ISSN 1897-1652 e-ISSN 2154-3747 POLISH JOURNAL OF PHILOSOPHY Volume XI, No. 1 (2017) Institute of Philosophy Jagiellonian University Kraków Published by the Institute of Philosophy of the Jagiellonian University POLISH JOURNAL OF PHILOSOPHY SPRING 2017 Articles Valeria Bizzari Aristotle, Phenomenology, and the Mind/Body Problem 7 Douglas Ian Campbell Against Lewis on ‘Desire as Belief’ 17 Bogna Choińska Following Desire as the Ethical Postulate of Psychoanalysis 29 Michael J. Shaffer Grounding Reichenbach’s Pragmatic Vindication of Induction 43 Adam Tamas Tuboly From ‘Syntax’ to ‘Semantik’ — Carnap’s Inferentialism and Its Prospects* 57 POLISH JOURNAL OF PHILOSOPHY Vol. XI, No. 1 (Spring 2017), 57-78. From ‘Syntax’ to ‘Semantik’ — Carnap’s Inferentialism and Its Prospects* Adam Tamas Tuboly (a) Institute of Philosophy, Hungarian Academy of Sciences (b) University of Pécs Abstract: The aim of this paper is to provide context for and historical exegesis of Carnap’s alleged move from syntax to semantics. The Orthodox Received View states that there was a radical break, while the Unorthodox Received View holds that Carnap’s syntactical period already had many significant semantical elements. I will argue that both of them are partly right, both of them contain a kernel of truth: it is true that Carnap’s semantical period started after his Logical Syntax of Language — in one sense of semantics. But it is also true that Carnap had already included semantical ideas in LSL: though not (just) in the sense that URV maintains. This latter sense of semantics is related to what is usually called inferentialism, and by getting a clearer picture of Carnap’s original aims, context, and concept-usage, we might be in a better position to approach his alleged inferentialism. 1. Introduction: The Scene and the Heroes Carnap held many views, though still not as many as Bertrand Russell — nonetheless, orthodox views about these changes emerged and stuck around during the last quarter of the twentieth century. One of them maintained that there was a radical break between Carnap’s syntactical and semantic period, the latter being initiated alone by Alfred Tarski. As A. J. Ayer (1982, p. 52) said, “Carnap’s eyes were dramatically opened at a congress in Paris in * This paper was supported by the MTA BTK Lendület Morals and Science Research Group; and by the János Bolyai Research Scholarship of the Hungarian Academy of Sciences. I am indebted to Jan Woleński and the Carnap Archives at Los Angeles (Rudolf Carnap papers (Collection 1029). UCLA Library Special Collections, Charles E. Young Research Library) and at Pittsburgh (Rudolf Carnap Papers, 19051970, ASP.1974.01, Special Collections Department, University of Pittsburgh) for the permission to quote the archive materials. All rights reserved. All the translations from the archive files are mine. I cite the Pittsburgh Archive files as follows: RC XX-YY-ZZ where XX is the box number, YY the folder number, and ZZ the item number; the UCLA archive as Carnap 1957, [UCLA] followed by box, folder, and page numbers. All the translations from the German letters are mine; the original, often idiosyncratic and ungrammatical English sentences of Carnap and Neurath’s letters are preserved. DOI: 10.5840/pjphil20171115 58 Adam Tamas Tuboly 1935 when Tarski presented an abstract of his semantic theory of truth.” This view may be labeled as the Orthodox Received View (ORV) of Carnap’s philosophical development. It was pointed out recently that this picture is highly oversimplified; furthermore, certain parts of it are entirely false. These new developments — reaching their peak in the 1990s — seem to be the new orthodoxy, thus I will refer to them as the Unorthodox Received View (URV). I will argue that both of the received views are partly right, that is, both of them contain a kernel of truth: it is true that Carnap’s (1937) semantical period started after his Logical Syntax of Language (LSL; originally in German, 1934) — in one sense of semantics. But it is also true that Carnap had already included semantical ideas in LSL: though not (just) in the sense that URV maintains. This latter sense of semantics is related to what is usually called inferentialism, and by getting a clearer picture of Carnap’s original aims, context, and concept-usage, we might be in a better position to approach his alleged inferentialism. One more bit of context: recently Jaroslav Peregrin (2014 and forthcoming) has uncovered the logical side of Carnap’s LSL and its relation to contemporary inferentialism. As he (forthcoming) argues, “from the current vantage point […] Carnap’s investigations from LSL are much more interesting than even the later Carnap himself would have appreciated.” The recent paper is not so much about the logical technicalities and subtleties of Carnap’s work, but about the historical settings: some more context and historical evidence will be provided for the recent understanding of Carnap. 2. The Orthodox Received View ORW has stuck around until the late 1980s, though one might still encounter it in textbook-type presentations. The main idea behind ORW is that Carnap’s characteristic ideas are either flawed, or turned out to be wrong. There are four main points that seem to be the core of this view and they could be made explicit as follows: (A) During the early 1930s, Carnap embraced the approach of ‘logical syntax’ that holds that every philosophical problem is a linguistic (syntactical) problem, and it deals exclusively with formal structures and syntactical categories, without taking into account any type of meaning and/or reference. “Philosophy is to be replaced by the logic of science,” said Carnap (1937, p. xiii), “that is to say, by the logical analysis of the concepts and sentences of the sciences, for the logic of science is nothing other than the logical syntax of the language of science” (emphasis in the original). In the year of the Syntax book, Carnap (1934a, p. 29) even From ’Syntax to ’Semantik’ 59 declared that he and his Viennese friends “pursue Logical Analysis, but no philosophy” (emphasis in the original). (B) The conception presented in Logical Syntax is full of serious philosophical and technical mistakes and errors; therefore the approach based on logico-syntactical analyses of language gives us a highly defective tool: it was doomed to failure. (C) There is a radical break in Carnap’s development: he was convinced by Alfred Tarski to give up his syntactical approach in favor of semantics even before the ink was dry on the Syntax-book. (D) Finally, Carnap accepted the semantic method, and quickly worked out various semantical systems, using such notions as ‘truth’, ‘truthcondition’, ‘reference’, ‘denotation.’ The Orthodox Received View was eagerly defended and propagated, from different perspectives, by A. J. Ayer (1982), Max Black (1945, p. 172), Ernest Nagel (1942, p. 469), Maria Kokoszyńska (1936 and 1939, p. 118) and Otto Neurath. Some of them thought that Carnap was guilty of accepting semantics despite its obvious unempiricist and metaphysical underpinnings and/or implications. For example, Black (1945, p. 172) said that “[i]t comes as something of a shock […] to find these new studies re-admitting wholesale those ostensible referents of symbolism which had seemed to have been permanently banished from ‘scientific philosophy’ [in LSL].” The American logical empiricist Ernest Nagel (1935, p. 357) also admired LSL, and thus raised in his review of Introduction to Semantics two important problems: first, it seemed that Carnap’s approach to designation leads to unnecessary and problematic metaphysical hypostatization, and secondly that “it appears doubtful […] whether a semantics which deliberately abstracts from all reference to the users of a language has much to contribute to the resolution of general philosophical issues” (Nagel, 1942, p. 472). Even before Nagel’s review of the book, Carnap was aware of his dismissal of the new semantical project. In Introduction to Semantics (1942, p. xi), Carnap stated that “while many philosophers today urge the construction of a system of semantics, others, especially among my fellow empiricists, are rather skeptical.” In the Neurath-Carnap correspondence, this is explained as follows: “When I wrote about fellow empiricists making objections against Semantics (in the Preface of vol. I) I was, of course, thinking in the first place of you and Nagel.”1 Carnap, as he claimed in the same letter, did not name or quote them because they did not publish anything that was explicitly against semantics. Nagel planned, however, to publish a paper on Charles S. Peirce around 1939: he aimed to attacked Carnap’s way of pursuing semantics, but 1 Carnap to Neurath, May 11, 1943 (RC 102-55-01). 60 Adam Tamas Tuboly it did not appear since The Journal of Unified Science (the follow-up of Erkenntnis) was cancelled during the war.2 Nonetheless, since the paper was read at the 1939 International Congress for the Unity of Science (at Harvard), Carnap was aware of its content. Nagel (1939/1954, p. 96) emphasized Peirce’s behavioristic treatment of language: “neither terms nor statements can be regarded as designating, independently of the habits involved in their use” (emphasis in the original). Besides reconstructing Peirce’s philosophy, Nagel accepted and recommended it with regard to the newly emerged semantics of Carnap as well; there is a long passage about how Nagel incorporated his fears into the narrative of the history of logical empiricism, and since it is quite unknown, I will quote it at full length: Some have suspected, perhaps unjustly, that the recently inaugurated discipline of semantics will open wide the door for the rehabilitation of Bolzano’s Saetze-anSich, Meinong’s objectives, Russell’s subsistens, and allied conceptions of the referends of signs. Though such doctrines have had fruitful historical roles, I think it would be a retrograde step if modern logical empiricism were to revive them in a new form; for the greater strength and promise of the movement has been its interpretation of the abstract in terms of the concrete, and its resolute turning from speculations which have no ascertainable consequences in issues of observable fact. (Nagel, 1939/1954, p. 96. Emphases in the original.) Carnap (1942, pp. xi-xii) indeed knew about these fears and was inclined to dissolve them. He even referred to Nagel’s talk implicitly: friends of the movement, he said, “are afraid that a discussion of propositions […] and truth […] will open the back door to speculative metaphysics, which was put out at the front door.” In January 1950, Carnap wrote a letter to his friend Franz Roh, an art historian from Munich: in that letter he happily accepted Roh’s ‘psychological and sociological explanation of misunderstandings’ and used examples from his own life to confirm the narrative about how his old friends tried to save him from the new changes and conversions. “[Take] for example my transition from the period of my ‘Logischen Syntax der Sprache’ to the later semantical period to which my latest books belong,” wrote Carnap to Roh, “[…] my friends always came and said, ‘But you have put everything such a good way; you cannot suddenly overthrow everything!’”3 Because so many of his friends were criticizing him for the propagation of semantics and they did not understand that it was a practical decision and not a theoretical question, he wrote up a paper to explain his stance: it was “Empiricism, Semantics, and Ontology” (ibid.). Nagel’s paper was published later in his selected essays: Sovereign Reason and Other Studies in the Philosophy of Science; see Nagel (1939/1954). 3 Carnap to Franz Roh, January 14, 1950 (RC 102-34-06). 2 From ’Syntax to ’Semantik’ 61 Thus the targets of Carnap’s famous paper were Nagel and Neurath, and not just the usual suspect W.V.O. Quine. Though the above passage might be seen as a digression, actually it documents well Carnap’s basic attitude towards philosophy and his friends: make everything as clear as it can be and do not let misunderstanding influence your judgment and make you look for theoretical gaps and clashes when only practical matters surface. The question of the nature and status of semantics was for Carnap such a milestone that it deserved much more attention. From the Nagel-Neurath pair, the latter presented a more difficult case: it was quite difficult to untangle Neurath’s arguments against semantics, as they amalgamated the requirement of clarity for empirical languages, a vehement anti-metaphysical attitude, and certain considerations against non-pluralistic (totalitarian) tendencies in science and especially in politics. While LSL was Neurath’s “logical Bible,”4 Introduction to Semantics presented just “all the Aristotelian metaphysics in full glint and glamour,” and the same absolutist attitude that might be found (at least Neurath detected it) behind all those searches for the real, one world, for the only Truth that characterized the entire tradition from Plato, through Descartes and Kant, to German idealism and finally to fascism and Nazism. 5 Others, mainly the Polish logicians, maintained that the earlier Carnap was a bad guy. Perhaps due to either being under the wrong influences, or working in a sort of isolation, Carnap promoted mistaken and false ideas in his syntactical phase, and it took “a Tarski” to enter the picture and single-handedly transform him into an advocate of semantics. This narrative, though it is diametrically opposed to Neurath’s and Nagel’s story, provided its own ‘radical break’ point in Carnap’s philosophical and intellectual development. Be as it may — I am not here to judge which Carnap was “the true and right Carnap” — both groups (and sometimes even Carnap!) claimed explicitly or implicitly that there was a break in his intellectual development, centered on the idea of semantics, initiated by Polish logicians, or by one particular logician, namely Tarski. 3. The Unorthodox Received View Neurath to Carnap, November 25, 1934 (RC 029-10-07). Neurath to Carnap, January 15, 1943 (RC 102-55-02). On semantics and politics see Neurath to Carnap, September 15, 1943 (RC 102-55-03), this letter contains the term “Plato-Hitler”) and September 22, 1945 (RC 115-07-66, unsent). Cf. Reisch, 2005. 4 5 62 Adam Tamas Tuboly Beginning with the works of Alberto Coffa (1976; 1991), Thomas Ricketts (1996), Stewe Awodey (2007), and Richard Creath (1990a; 1996; 1999), a new understanding of Carnap emerged. Creath (1996, p. 251) argued that “for the most part, […] Carnap’s syntax is really semantics or at least that the difference between what Carnap provides and a full-blown semantics would be infinitesimal were there infinitesimals.” Their points were the following: (1*) Even though Carnap argued for a syntactical conception of language, logic, and philosophy, his project involved many elements and techniques which later came to be called semantics. Creath summarized (1990a, p. 410) this as follows: “Not only is the logical consequence relation itself semantical, as we use the term, but so are truth tables, interpretation, and analyticity, all of which Carnap discusses. The treatment, especially of the last of these, is surprisingly close to a full semantical account.” Even Tarski noted in his “On the Concept of Logical Consequence” (1936/1956, p. 413) and in the “Historical Notes” for “The Concept of Truth in Formalized Languages” (1935/1956, pp. 277278) that Carnap achieved very similar results via very similar methods regarding truth and semantics. (2*) Carnap’s main metaphilosophical commitments (conventionalism, holism, or in Creath’s terms, functionalism) remained intact in the alleged change of views, indicating that his radical break was not that radical. Carnap suggested a pragmatic approach to logic, syntax, and semantics in his later works, as well as in his famous principle of tolerance in LSL (1937, §17). He claimed that semantics is to be regarded as a tool that is used to structure and systematize knowledge, and its use should be decided by its expedience (Carnap, 1950/1956, p. 221). It would be important to mention that there were scholars already in 1942, who claimed that — from a general point of view — “the continuity with Carnap’s previous thinking is evident from the beginning” since he ignores “psychologistic pseudoproblems” (J.M.R., 1942, p. 282).6 (3*) Therefore, given the metaphilosophical continuation and the presence of semantical-technical matters in LSL, the Unorthodox Received View holds that the change between Carnap’s syntactical and semantical period is a relatively small one, and not a radical break. Actually, Carus (1999) claimed that there isn’t any radical break between the syntactical and the semantical Carnap — or at least surely not in the sense of ORV — since all those conceptions, like theories of truth, coherence, correspondence, that are used by ORV to mark the gaps in Carnap’s thinking were rejected by Carnap entirely. 6 From ’Syntax to ’Semantik’ 63 Ricketts (1996, p. 231) even claimed, “the step is so small that we should ask why Carnap needed the impetus of Tarski’s work to take it.”7 I tagged this approach as “Unorthodox Received View” because even though the aforementioned scholars made many new, important, and revealing discoveries and remarks, calling our attention to a much more interesting picture and context of Carnap’s work, their views have apparently become the new orthodoxy. Nevertheless, there is still much more to Carnap’s syntactical and semantical period, and using a bit more context and materials, we might be able to show (at least as a first approximation) that we still have to refine our new Carnap-picture. 4. A Timeline: From ‘Metalogik’, through ‘Semantik’, to ‘Syntax’ In this section, in order to solve some conceptual issues and to see how Carnap’s systems and ideas evolved, I will discuss three notions that he used frequently in the early 1930s, namely ‘Metalogik,’ ‘Semantik’ and ‘Syntax.’ When dealing with the Wittgensteinian problem of how to account for and talk about language, Carnap (1963, pp. 53-54) had a famous “sleepless night” in January 1931, when he formulated his new project: “Versuch eine Metalogik,” i.e. “Attempt at a Metalogic.” In his first Vienna Circle lectures on metalogic, delivered in June, he defined the notion as “the theory of the forms which appear in a language, thus the representation of the syntax of language. In it one must not — to follow the formulation of the Warsaw group — make reference [Bezug] to the meanings [Bedeutung] of the signs” (Stadler, 2001/2015, p. 107).8 Thus “Metalogik” captured Carnap’s intention: he wanted to talk about language without considering the various non-verbal elements; however, since language is determined by logic, the first step was to talk about logic, to work out a metalogical framework. During 1931, Carnap was thinking about his new project under the title and category of “Metalogik.” This might be seen also in his famous “The Elimination of Metaphysics Through Logical Analysis of Language” paper: he talks there about the “metalogical” formulations of language (1932/1959, p. 62). By the time of 7 It should be mentioned that Tarski did not ‘open Carnap’s eye dramatically’ in 1935, since it is known that Carnap invited Tarski to Paris, and that he was involved in the German translation of Tarski’s 1933 truth paper. On this, see Gruber (2016) and Woleński (2017). 8 Carnap’s three lectures on metalogic, introducing his new ideas to the Circle, have been translated and published in Stadler (2001/2015, pp. 107-111 (first lecture), pp. 111-116 (second lecture), pp. 116-123 (third lecture and discussion)). 64 Adam Tamas Tuboly the publication of this paper, Carnap wrote from Prague to Schlick that “of my ‘Metalogic’, the main part is done.”9 Even though the title of the manuscript was still “Metalogik — Die logische Syntax der Sprache” in 1932, important changes were lurking around. He reported to Schlick in June that he “accepted the suggestion of Gödel and Behmann to replace ‘Metalogic’ with the word ‘Semantics.’”10 It is known from Carnap’s correspondence that he asked Gödel (January 1) to read the manuscript and make a “few critical remarks.”11 Gödel was interested and Carnap sent him via Hempel the “Metalogik” manuscript (February 23, 1932; Gödel, 2003, p. 343). Their letters dealt with Carnap’s definition of “analyticity,” but nothing was said about Syntax/Semantics, only Carnap’s diaries testify that Gödel suggested the term “Semantik”; actually it happened on the same day when Neurath proposed “Universal Syntax, logical foundations of Unified Science” (“Universelle Syntax, logische Grundlagen der Einheitswissenschaft”) as a new title to emphasize connections to the movement. 12 To see why Gödel might have suggested the term “Semantik,” another player shall enter the scene now. The German mathematician and logician Heinrich Behmann also got the manuscript before January 1932: that time Carnap asked him terminological questions, for example, about the title: “Is ‘Metalogic’ good? or is ‘logical’ Syntax better?”13 Behmann read the manuscript, made many technical remarks that will not concern us here, and responded to Carnap’s title-questions. Interestingly, Behmann indicated that, the problem was more than just a terminological issue. He claimed on March 23 that “metalogic” is too narrow a term, since what captures Carnap’s intentions is not an investigation about logic, as the term would suggest, but about language, so analogously a term like “Meta-Sprache,” “Meta-language” would be a better fit.14 However, three days later (March 26), Behmann added some notes to the same letter; he came up with a new and better suggestion: what actually captures Carnap’s new project is “’Semantics, as ‘the study of representation [Darstellung] through signs’” (ibid.). The term “Semantik” stuck with Carnap: in the two lectures that he prepared for Hans Reichenbach’s Berlin-colloquium in July 1932, he talked about the “Formal Questions of Semantics” and “Semantics as the Carnap to Schlick, December 7, 1931 (RC 029-029-15). Carnap to Schlick, June 30, 1932 (RC 029-029-11). 11 Carnap to Gödel, January 1, 1932 (RC 115-08-06). 12 Carnap’s diary entry and margin note, March 26, 1932 (RC 025-75-10). 13 Carnap to Behmann, January 20, 1932 (RC 115-10-16). 14 Behmann to Carnap, March 21, 1932 (RC 115-10-18). 9 10 From ’Syntax to ’Semantik’ 65 Foundation of Scientific Philosophy.”15 In the second lecture, we read the following: What is semantics? Hilbert’s metamathematics or proof-theory [Beweistheorie]: signs as objects = semantics of mathematical systems of formulas. Semantics = theory of forms of the complete language [Formtheorie der Gesamtsprache], combinatorics of the sign-complexes, without reference [Bezugnahme] to the referents [Bedeutung] of signs! (Like chess). Calculus. 2 parts: 1) Syntax, 2) theory of deduction [Schlusslehre] […] Thesis: Scientific Philosophy is Semantics. […] […] 2) Philosophy handles the language-forms, without reference to the referents! Example: Consequence-relation [Folgebeziehung], the most important concept of Semantics. (RC110-07-23, p. 1. Underlining in the original.) Even though Carnap crossed “Semantik” in the titles and wrote “Syntax” everywhere, there was apparently a point when he was more than positive about semantics: it was just the foundation of scientific philosophy! This period, however, might be characterized as “oscillation”: in the reply to his fellow Vienna Circle member Edgar Zilsel and to the Gestalt psychologist Karl Duncker, written mainly during the summer of 1932, Carnap (1932, p. 177) mentioned semantics, logical syntax, and metalogic at the same time, since all of them study the structure of sentences of a given language. At this time ‘metalogic’ was the least favored term and Carnap was caught between ‘syntax’ and ‘semantics.’ In April 1932, he replied to Behmann’s letter as follows: “I like the term ‘semantics’; Gödel also suggested it. Neurath, on the other hand, thinks it is unappealing and pedantic; he suggests ‘syntax.’ In order to avoid confusion with syntax in philology, one would probably often have to call it ‘logical syntax.’”16 He wrote the same to Schlick: “I replaced the word ‘metalogic’ with ‘semantics’, accepting Gödel’s and Behmann’s suggestion.”17 Furthermore, in the table of contents, written for an earlier version of the Syntax book, Carnap had the following handwritten note at the title (which was at that time “Metalogik: Die logische Syntax der Sprache”): “Later ‘metalogic’ was replaced by ‘semantics’, but since Neurath rejected the word ‘semantics’, I kept ‘syntax’” (RC 110-04-07, p.1). Nevertheless, until the end of 1932, Carnap referred to his manuscript (almost) always as ‘Semantik’ in his diaries. Though in February 1933 he 15 “Formale Fragen der Semantik Syntax (Für Vortrag in Reichenbachs Seminar, Berlin, 1.7.32.” (RC 110-07-20); and “Die Semantik Syntax als Grundlage der wissenschaftlichen Philosophie. Referat in Reichenbachs Kolloquium, Berlin, 4.7.32.” (RC 110-07-23). 16 Carnap to Behmann, April 17, 1932 (RC 115-10-19). 17 Carnap to Schlick, June 30, 1932 (RC 029-29-11). 66 Adam Tamas Tuboly wrote to Quine that during his “Logik II” course in Prague he “will discuss [the] new research on logical syntax (metalogic, semantics),”18 the primary category of identification became ‘syntax’ after that spring. Actually, when Carnap delivered a talk at the meeting of the Vienna Circle (July 10, 1933), he talked only about syntax and argued that “(philosophy =) logic of science = Syntax” (RC 110-07-22). After these lectures — where Carnap presented the famous “principle of tolerance” for the very first time — the main ideas of LSL were settled just as we know them today. Before I move on, one more thing should be mentioned. In his intellectual autobiography, Carnap (1963, p. 54) remembered the discussed changes as follows: “At that time I defined the term ‘metalogic’ as the theory of the forms of the expressions of a language. Later I used the term ‘syntax’ instead of ‘metalogic’ […].” ‘Semantics’ is obviously and notably missing from these remarks — thus vindicating the narrative of ORW. The original and unpublished autobiography, however, is a bit different: “At that time I defined the term ‘metalogic’ as ‘the theory of the forms of the expressions of a language, that is, the description of the syntax of language’ […]. Later I used instead of ‘metalogic’ the term ‘semantics’; but I soon abandoned it in favor of ‘syntax’ […].”19 Thus vindicating URV. It would be interesting to hypostasize about why Carnap deleted that clause, but that may lead too far into the history of American philosophy of science and language. What conclusions shall we draw? (1) Though Carnap had various names for his project, all of them captured something. ‘Metalogik’ captured the idea that language is to be accounted and approached from a logical point of view. ‘Semantik’ was apt because it was connected to the idea (as we will see in the following sections) that Carnap wanted to talk about meaning and content, and these were semantical notions. Finally, ‘Syntax’ captured the idea that philosophy deals only with structures, formal rules and not extra linguistic entities. The different names thus documented the changes and development of Carnap’s perspectives and emphases. 5. “Semantik” in the 1920s and 1930s Around the fin-de-siècle and shortly after, many words and concepts were in use to mark the studies of language, meaning, and reference: semasiology, rhematic, sematology, glossology, comparative ideology, sensifics, significs, rhematology, semiotic, semiology, orthology, and science of idiom (Read, 1948; Burgess, 2008 and ms.; Woleński, 1999). Carnap to Quine, February 6, 1933; quoted from Creath (1990b, p. 109). Carnap, Rudolf 1957 [UCLA]: “Autobiography. Part Two. Philosophical Problems.” Box 2, CM3: M-A4. 18 19 From ’Syntax to ’Semantik’ 67 The application of some of these became regular only in the second half of the 19th century and they covered the investigations of the historical changes of word-meaning as an extension of the etymological method, or the method of dealing with sentences and how they function as signs in communication. So the study of language had many forms and various conceptions in actual usage, but none of them formed a detailed research program or a discipline among philosophers in continental Europe. “Semantik” and closely related terms came to be used widely only relatively late in the 1920s, mainly by Polish logicians and philosophers. As Jan Woleński (1999 and 2009) pointed out, there were two main trends in the Lvov-Warsaw School. One of them originated from and was influenced by the works of Kazimierz Twardowski, a pupil of Franz Brentano. Among his most important students and colleagues, we find Jan Łukasiewicz, Stanisław Leśniewski, and Tarski himself. All of them understood semantics as a study of the word-world relation: “[P]olish philosophers, influenced by Brentano via Twardowski,” said Woleński (2009, p. 52), “transformed the intentional conception of mental acts into a referential treatment of language.” Interestingly, Neurath wrote the following to Carnap in January 1943: Scholasticism created Brentanotism, Brentano beget Twardowski, Twardowski beget Kotarbinski, Lukasiewicz (you know his direct relations to the NeoScholasticism in Poland), both together beget now TARSKI etc and now they are God fathers of OUR Carnap too, in this way THOMAS AQUINAS enters from another door Chicago, where he entered already via [Mortimer] ADLER. (Neurath to Carnap, January 15, 1943. RC 105-55-02.) For Neurath, this line of heritage represented such a grandiose metaphysical inheritance that it threatened logical empiricism with pseudo-scientific system building. He expressed his fears in the context of semantics and the new correspondence-like theory of truth claiming that a new “METAPHYSICA MODO LOGISTICA DEMONSTRATA,” that is, a new metaphysics, based on strict logic, is hanging over the movement’s head, threatening even its elementary integrity. But that is another story.20 The referential treatment of language is clearly seen in Tarski’s ideas: for him, the semantic method involved such concepts as denotation, satisfaction, definition, and truth. Most of these items were simply neglected at the time by those English and French-speaking scholars who tagged their projects with the above-mentioned categories of semantics. John Burgess (ms.) even claimed that “Tarski could have called what he actually calls ‘semantic’ notions ‘referential’ notions instead.” It is quite interesting that he excluded almost all talk of meanings from his semantics. See Neurath to Carnap, July 1, 1936 (RC 102-52-23) and April 30, 1935 (RC 029-09-57). 20 68 Adam Tamas Tuboly One cannot find the term synonym, homonym, or ambiguity in Tarski, notes Burgess, “which surely would all have been near the top of a list of typical ‘semantic’ notions for [Michel] Bréal,” who was the father of semantics with his lectures and book at the end of the 19th century. In 1945, the famous lexicographer and scholar of English slang, Eric Partridge complained on the pages of the Journal of the Royal Society of Arts that the concept of semantics, developed by Bréal and others was […] not good enough for Professor Carnap and his Polish, Austrian and American fellow logicians. They must adopt it, as though no other designation could have been found to suit their non-philological, their psychological, especially their logistical (rather than logical) purposes, and adapt it to their non-linguistic ends. For them, the word [semantics] means something quite different from the denotation intended and explained by Bréal, Jespersen, and all other true philologists. (Partridge, 1945, p. 455.) So, obviously, not everybody was happy with the transformation of the old science of semantics — studying communication, information-production, signifying etc. — into the formal, mathematical and logical study of an abstract word-world relation. Carnap even noted in his unpublished autobiography that after the first 1935 Paris congress a reporter wrote about Tarski’s lecture as follows: “A learned philosopher came all the way from Warsaw to Paris in order to give to us the great revelation that snow is white if and only if snow is white.”21 Let’s stop for a minor and short digression. As can be seen from the remarks above, the opinion on logical semantics among professionals and laymen varied from acceptance to rejection and incomprehension. Another good example is the American economist, social theorist and popular writer Stuart Chase who published his Tyranny of Words in 1938. Propagating Alfred Korzybski’s theory of semantics, Chase (1938, p. 160) mentioned at one point that the University of Chicago Press will publish a new encyclopedia under the leadership of Otto Neurath, who “is a kind of pioneer in semantics. He believes in going to things wherever possible, rather than to words.” Neurath wrote to Carnap that after he moved to England, he rebuilt his library, which now included many new books, among them items from Stuart Chase.22 It is unknown whether Tyranny of Words was among them, or if it was, whether Neurath ever read it carefully (though the title might have caught his attention): what is sure, however, is that Neurath would never accept that he is a pioneer in semantics, or that he believed in going to things and not to words — or at least not without major restrictions and qualifications. 21 22 Carnap 1957 [UCLA], Box 2, CM-3: MA-4, M7-M8. Neurath to Carnap, July 17, 1942, (RC 102-56-04). From ’Syntax to ’Semantik’ 69 Thus it does not seem unreasonable that Carnap did not want to go public with the debate concerning semantics. In his opinion, it would not have helped their common cause. As he wrote to Neurath, “I think for the sake of the movement it would be much better if we were more tolerant towards each other. If your intolerance would become the general custom, then I am afraid you would be among the first to be declared a heretic and excommunicated.”23 The major problem was that they couldn’t even agree on what exactly they disagreed about. In Neurath’s narrative they had a theoretical debate with some political and social consequences; in Carnap’s eyes, however, they had some practical issues and he thought that “it would be best for the development of science if the people on the one side who see more the turbulent whirl of material in all its colorfulness and vagueness, and those on the other side who love nice structural schemata would not polemized against each other but rather realize that the work of both is necessary for science.”24 Neurath (1941/1983, pp. 221-222; 1944, pp. 9-14) did not publish much of his polemics against semantics, and even those papers that appeared did not become influential: nonetheless Neurath was quite good in building and maintaining relationships, organizing events and institutions, so Carnap might have feared that Neurath’s attacks against semantics could have influenced the movement behind the curtains. Let’s get back to the main line of argument of this paper. Besides the Tarski-style, referentially oriented version of semantics, there was another line of thought in Poland (Woleński, 2009): namely Leon Chwistek’s approach, which was called in 1924 “Semeiotics” and, later, “Semantik” in 1932/1933 when his paper appeared in Erkenntnis and when Carnap mentioned it in Syntax. Chwistek’s semantics was, however, quite different from Tarski’s and from the Twardowski-school: it is best compared to Hilbert’s theory of proof, to his formal method, since it disregards the relation of a sign to its reference; thus, as Carnap noted in LSL (§68), “Chwistek’s system of so-called semantics is, on the whole, dedicated to the same task as our syntax” (emphasis in the original). Again, what conclusion shall we draw now? Given the various terms and ranges of the scientific inquiry about language, meaning, and their relation to the extra-linguistic sphere, one shall distinguish two very general trends (painted with broad strokes): one type of semantics concerned the word-world relation, while the other treated the meaningful component, the information, and content-elements, which were conveyed through the use of signs. Notably, however, these trends were not always recognized or treated separately, or even if they were, the assessment of their value varied 23 24 Carnap to Neurath, January 29, 1943 (RC 115-07-62). Carnap to Neurath, February 4, 1944 (RC 102-55-04). 70 Adam Tamas Tuboly quite frequently, not helping the general causes and aims of Carnap and other defenders of the new semantical project(s). 6. Carnap’s usage of the terms In order to draw conclusions about Carnap, we have to note whether there is any congruence in his and others’ concept usage. As a start, let’s take Carnap’s official semantical writings. Both in Foundations of Logic and Mathematics (1939) and in Introduction to Semantics (1942), Carnap starts from the idea of semiotics as a tripartite inquiry of signs. The three levels are well known: pragmatics, semantics, and syntax. Pragmatics deals with signs and their users, in semantics we abstract from users and treat only the sign-world relation, and finally in syntax the relations of signs to signs become the primary target (Carnap, 1942, p. 9). This approach, actually, corresponds nicely to Charles Morris’s (1938) division of semiotic. But then we face some peculiarities! “With no real room for meanings under any label in [Morris’s] system,” says Burgess (ms.), “there is naturally no room in his sogenant semantics for anything like semantics as it had been conceived before World War I.” Even though Morris became one of the main propagators of semantics in the United States and in (some parts of) Europe, his semantics was not the one which was already established outside of Poland. The same seems to be true for the mature Carnap as well, after he adopted Morris’s program. More importantly, however, Carnap’s definition of semantics is very similar to that of Tarski. In his “The Concept of Truth in Formalized Languages,” Polish edition 1933, German 1935, Tarski defined semantics as follows: “a characteristic feature of the semantical concepts is that they give expression to certain relations between the expressions of language and the objects about which these expressions speak” (Tarski, 1935/1956, p. 252). Interestingly though, while Carnap and Morris always started their discussion with the definition of ‘semantics’, Tarski defined it only at a very late point in the last part of his paper, thus somehow indicating that ‘semantics’ need not be defined (given presumably its well-known character). Viewing matters from this perspective, it seems to be obvious that Carnap was under the philosophical and technical aegis of both Morris’s and Tarski’s semantics. It should be kept in mind, however, that there were important distinctions between Morris and Tarski. On November 22, 1936, Carnap visited Morris and talked with him about the International Encyclopedia of Unified Science and Tarski’s semantics. He noted the following in his diaries: “Morris seems to be infected by Neurath’s qualms, takes the notions, even my clarifications to be metaphysical, he From ’Syntax to ’Semantik’ 71 wants another — ‘semantics’ (?).”25 Even though Morris fought for semantics, he was committed to behaviorism as well: just like Neurath. This relation is made explicit in Morris’s controversial Sign, Language, and Behavior, where he said that “a science of signs can be most profitably developed on a biological basis and specifically within the framework of the science of behavior (a field which, following a suggestion of Otto Neurath, may be called behavioristics)” (Morris, 1946, p. 1). Note, however, that this was a quite typical move among American pragmatists; Nagel said in his Harvard lecture that I can think of no better way to still these suspicions [about the metaphysical underpinning of semantics] than by placing the study of semantics into a behavioral context, and by instituting an analysis of such key semantic terms as ‘designation’ and ‘truth’ as used in specific contexts, in order to reveal the modes of action they signify. (Nagel, 1939/1954, p. 96. Emphasis in the original.) So, semantics had the same agenda for Nagel and Morris as for Tarski and Carnap, but with different accounting: semantic notions shall be treated not as logical, but as behavioral concepts, thus freeing the project from actual and/or possible metaphysical components. Seemingly, there was as much agreement as disagreement among those who pursued semantics under the banner of logical empiricism. It is not at all evident, after all, whether it was Morris or Tarski who had the most significant influence on Carnap after the mid-1930s. Given that Morris (1936) did not use ‘semantics’ or ‘syntax’ in his 1935 lecture at the International Congress for the Unity of Science in Paris,26 but used instead the concept of ‘existential dimension’ and ‘formal dimension of meaning’, and given that Tarski presented his semantical ideas there in detail (invited by Carnap to do so), he might have influenced Morris somehow. At least Morris seems to have taken ‘semantics’ from Tarski and ‘syntax’ maybe from Carnap. In his unpublished lecture, entitled “Über den semantischen Wahrheitsbegriff” prepared for a discussion with Neurath in Paris, 1937, Carnap did not consider the tripartite approach of semiotic (so characteristic of Morris), or the peculiar logical ideas of Tarski. He was busy with the distinction of truth and confirmation (RC 080-32-01). Though the character of Carnap and Neurath’s debate could explain this, it still documents Carnap’s interests and concept-usage before his official semantical books. Note also that although Tarski was a regular figure in Carnap’s discussions (he even visited Carnap in Prague), the latter’s famous “Truth Carnap’s diary entry, November 22, 1936 (RC 025-82-01). The various events at and after the 1935 Paris Congress are described and analyzed from the viewpoint of the history of philosophy by Jan Woleński (forthcoming). 25 26 72 Adam Tamas Tuboly and Confirmation” paper (1949) — presented at the 1935 Paris Congress — mentions Tarski only once and doesn’t consider semantics per se, just the difference between truth and confirmation. What may cause some historical and interpretational trouble is that the 1949 English translation (prepared by “H.F.” which stands presumably for “Herbert Feigl”) is actually a mixture of the original “Wahrheit und Bewährung” paper and the 1946 “Remarks on Induction and Truth” paper (written for the discussion with Felix Kaufmann at a symposium) which deals in more detail with Tarski and semantics in the fashion and emphasis of the 1940s (Kaufmann presented Neurath-style arguments against Carnap’s semantics). But anyway, what is certain is that Carnap thought that semantics should deal with signs, especially linguistic artifacts and their denotation or referents in the world. Later, in his posthumously published “Notes on Semantics,” Carnap (1972, p. 8) called his semantics “pure, designative semantics,” dealing with the “designative meaning component” which is “relevant for questions of truth.” To move along, we shall ask the question whether this is the semantics that is to be found in Carnap’s LSL? Let’s see first what Carnap did not talk about. Obviously, he was not dealing with the referents, denotation, or, to be more precise, the Bedeutung of expressions — syntax shall abstract from these. Since ‘abstracting from the referents’ meant being formal, ‘formal semantics’ was just an oxymoron. Interestingly, Carnap did not talk about the meaning, that is, the ‘Sinn’ of expressions. Or to be more precise, he did not consider meanings to be separate, abstract entities in their own right, to be grasped and expressed; and he did not take meaning to be such a function or mode of presentation that determines the Bedeutung — from this perspective, Frege was not an ideal in Syntax. What Carnap did talk about is content, or in German, ‘Gehalt’/‘Inhalt’. He distinguished two components of expressions, namely Sinn/Gehalt/Inhalt and Bedeutung. Regarding their English translations, some confusion might arise. ‘Gehalt’ and ‘Inhalt’ is always translated as ‘content’; ‘Bedeutung’, however, is often rendered as ‘meaning’, but in many cases, ‘meaning’ is the word for ‘Sinn’, too. In other cases, ‘Sinn’ is just ‘sense’. As a result, often when Carnap speaks about how syntax abstracts from ‘meaning’, he means abstracting from ‘Bedeutung’, that is, the extra-linguistic element, not from ‘Sinn’, the content or information conveyed by expressions. Logical syntax does not consider Bedeutung, but what is the case with ‘Sinn’, ‘Inhalt’ or ‘Gehalt’? Carnap asks whether it is possible to approach and define this component from the formal, or syntactical point of view, without taking into account the Bedeutung. His account of the content of a proposition is quite surprising (from the point of view of ORV, and even From ’Syntax to ’Semantik’ 73 URV) and interesting: the content of a proposition is what we can learn from it, what it conveys to us, that is, and here is the point, ‘what we can deduce from S’, and ‘from what S follows.’ ‘Folgerung’, ‘entailment’, ‘consequence’ are formal notions in the sense that they do not rely on Bedeutung. As Carnap (1934b, p. 11) says in “On the Character of Philosophic Problems” (written for the first issue of Philosophy of Science), “the question, whether a certain proposition is an inference [Folgerung] of certain other propositions or not, is therefore completely analogous to the question whether a certain position in chess can be played from another or not.” Since content or Sinn is definable by Folgerung or consequence, it is also a formal notion. Actually, this is the main tenet of so-called inferentialism, as Jaroslav Peregrin (2014 and forthcoming) stated recently: the meaning of (logical and non-logical) expressions is exhausted by their inferential roles. By the content of a sentence S Carnap meant the class of non-valid consequences of S, or the non-analytic consequences of S. Part 4 of Semantik’s original manuscript contained a section entitled ‘Theory of the Content of Formulas’ with such subsections as ‘The General Theory of Contents’, ‘Logical Content’, ‘Physicalistic Content’, ‘Empiristic Content,’ and so on, but more importantly, its last section was: “Sinn = theoretischer Gehalt” (RC 110-04-07/p.3. IV/B/1-2). In fact, when he refers to his new “Attempt” manuscript in the “Overcoming Metaphysics” paper, Carnap (1932/1959, p. 62) says that “it is planned to give elsewhere a detailed exposition of metalogic as the theory of syntax and meaning [Sinnes], i.e. relations of deducibility [Ableitungsbeziehung].” This happened, after all, in the Syntax book, though under different names. 7. Concluding remarks The aim of this paper was to show that defenders of ORV and URV are both right in a sense: Carnap indeed had a radical break in the mid-1930s, after the publication of his Logical Syntax of Language. What surfaces in his later semantical writings — the semantical concept of truth, based on denotation, reference, satisfaction etc. — is missing from LSL. But it must be admitted as well that Carnap already had some semantical insights in the context of the syntax project. These were partly connected to his later works (as it was emphasized by URV), but mainly they constituted semantics in a different sense: namely, as semantics dealing with content, information and with the ‘Sinn’ of expressions. Referentially conceived semantics as 74 Adam Tamas Tuboly a research project was pursued only later, and it is a further question whether Tarski or Morris represented the major line for Carnap.27 Semantics for Carnap, in the early 1930s, was very similar to what later became called inferentialism, so, finally, I would like to point out some further historical and general ideas that may help us assessing Carnap’s inferentialism, and provide perhaps further discussion. Peregrin claimed (2014, pp. 27-28) that while Carnap accepted the usual empiricist dogma that the choice of conceptual frameworks or schemes is wholly conventional and arbitrary, and “the world can have its say only after we accomplish the choice,” Sellars and others “believed that the world influences our putting together the framework” (emphasis in the original). But that is something entirely accepted by Carnap! In his “Theoretical Questions and Practical Decisions,” published in the same year as Syntax, Carnap (1934c, p. 258) claimed that the world always has its say: while the development of linguistic frameworks, logical systems, and conceptual tools is a matter of convention, we are always restricted by already established empirical findings and by our wishes, aims, and goals pursued in a natural setting. In 1970, Carnap wrote the following to Gerhard Kroiss; it is worth quoting the letter at some length: The question how and where we should begin with the construction of our language and our system of concepts is indeed an important problem and a rather puzzling one. Wherever we might begin, we feel the need for a clarification of our initial concepts, and this a need to use other concepts which precede the “initial” concepts. This is the old puzzle of philosophy: how and where should we make the absolute beginning? I believe that the question in this form is unanswerable. It seems to me that the best description of the problem situation and of the required procedure was given by Neurath by his comparison with the task of repairing a ship. Our language, the whole conceptual system with which we wish to work, is like a ship on the open ocean; there is no dry-dock in which we could make a construction from an absolute beginning. We have only the ship itself, and we cannot proceed otherwise than to reconstruct parts while we are supported by the other parts. (Carnap to Gerhard Kroiss, February 5, 1970. RC 027-31-04. Typos corrected; pencil underlining by Carnap.) Carnap thought (together with Neurath) that the world always has its say; we cannot free ourselves entirely to find an abstract realm with free-floating conventions. Carnap embraced — as André Carus (2007) called it — a “dialectical picture”: setting up languages is a task for linguistic engineering and thus a practical task. How we set up our languages is based on how we set up our rules of formation and transformation. We may be Though Carnap always referred to Tarski, the tripartite conception of pragmatics, semantics, and syntax, and the conceptual settings of his works connect Carnap mainly to Morris. Furthermore, there are some interesting differences between Tarski and Carnap on which see Wagner 2017. 27 From ’Syntax to ’Semantik’ 75 motivated by theoretical virtues and facts of the matter, but these facts shall be captured already in a certain framework. So theories (i.e., knowledge and facts) influence our practices (building up languages), but our practices (e.g. our decisions) influence the theoretical level (since you can know only those things that can be expressed in a language). Peregrin (2014, p. 28) also says, that while for logical empiricists natural laws “must be simply empirical generalizations” and become like freefloating conventions, “Sellars noticed that, as a matter of fact, natural laws do take part in conferring roles on the terms they involve and hence that they do constitute concepts” (emphasis in the original). Though Carnap (1937, §82) accepts that we deal with conventions here, he says that they are not arbitrary. “The choice of them is influenced, in the first place, by certain practical methodological considerations […],” but later “all must be tested by experience.” Of course, testing is not a unidirectional process, since after testing “P-rules [that is, natural laws] can be altered […], or the protocol-sentence can be taken as being non-valid; or again the Lrules which have been used in the deduction can also be changed.” After all, for Carnap, the fact that “the hypotheses, in spite of their subordination to empirical control by means of the protocol-sentences, nevertheless contain a conventional element is due to the fact that the system of hypotheses is never univocally determined by empirical material, however rich it may be.” Thus it is not entirely correct to say, as Peregrin does, that Carnap did not consider empirical languages and expression in his Syntax-project. It is true, that he focused on logic and mathematics, but his P-rules, functioning as transformation rules, were just rules that determined the meaning and content of physical expressions via inferential patterns. Rules, rules, and rules: in the semantical writings, Carnap claimed that “rules determine the meaning or sense of the sentences” (1942, §7; emphases in the original). Whether to accept a rule or to construct a new one is a question of expedience, as was everything for Carnap, including semantics and syntax (in whichever sense). References Awodey, S. (2007). Carnap’s Quest for Analyticity: the Studies in Semantics. In Friedman, M. & Creath, R. (eds.), The Cambridge Companion to Carnap (pp. 226-247). Cambridge: Cambridge University Press. Ayer, A. J. (1982). The Vienna Circle. In Gadol, E. T. (Ed.), Rationality and Science. A Memorial Volume for Moritz Schlick in Celebration of the Centennial of His Birth (pp. 36-54). Wien/New York: Springer. 76 Adam Tamas Tuboly Black, M. (1945): Review of Introduction to Semantics and Formalization of Logic. Mind, 54 (214), 171-176. Burgess, J. (2008). Tarski’s Tort. In Mathematics, Models, and Modality. Selected Philosophical Essays. (149-168). Cambridge: Cambridge University Press. Burgess, J. (ms.). Episodes in the History of ‘Semantics.’ Unpublished conference talk for the Princeton Conference: ‘Semantics: Mathematics or Psychology?’ September 15, 2013. Carnap, R. (1932). Erwiderung auf die vorstehenden Aufsätze von E. Zilsel und K. Duncker. Erkenntnis, 3, 177-188. Carnap, R. (1932/1959). The Elimination of Metaphysics Through Logical Analysis of Language. In Ayer, A. J. (Ed.), Logical Positivism (pp. 6081). New York: The Free Press. Carnap, R. (1934a). The Unity of Science. (M. Black, Trans.) London: Kegan Paul. (Original work published in 1932.) Carnap, R. (1934b). On the Character of Philosophic Problems. Philosophy of Science, 1, 5-19. Carnap, R. (1934c). Theoretische Fragen und praktische Entscheidungen. Natur und Geist, 2 (9), 257-260. Carnap, R. (1937): Logical Syntax of Language. (A. Smeaton, Trans.) London: Routledge & Kegan Paul Ltd. (Original work published in 1934.) Carnap, R. (1949). Truth and Confirmation. In Feigl, H. & Sellars W. (Eds.), Readings in Philosophical Analysis (pp. 119-127). New York: Appleton-Century-Crofts, Inc. Carnap, R. (1950/1956). Empiricism, Semantics, and Ontology. Reprinted in Meaning and Necessity (pp. 205-221). Chicago: University of Chicago Press. Carnap, R. (1972). Notes on Semantics. Philosophia, 2, 3-54. Carus, A. (1999). Carnap, Syntax, and Truth. In Peregrin, J. (Ed.) Truth and its Nature (If any) (pp. 15-35). Dordrecht: Springer. Carus, A. (2007). Carnap in Twentieth-Century Thought: Explication as Enlightenment. Cambridge: Cambridge University Press. Coffa, A. (1976). Carnap’s Sprachanschauung Circe 1932. PSA, 2, 205-241. Coffa, A. (1991). The Semantic Tradition from Kant to Carnap: To the Vienna Station. Cambridge: Cambridge University Press. Creath, R. (1990a). The Unimportance of Semantics. PSA, 2, 405-416. Creath, R. (1990b). (Ed.). Dear Carnap, Dear Van: The Quine-Carnap Correspondence and Related Work. Berkeley: University of California Press. Creath, R. (1996). Languages Without Logic. In Giere, R. N. & Richardson, A. W. (eds.) Origins of Logical Empiricism (pp. 251-265). Minneapolis: University of Minnesota Press. From ’Syntax to ’Semantik’ 77 Creath, R. (1999). Carnap’s Move to Semantics: Gains and Losses. In Woleński, J. & Köhler, E. (eds.) Alfred Tarski and the Vienna Circle – Austro-Polish Connections in Logical Empiricism (pp. 65-76). Dordrecht: Kluwer Academic Publishers. Gruber, M. (2016). Alfred Tarski and the “Concept of Truth in Formalized Languages.” A Running Commentary with Consideration of the Polish Original and the German Translation. Dordrecht: Springer. J. M. R. (1942). Review of Carnap’s Introduction to Semantics. Philosophy of Science, 9 (3), 281-282. Kokoszyńska, M. (1936). Syntax, Semantik und Wissenschaftslogik. Actes du Congrès International de Philosophie Scientifique, fasc. 3 (pp. 9-14). Paris: Hermann & Édituers. Kokoszyńska, M. (1939). Review of Carnap’s Foundations of Logic and Mathematics. The Journal of Symbolic Logic, 4 (3), 117-118. Morris, Ch. (1936). The Concept of Meaning in Pragmatism and Logical Positivism. Actes du Congrès International de Philosophie Scientifique, fasc 8 (pp. 130-138). Paris: Hermann & Édituers. Morris, Ch. (1946). Sign, Language, and Behavior. New York: Prentice Hall. Nagel, E. (1935). Review of Logische Syntax der Sprache. Journal of Philosophy, 32, 49-52. Nagel, E. (1939/1954). Charles S. Peirce, Pioneer of Modern Empiricism. In Sovereign Reason and Other Studies in the Philosophy of Science (pp. 89-100). Glencoe, Illinois: The Free Press. Nagel, E. (1942). Review of Introduction to Semantics. The Journal of Philosophy, 39 (17), 468-473. Neurath, O. (1941/1983). Universal Jargon and Terminology. In Cohen, R. S. & Neurath, M. (Eds.) Otto Neurath: Philosophical Papers 19131946 (pp. 213-229). Dordrecht: D. Reidel. Neurath, O. (1944). Foundations of Social Sciences. Chicago: University of Chicago Press. Partridge, E. (1945). Review: Introduction to Semantics by Rudolf Carnap. Journal of the Royal Society of Arts, 93, 455-456. Peregrin, J. (2014). Inferentialism. Why Rules Matter. Basingstoke: Palgrave Macmillan. Peregrin, J. (forthcoming). Carnap’s Inferentialism. In Schuster, R. (ed.) Vienna Circle in Czechoslovakia. Dordrecht: Springer. Read, A. W. (1948). An Account of the Word ‘Semantics.’ WORD, 4 (2), 78-97. Reisch, G. (2005). How the Cold War Transformed Philosophy of Science: to the Icy Slopes of Logic. Cambridge/New York: Cambridge University Press. 78 Adam Tamas Tuboly Ricketts, Th. (1996). Carnap: From Logical Syntax to Semantics. In Giere, R. N. & Richardson, A. W. (eds.) Origins of Logical Empiricism (pp. 213-250). University of Minnesota Press. Tarski, A. (1935/1956). The Concept of Truth in Formalized Languages. In Tarski, A. Logic, Semantics, Metamathematics. Papers from 1923 to 1938 (pp. 152-278). Oxford: Clarendon Press. Tarski, A. (1936/1956). On the Concept of Logical Consequence. In Logic, Semantics, Metamathematics. Papers from 1923 to 1938 (pp. 409-420). Oxford: Clarendon Press. Wagner, P. (2017). Carnapian and Tarskian Semantics. Synthese, 194 (1), 97-119. Woleński, J. (1999). Semantic Revolution: Rudolf Carnap, Kurt Gödel, Alfred Tarski. In Woleński, J. & Köhler, E. (eds.) Alfred Tarski and the Vienna Circle (pp. 1-15). Dordrecht: Kluwer Academic Publishers. Woleński, J. (2009). The Rise and Development of Logical Semantics in Poland. In Lapointe, S. & Woleński, J. & Marion, M. & Miskiewicz, W. (eds.) The Golden Age of Polish Philosophy: Kazimierz Twardowski’s Philosophical Legacy (pp. 43-59). Springer. Woleński, J. (2017). The Story of the German Translation of Tarski’s Book of Truth. In Bar-Am, N. & Gattei, S. (eds.) Encouraging Openness. Essays for Joseph Agassi on the Occasion of His 90th Birthday (pp. 8190). Dordrecht: Springer. Woleński, J. (forthcoming). On Controversy on Semantics at the Paris Congress 1935.