar implicatures. Journal of Semantics, 28(3), 359–400.
- Chierchia, G. (2004). Scalar implicatures, polarity phenomena, and the syntax/pragmatics interface. In A. Belletti (Ed.), Structures and beyond (pp. 39–103). Oxford, UK: Oxford University Press.
- Chierchia, G., Fox, D., & Spector, B. (2012). Scalar implicatures as a grammatical phenomenon. In C. Maienborn, K. von Heusinger, & P. Portner (Eds.), Semantics: An international handbook of natural language meaning (Vol. 3, pp. 2297–2331). Berlin, Germany: De Gruyter.
- Clark, P., & Hale, B. (Eds.). (1994). Reading Putnam. Oxford, UK: Blackwell.
- Collins, J., & Dobler, T. (Eds.). (2018). The philosophy of Charles Travis: Language, thought, and perception. Oxford, UK: Oxford University Press.
- Colloque philosophique de Royaumont. (1962). La Philosophie analytique: 4ème Colloque philosophique de Royaumont 1958. Alençon, France: Éditions de Minuit.
- Crary, A., & Read, R. (2000). The new Wittgenstein. London, UK: Routledge.
- Davidson, D. (1967). Truth and meaning. Synthese, 17, 304–323.
- Davidson, D. (1969). True to the facts. Journal of Philosophy, 66, 748–764.
- Davidson, D. (1970). Semantics for natural languages. Reprinted in D. Davidson, Inquiries into truth & interpretation, 1985 (pp. 55–64). Oxford, UK: Oxford University Press.
- Davidson, D. (1985). Inquiries into truth & interpretation. Oxford, UK: Oxford University Press.
- Davidson, D. (1989). A coherence theory of truth and knowledge. In E. Lepore (Ed.), Truth and interpretation: Perspectives on the philosophy of D. Davidson (pp. 307–332). Oxford, UK: Basil Blackwell.
- Davidson, D. (1990). The structure and content of truth. Journal of Philosophy, 87, 279–328.
- Diamond, C. (1995). The realistic spirit: Wittgenstein, philosophy and the mind. Cambridge, MA: MIT Press.
- Donnellan, K. (1966). Reference and definite descriptions. The Philosophical Review, 75(3), 281–304.
- Donnellan, K. (1972). Proper names and identifying descriptions. In D. Davidson & G. Harman (Eds.), Semantics of natural language. Dordrecht, The Netherlands: D. Reidel.
- Donnellan, K. (1979). The contingent a priori and rigid designator. In P. French, T. Uehling, & H. Wettstein (Eds.), Contemporary perspectives in the philosophy of language. Minneapolis: University of Minnesota Press.
- Ducrot, O. (1972). Dire et ne pas dire. Paris, France: Herman.
- Dummett, M. (1973). Frege: Philosophy of language. London, UK: Duckworth. Second edition (1981). Cambridge, MA: Harvard University Press.
- Dummett, M. (1978). Truth and other enigmas. Oxford, UK: Clarendon Press.
- Dummett, M. (1993). Origins of analytical philosophy. Cambridge MA: Harvard University Press.
- Evans, G. (1976). The causual theory of names. Proceedings of the Aristotelian Society, 80, 187–208.
- Evans, G., & McDowell, J. (Eds.). (1976). Truth and meaning. Oxford, UK: Clarendon Press.
- Fann, K. T. (Ed.). (1969). Symposium on J. L. Austin. London, UK: Routledge & Kegan Paul.
- Fillmore, C. (1969). Verbs of judging: An exercise in semantics description. Paper in Linguistics, 1(1), 91–117.
- Floyd, J., & Shieh, S. (Eds.). (2001). Future pasts: The analytic tradition in twentieth-century philosophy. New York, NY: Oxford University Press.
- Foster, J. (1976). Meaning and truth theory. In G. Evans & J. McDowell (Eds.), Truth and meaning (pp. 1–32). Oxford, UK: Clarendon Press.
- Fox, D. (2007). Free choice and the theory of scalar implicatures. In U. Sauerland & P. Stateva (Eds.), Presupposition and implicature in compositional semantics (pp. 537–586). New York, NY: Palgrave Macmillan.
- Frege, G. (1879). Begriffsschrift, eine der arithmetischen nachgebildete Formelsprache des reinen Denkens. Halle, Germany: Nebert. Partial translation in Geach, P. T., & Black, M. (Eds.). (1970). Translations from the philosophical writings of Gotttlob Frege. Oxford, UK: Blackwell.
- Frege, G. (1884). Die Grundlagen der Arithmetik. Breslau, Germany: Koebner. English translation (2007). The foundations of arithmetic. New York: Pearson Longman.
- Frege, G. (1892). Über Sinn und Bedeutung. Zeitschrift für Philosophie und philosophische Kritik, 100, 25–50. Translation in Geach, P. & Black, M. (1970). Philosophical writings of Gottlob Frege (pp. 56–78). Oxford, UK: Basil Blackwell.
- Frege, G. (1893). Grundgesetze der Arithmetik (Vols. 1 and 2). Iena: Pohle. Partial translation in Furth, M. (Ed.). (1964). The basic laws of arithmetic. Berkeley: University of California Press.
- Frege, G. (1923). Gedankengefüge. Beiträge zur Philosophie des deutschen Idealismus, 3, 36–51. Translation in McGuinness, B. (Ed.). (1984). Gottlob Frege: Collected papers. Oxford, UK: Blackwell.
- Frege, G. (1977). Logical investigations (edited with a preface by P. T. Geach). Oxford, UK: Basil Blackwell.
- Geurts, B., & Pouscoulous, N. (2008). No scalar inferences under embedding. In P. Egré & G. Magri (Eds.), Presuppositions and implicatures (pp. 106–117). Boston, MA: MIT Working Papers in Linguistics.
- Glock, H.J. (1996). A Wittgenstein dictionary. Oxford, UK: Blackwell.
- Godart-Wendling, B., & Goubier, F. (2015). A la recherche d’une notion non stabilisée: La présupposition à l’aune des exponibilia et des approches contemporaines. Cahiers de lexicologie, 2(107), 203–222.
- Godart-Wendling, B., & Raïd, L. (2016). A la recherche de la présupposition, London, UK: ISTE Editions.
- Grice, P. H. (1989). Studies in the way of words. Cambridge, MA: Harvard University Press.
- Gupta, A. (1980). The logic of common nouns. New Haven, CT: Yale University Press.
- Hacker, P. (1975). Insight and illusion: Wittgenstein on philosophy and the metaphysics of experience. London, UK: Oxford University Press.
- Hacker, P. (1996). Wittgenstein’s place in twentieth-century analytic philosophy. Oxford, UK: Blackwell.
- Heijenoort, J. van. (1967). From Frege to Gödel. A source book in mathematical logic, 1879–1931. Cambridge, MA: Harvard University Press.
- Hughes, C. (2004). Kripke: Names, necessity, and identity. Oxford, UK: Oxford University Press.
- Kaplan, D. (1977). Demonstratives: An essay on the semantics, logic, metaphysics, and epistemology of demonstratives and other indexicals. Reprinted in J. Almog, J. Perry, & H. Wettstein (Eds.), Themes from Kaplan, 1989, (pp. 481–563). New York, NY: Oxford University Press.
- Kaplan, D. (1978). Dthat. In P. Cole (Ed.), Syntax and Semantics: Vol. 9.Pragmatics (pp. 221–243).
- Kaplan, D. (1979). On the logic of demonstratives. Journal of Philosophical Logic, 8(1), 81–98.
- Kaplan, D. (1990). Thoughts on demonstratives. In P. Yourgau (Ed.), Demonstratives. Oxford, UK: Oxford University Press.
- Karttunen, L. (1974). Presupposition and linguistic context. Theoretical Linguistics, 1, 3–44.
- Kempson, R. M. (1975). Presupposition and the delimitation of semantics. Cambridge, UK: Cambridge University Press.
- Kneale, W., & Kneale, M. (1962). The development of logic. Oxford, UK: Clarendon Press.
- Kripke, S. A. (1963). Semantical analysis of modal logic. Mathematical Logic Quarterly, 9(5–6), 67–96.
- Kripke, S. A. (1972). Naming and necessity. In D. Davidson, G. Harman, Dordrecht, D. Reidel Publishing Company (Eds.), Semantics of natural language. Reprinted in 1980 by Harvard University Press, Cambridge MA.
- Kripke, S. (1982). Wittgenstein: On rules and private language. Oxford, UK: Basil Blackwell.
- Laugier, S. (1992). L’anthropologie logique de Quine: L’apprentissage de l’obvie. Paris, France: Vrin.
- Lepore, E. (Ed.). (1986). Truth and interpretation: Perspectives on the philosophy of D. Davidson. Oxford, UK: Basil Blackwell.
- Lepore, E., & Ludwig, K. (2005). Donald Davidson: Meaning, truth, language, and reality. Oxford, UK: Oxford University Press.
- Levinson, S. C. (1983). Pragmatics. Cambridge, UK: Cambridge University Press.
- Lewis, D. (1968). Counterpart theory and quantified modal logic. Journal of Philosophy, 65(5), 113–126.
- Lewis, D. (1973). Counterfactuals. Oxford, UK: Blackwell.
- Lewis, D. (1979). Scorekeeping in a language-game, Journal of Philosophical Language, 8, 339–359.
- Lewis, D. (1983). Philosophical papers. Oxford, UK: Oxford University Press.
- Lewis, D. (1986). On the plurality of worlds. Oxford, UK: Blackwell.
- Lewis, D. (2008). Convention: A philosophical study. Hoboken, NJ: Wiley.
- Linsky, L. (1967). Referring. London, UK: Routledge & Kegan Paul.
- Linsky, L. (1977). Names and descriptions. Chicago, IL: University of Chicago Press.
- Ludlow, P. (1997). Readings in the philosophy of language. Cambridge, MA: MIT Press.
- Mey, J. (1985). Whose language? A study in linguistic pragmatics. Amsterdam, The Netherlands: John Benjamins.
- Mill, J. S. (1843). A system of logic: Ratiocinative and inductive, London, Longmans, Green and Co. Reprinted in 2002 by the University Press of the Pacific.
- Monk, R. (1991). Ludwig Wittgenstein: The duty of genius. London, UK: Vintage Books.
- Montague, R. (1968). Pragmatics. Reprinted in R. H. Thomason (Ed.), Formal philosophy: Selected papers of Richard Montague, 1974, (pp. 95–118). New Haven, CT: Yale University Press.
- Moore, G. E. (1966). Lectures on philosophy. London, UK: Allen & Unwin.
- Mounce, H. O. (1997). The two pragmatisms. London, UK: Routledge.
- Passmore, J. (1966). A hundred years of philosophy (2nd ed.). London, UK: Penguin.
- Putnam, H. (1975). The meaning of “meaning.” In Putnam (Ed.), Philosophical papers II: Mind, language and reality (pp. 131–193). Cambridge, UK: Cambridge University Press.
- Putnam, H. (1981). Reason, truth and history. Cambridge, UK: Cambridge University Press.
- Putnam, H. (1983). Realism and reason. Cambridge, UK: Cambridge University Press.
- Quine, W. V. O. (1934). “Ontological remarks on the propositional calculus.” Mind, XLIII(172), 472–476.
- Quine, W. V. O. (1936). Truth by convention. In O. H. Lee (Ed.), Philosophical essays for A. N. Whitehead (pp. 90–124). New York, NY: Longmans, Green.
- Quine, W. V. O. (1948). On what there is. Review of Metaphysics, 2, 21–38.
- Reprinted in Quine, W. V. O. (1953). From a logical point of view (pp. 1–19). Cambridge, MA: Harvard University Press.
- Quine, W. V. O. (1951). Two dogmas of empiricism. Philosophical Review, 60, 20–43.
- Reprinted in Quine, W. V. O. (1953). From a logical point of view (pp. 20–46). Cambridge, MA: Harvard University Press.
- Quine, W. V. O. (1953). From a logical point of view. Cambridge, MA: Harvard University Press.
- Quine, W. V. O. (1958). Le mythe de la signification. Reprinted in Cahiers de Royaumont. La philosophie analytique. 1962, Paris, France: Editions de Minuit.
- Quine, W. V. O. (1960a). Carnap on logic truth. Synthese, 12, 350–374.
- Reprinted in Quine, W. V. O. (1966). The ways of paradox and other essays (pp. 100–125). New York, NY: Random House.
- Quine, W. V. O. (1960b). Word and object. Cambridge MA: MIT Press.
- Quine, W. V. O. (1966). The ways of paradox and other essays. New York, NY: Random House. Second edition, enlarged (1976). Cambridge MA: Harvard University Press.
- Quine, W. V. O. (1969). Ontological relativity and other essays. New York, NY: Columbia University Press.
- Quine, W. V. O. (1970). Philosophy of logic. Englewood Cliffs, NJ: Prentice-Hall.
- Quine, W. V. O. (1974). The roots of reference. La Salle, IL: Open Court.
- Quine, W. V. O. (1990). Pursuit of truth. Cambridge, MA: Harvard University Press.
- Quine, W. V. O. (1991). Two dogmas in retrospect. Canadian Journal of Philosophy, 21, 265–274.
- Ramsey, F. P. (1927). Facts and propositions. Proceedings of the Aristotelian Society, Supplementary Volume 7, 6–20.
- Recanati, F. (2010). Truth-conditional pragmatics. Oxford, UK: Oxford University Press.
- Rivenc, F. (1998). Sémantique et vérité: De Tarski à Davidson. Paris, France: Presses Universitaires de France.
- Rorty, R. (Ed.). (1967). The linguistic turn. Chicago, IL: University of Chicago Press.
- Russell, B. (1903). Principles of mathematics. London, UK: Allen and Unwin.
- Russell, B. (1905). On denoting. Mind, 14(56), 479–493.
- Russell, B. (1918–1919). The philosophy of logical atomism. The Monist, 28, 495–527; 29, 190–222 and 344–380. Reprint in 2010, London, UK: Routledge.
- Russell, B. (1940). An inquiry into meaning and truth. London, UK: George Allen & Unwin.
- Sauerland, U. (2004). Scalar implicatures in complex sentences. Linguistics and Philosophy, 27, 367–391.
- Sbisà, M., & Turner, K. (2013). Pragmatics of speech actions. Berlin, Germany: De Gruyter Mouton.
- Schilpp, P. A. (Ed.). (1963). The philosophy of Rudolf Carnap. La Salle, IL: Open Court.
- Searle, J. R. (1958). Proper names. Mind, 67, 166–173.
- Searle, J. R. (1962). What is a speech act? In M. Black (Ed.), Philosophy in America. Ithaca, NY: Cornell University Press.
- Searle J. R. (1969). Speech acts. Cambridge, UK: Cambridge University Press.
- Searle, J. R. (1983). Intentionality: An essay in the philosophy of mind. Cambridge, UK: Cambridge University Press.
- Searle, J. R. (1995). The construction of social reality. New York, NY: The Free Press.
- Sellars, W. (1975). Action, knowledge and reality. Indianapolis, IN: Bobbs-Merrill.
- Soames, S. (2003). Philosophical analysis in the twentieth century (2 vols.). Princeton, NJ: Princeton University Press.
- Sperber, D., & Wilson, D. (1986). Relevance: Communication and cognition. Oxford, UK: Blackwell.
- Stalnaker, R. (1970). Pragmatics. Synthese, 22, 272–289.
- Stalnaker, R. (1973). Presupposition. Journal of Philosophical Logic, 2(4), 447–457.
- Stalnaker, R. (1974). Pragmatic presuppositions. In M. K. Munitz & P. K. Unger (Eds.), Semantics and philosophy: Essays (pp. 47–62). New York: New York University Press.
- Stalnaker, R. (1978). Assertion. In P. Cole (Ed.), Syntax and semantics 9. New York, NY: Academic Press.
- Stalnaker, R. (1999). Context and content. Oxford, UK: Oxford University Press.
- Stalnaker, R. (2002). Common ground. Linguistics and Philosophy, 25(5–6), 701–721.
- Strawson, P. F. (1949). Truth. Analysis, 9(6), 83–97.
- Strawson, P. F. (1950). Truth. Proceedings of the Aristotelian Society, Supplementary, 24.
- Strawson, P. F. (1950). On referring. Mind, 59, 320–344. Reprinted in (1971). Logico-linguistic papers (pp. 1–27). London, UK: Methuen.
- Stroll, A. (2000). Twentieth-century analytic philosophy. New York, NY: Columbia University Press.
- Tarski, A. (1933). Der Wahrheitsbegriff in den formalisierten Sprachen. Studia Philosophica, 1, 261–405. Translation in (1944). The semantic conception of truth. Philosophy and Phenomenological Research, 4, 341–375.
- Tarski, A. (1956). The concept of truth in formalized languages. In Tarski (Ed.), Logic, Semantics, Metamathematics. Oxford, UK: Clarendon Press.
- Tarski, A. (1967). Truth and proof. Scientific American, 220, 63–77.
- Travis, C. (1996). Meaning’s role in truth. Mind, 105(419), 451–466.
- Travis, C. (1997). Pragmatics. In C. Wright & B. Hale (Eds.), A companion to the philosophy of language. Oxford, UK: Blackwell.
- Urmson, J. O. (1956). Philosophical analysis: Its development between the two world wars. Oxford, UK: Oxford University Press.
- Van Fraassen, B. (1968). Presupposition, implication, and self-reference. Journal of Philosophy, 65(5), 136–152.
- Van Rooij, R., & de Jager, T. (2012). Explaining quantity implicatures. Journal of Logic, Language and Information, 21(4), 461–477.
- Waismann, F. (1945). Verifiability. Proceedings of the Aristotelian Society, Supplementary 19, 101–164.
- Wanderer, J. (2008). Robert Brandom. Stocksfield, UK: Acumen.
- Weiss, B. (2002). Michael Dummett. Stocksfield, UK: Acumen.
- Whitehead, A. N., & Russell, B. (1910–1913). Principia mathematica (3 vols.). Cambridge, UK: Cambridge University Press. 2nd ed., 1925 (Vol. 1), 1927 (Vols. 2, 3); abridged as (1962). Principia Mathematica to *56, Cambridge, UK: Cambridge University Press.
- Wittgenstein, L. (1921). Tractatus logico-philosophicus. In Annalen der Naturphilosophie. Reprinted (1922). London, UK: Routledge & Kegan Paul.
- Wittgenstein, L. (1953). Philosophical investigations. Oxford, UK: Basil Blackwell.
- Wittgenstein, L. (1958). The blue and brown books. New York, NY: Harper Perennial.
- Wittgenstein, L. (1969). On certainty. New York, NY: Harper & Row.
1. Here, we can put a diagram so that the reader becomes aware of the difficulty in reading Frege’s formalism.