: A Grammar for

Schaar_Prelims.indd 1 9/23/2015 5:21:06 PM Poznań Studies in the Philosophy of the Sciences and the Humanities

Founding Editor Leszek Nowak (1943–2009)

Editor-in-Chief Katarzyna Paprzycka (University of Warsaw)

Editors Tomasz Bigaj (University of Warsaw) – Krzysztof Brzechczyn (Adam Mickiewicz University) – Jerzy Brzeziński (Adam Mickiewicz University) – Krzysztof Łastowski (Adam Mickiewicz University) – Joanna Odrowąż-Sypniewska (University of Warsaw) – Piotr Przybysz (Adam Mickiewicz University) – Mieszko Tałasiewicz (University of Warsaw) – Krzysztof Wójtowicz (University of Warsaw)

Advisory Committee Joseph Agassi (Tel-Aviv) – Wolfgang Balzer (München) – Mario Bunge (Montreal) – Robert S. Cohen (Boston) – Francesco Coniglione (Catania) – Dagfinn Føllesdal (Oslo, Stanford) – Jaakko Hintikka✝ (Boston) – Jacek J. Jadacki (Warszawa) – Andrzej Klawiter (Poznań) – Theo A.F. Kuipers (Groningen) – Witold Marciszewski (Warszawa) – Thomas Müller (Konstanz) – Ilkka Niiniluoto (Helsinki) – Jacek Paśniczek (Lublin) – David Pearce (Madrid) – Jan Such (Poznań) – Max Urchs (Wiesbaden) – Jan Woleński (Kraków) – Ryszard Wójcicki (Warszawa)

VOLUME 103

Polish Analytical Philosophy

Editor-in-Chief Jacek Juliusz Jadacki (University of Warsaw)

Editors Jacek Paśniczek (Maria Curie-Skłodowska University, Lublin) – Jan Woleński (Professor Emeritus, Jagiellonian University, Kraków) – Ryszard Wójcicki (Professor Emeritus, Polish Academy of Sciences)

The titles published in this series are listed at brill.com/paph

Schaar_Prelims.indd 2 9/23/2015 5:21:06 PM Kazimierz Twardowski: A Grammar for Philosophy

By

Maria van der Schaar

leiden | boston

Schaar_Prelims.indd 3 9/23/2015 5:21:06 PM The book was prepared within the project 11H 11 004280, Polish Philosophy of 19th and 20th Centuries (research module 11.1. of National Program for the Development of Humanities of the Polish Ministry of Science and Higher Education).

Poznań Studies is sponsored by the University of Warsaw.

Cover illustration: Kazimierz Twardowski among his pupils (Lwów, ca. 1910). Sitting from the left: Maria Fränklówna (2), Seweryn Stark (4), Kazimierz Twardowski (5), Karol Frenkel (6), (7), and Zofia Pasławska-Drexlerowa (8); standing from the left: Edmund Gromski (2), Mieczysław Tretter (5), Irena Jawicówka-Pannenkowa (9), Stanisław Leśniewski (11), Daniela Tennerówna-Gromska (14), and Tadeusz Kotarbiński (15).

ISSN 1389-6768 isbn 978-90-04-30402-4 (hardback) isbn 978-90-04-30403-1 (e-book)

Copyright 2016 by Koninklijke Brill nv, Leiden, The Netherlands. Koninklijke Brill nv incorporates the imprints Brill, Brill Hes & De Graaf, Brill Nijhoff, Brill Rodopi and Hotei Publishing. All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher. Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill nv provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, ma 01923, usa. Fees are subject to change.

This book is printed on acid-free paper.

Schaar_Prelims.indd 4 9/23/2015 5:21:06 PM POLISH ANALYTICAL PHILOSOPHY

Editors:

Jacek Juliusz Jadacki, editor-in-chief (University of Warsaw) Jacek Paśniczek (Maria Curie-Skłodowska University, Lublin) Jan Woleński (Jagiellonian University, Cracow) Ryszard Wójcicki (Professor Emeritus, Polish Academy of Sciences)

The volumes in the subseries present the heritage of the analytical movement in Polish philosophy in general, and the achievements, traditions and continuations of the Lvov-Warsaw School in particular.

Other volumes in the subseries: I: K. Twardowski, On Action, Products and Other Topics in Philosophy, ed. J.L. Brandl and J. Woleński (1999, vol. 67) II: T. Czeżowski, Knowledge, Science and Values: A Program for Scientific Philosophy, ed. L. Gumański (2000, vol. 68) III: Polish Philosophers of Science and Nature in the 20th Century, ed. W. Krajewski (2000, vol. 74) IV: J. Salamucha, Knowledge and Faith, ed. J. Jadacki and K. Święto- rzecka (2003, vol. 77) V: A. Wiegner, Observation, Hypothesis, Introspection, ed. I. Nowakowa (2005, vol. 87) VI: The Lvov-Warsaw School – the New Generation, ed. J. Jadacki and J. Paśniczek (2006, vol. 89) VII: A. Brożek, Theory of Questions. Erotetics through the Prism of Its Philosophical Background and Practical Applications (2011, vol. 99) VIII: K. Twardowski, On Prejudices, Judgments and Other Topics in Philosophy, ed. A. Brożek and J. Jadacki (2014, vol. 102)

Contents

1. Introduction. Twardowski as A Pupil and A Teacher ...... 9 2. Questions of Method. From Descriptive Psychology to Philosophical Grammar ...... 16 1 . Descriptive Psychology ...... 16 2 . A Philosophical Grammar ...... 24 3 . The Grammatical Distinction Between Internal and External ...... 32 4 . Modifying Terms ...... 35 3. Content and Object. From Psychology to Metaphysics ...... 50 1 . The Distinction between Content and Object ...... 50 2 . The Content of the Act ...... 55 3 . The Object of the Act ...... 59 From Psychology to Metaphysics ...... 59 Husserl’s Reaction to Twardowski’s Account of . . . . 61 Metaphysics and Mereology ...... 68 General Objects ...... 74 4 . Images and Concepts ...... 80 4...... Actions and Products ...... 84 1 . The Historical Background of Twardowski’s Theory of Judgement ...... 84 2 . Some Conceptual Distinctions ...... 91 3 . A Development in Twardowski’s Early Account of Judgement . . . . 97 4 . Actions and Products ...... 103 5 . Twardowski’s Critique of Russell’s Multiple Relation Theory of Judgement ...... 113 5. Knowing and Prejudice. An Educational Mission ...... 117 1 . Some Conceptual Distinctions ...... 117 2 . Brentano and Bolzano on Knowledge ...... 119 3 . Knowledge, Science and the Cognitive Act ...... 122 4 . Prejudice and the Critical Mind ...... 126 6. Truth and Time. Twardowski’s Impact on his Students ...... 129 1. The Correspondence Definition of Truth ...... 129 2 . The Absoluteness of Truth and the Logical Principles ...... 135 3 . Determinism and the Relativity of Truth to Time ...... 150 Truth and Time ...... 150 Jan Łukasiewicz ...... 152 Tadeusz Kotarbiński ...... 154 Leśniewski’s and Twardowski’s Reaction to Kotarbiński ...... 157 Conclusion ...... 160 Bibliography ...... 163 Name Index ...... 171 1.

INTRODUCTION. TWARDOWSKI AS A PUPIL AND A TEACHER

Kazimierz Twardowski (, 1866 – Lwów, 1938) is the founder of the Lvov‑Warsaw School with its strong tradition in and its scientific ap‑ proach to philosophy . Twardowski’s unique way of doing philosophy, his method, is of central importance for understanding his impact as a teacher . This method may be called “linguistic phenomenology,” to borrow a phrase from J L. . Austin, taking its starting point in ’s descriptive psychology and using a linguistic approach to philosophical problems . Because the term “phenomenology” is already in use for another tradition, I prefer to call Twardowski’s method a philosophical grammar, a term used by Leibniz for his universal language of thought 1. This focus on method is also the reason why Twardowski is still of im‑ portance for philosophy today . in the twentieth century can be characterised by its opposition to psychologism, on the one hand, and its opposition to metaphysics, on the other hand . This is changing now, as questions within the philosophy of mind and metaphysics are raised by philosophers standing in the analytic tradition . Do we need a new method in answering these questions? Analytic philosophers interested in these ques‑ tions may broaden their view by looking at a philosophical tradition that is “analytic” in a wider sense of the term . We can learn something from Polish logic and philosophy, as Twardowski and his pupils do not have a negative attitude towards analytic metaphysics, and certainly Twardowski himself was not opposed to a psychological approach to questions concerning judge‑ ment and intentionality . As I hope to show in the following chapters, we can improve our analytic methods by having some knowledge of Twardowski’s philosophical grammar . *

1 The term is also used by Anton Marty in his work on the philosophy of language (Marty 1908, pp . 67, 83) . Cf . (Mulligan 1990) . 10 Maria van der Schaar

Twardowski’s parents belonged to the Polish community in Vienna . Vienna at the time was the political and cultural centre of the Austrian‑Habsburg Empire . There was a tolerant politics, and people from all corners of the empire came to Vienna . Although the main language was German, a lot of people had Czech, Polish or Hungarian as their mother tongue . The Polish aristocracy played both a political and a cultural role in Vienna, and many of them went to the university . The Twardowski family, Ritter von Ogończyk, and the family of Kazimierz’ mother had played a role in the public life in . Twardowski’s father came to Vienna in 1851 to continue his study in law, and he later became a high official there . The family possessed no property from which it could have some extra income, and when Kazimierz decided to study philosophy rather than law, he did not want to burden the income of his parents . Twardowski went to the famous Theresianum gymnasium, which he fin‑ ished in 1885 . Obligatory for all gymnasia in Austria were lectures in logic and psychology . The official textbook was Robert Zimmermann’s Philo‑ sophische Propädeutik (1853) . Zimmermann was professor in Prague, and later got a chair in Vienna, where he supported Franz Brentano’s appoint‑ ment . As a young man he was ’s Herzensjunge, and the first edition of the book is strongly influenced by Bolzano’s ideas . The second edition that was used at the time Twardowski was a pupil at the gymnasi‑ um does not seem to follow Bolzano, and shows influences of J .F . Herbart, a psychologist and philosopher, who also defended a form of logical real‑ ism . In the textbook Bolzano is not mentioned, perhaps because Bolzano’s writings were forbidden by Rome and Vienna . Twardowski has always been an acute reader of Bolzano, and the teaching at the Theresianum probably had some influence . One of the teachers there was Alois Höfler, a student of Brentano, but to his regret Twardowski did not get Höfler as a teacher . As a student, Twardowski was tutor at the house of Count Dzieduszycki, who was politically active and liked to discuss philosophy . Twardowski met Kazimiera Kołodziejska at the house of the count, and married her in 1892. Twardowski had always enjoyed the company of women, and his choice was on a lady who was a good conversationalist, had a cheerful character, and was deeply religious (Brożek 2011, p. 191). The latter aspect of her character might have given some tension with the rather areligious Twar‑ dowski, but he never opposed any questions of belief by separating faith and knowledge . Twardowski loved music, and played the piano; there are also some compositions of his hand in the style of Hugo Wolf 2. Twardowski was the youngest of Brentano’s students, and was involved in discussions with other Brentano students, such as Alois Höfler . In private

2 Two songs are edited and recorded in (Jadacki 2005) . Introduction 11 he studied Aristotle together with Aquinas’ commentary, and the British phi‑ losophers . Twardowski got his PhD in 1892 on the basis of a thesis on idea and perception in Descartes, which he had finished in 1891 . He got a one year travel grant, and used it for a study in psychology in 1892 . He spent some time with Carl Stumpf in Munich and at Wilhelm Wundt’s psycholog‑ ical laboratory in Leipzig . As a topic for his Habilitationsthesis Brentano suggested Aristotle’s division of the sciences, but Twardowski preferred to write a more systematic treatment . Intrigued by a remark in Höfler’s Logik (1890), and probably stimulated by his reading of Bolzano’s Wissen‑ schaftslehre (1837), he decided to write on the distinction between content and object . As Brentano no longer occupied a chair, the official supervisor was Zimmermann . By submitting his thesis at the end of 1893, Twardowski earned the right to teach philosophy at the in 1894, and started in the winter of 1894/1895 with a course on logic, followed by a course on Hume’s Enquiry . As a Pole, Twardowski could not be appointed to a chair at a German University in Austria . Instead, he was asked to occupy a Chair in Lemberg (Lwów), giving his inaugural lecture on November 15, 1895, and obtaining a full professorship three years later . At the university, half of students were Poles, a third was Jewish, and a quarter was Ukrainian . Lwów was one of the cultural centres of the Austrian‑Hungarian Empire and the cultural cen‑ tre of Galicia . Four languages were spoken: German, Polish, Ukrainian and Yiddish . It is for this reason that the town is also known as Lemberg in the German language, as in Ukrainian, and as Lvov in Russian . They are different varieties of the same name .3 I will use the Polish name Lwów, but I will use Lvov in the name “Lvov‑Warsaw School,” as this is already an established usage . Twardowski was expected to give his lectures in Polish, and the main part of his later writings was written in Polish, although he took care that the more important papers were available in a German version . Having his roots in Vienna and exerting his influence in Poland, Twardowski was a typical example of a Central‑European philosopher . Other Brentano students had already become professor in Habsburg centres, such as Anton Marty, who got a chair in Prague around 1880 . For Twardowski, the older Brentano student Marty (1847‑1914) was of special importance, both for

3 This need not contradict Frege’s thesis that the names “Afla” and “Ateb” are different names of the same mountain . In Frege’s example the two names are not related to each other; they are independently given . The discovery that the two names refer to same mountain is a ge‑ ographical one – we have learned something about the world, that the Ateb is the Afla; the discovery that the four names refer to the same town is a linguistic one . In the example of the different names for Lwów the names are meant to be translations of each other . In a similar way, “Kasimir” is the German variant of Twardowski’s first name . 12 Maria van der Schaar his Brentanist philosophy of language, and as adviser for grounding a phil‑ osophical institute in Lwów. Cf. (Woleński 1990, p. 217). Twardowski was a great organiser, being rector of the university from 1914 until 1917 . Before the war, Twardowski started the Polish Philosophical Society in 1904 and the journal Ruch Filozoficzny in 1911 . The First World War started soon after he had become rector, and the Russians occupied Lwów, while Twardowski was not in town . He could not turn back to Lwów until July 1915 . Twardowski stayed mainly in Vienna during the war, helping those Polish students who had fled to Vienna 4. With some pride Twardowski speaks about the “Lvov School of Polish Philosophy” (Twardowski 1926, p . 28), as he was able to attract a circle of brilliant students . After the war philosophers and logicians of the first generation of Twardowski’s students obtained a chair in Warsaw, and we now speak of the Lvov‑Warsaw School . Twardowski was able to reach both students with a specialised interest in philosophy, and a wider public inside and outside the university . He was primarily a teacher; his lectures and seminars form the basis of most of the publications and unpublished material . He devoted ample time to his students; he had an intellectual and moral mission, and must have been very demanding . This mission became a characteristic of the Lvov‑Warsaw School. Cf. (Woleński 1989, p. 5). Twardowski accepted no dogma in his teaching; he had learned from Brentano a scientific attitude towards philosophical problems . Twardowski described his style as “from the bottom up,” starting from an exact description of ex‑ perience, with an open eye towards the sciences, especially psychology and linguistics. It is what Kotarbiński is to call ‘small philosophy’. He exerted influence by means of his philosophical style, not by pronouncing a phil‑ osophical doctrine . Twardowski’s aim is not to assume any metaphysical axioms . The distinction between content and object or the distinction be‑ tween presentation and judgement are meant to be philosophically neutral . His philosophical grammar is a universal grammar . Some of Twardowski’s students became well known phenomenologists, such as and , and some became psychol‑ ogists, such as Władysław Witwicki and Stefan Baley. Like Twardowski, Blaustein understood that philosophy is in need of psychological research (Miśkiewicz 2009, p. 187). Twardowski also attracted female students, for example, Izydora Dąmbska and Maria Kokoszyńska, who got important po‑ sitions as well. As Jan Woleński is a student of Dąmbska, the tradition still exists, and will be continued by a new generation with philosophers such as Tomasz Placek. Woleński’s work, such as his Logic and Philosophy in the Lvov‑Warsaw School (Woleński 1989), has played an important role in making

4 A fascinating chapter on this period can be found in the biography of Twardowski by Anna Brożek (2011). Introduction 13 the school known to the world . At least ten women of the school were appoint‑ ed at a university . Janina Hosiasson‑Lindenbaumowa (1899‑1942), a pupil of Kotarbiński and wife of Adolf Lindenbaum, seems to have been especially gifted, but was killed, like Lindenbaum, by the Nazis 5. Tadeusz Kotarbiński and Kazimierz Ajdukiewicz obtained their Ph D. . under Twardowski in, re‑ spectively, 1912 and 1913; they worked on philosophy of language, logic, and philosophy of science . Twardowski’s method of philosophical grammar plays an important role in Kotarbiński’s work.6 Ajdukiewicz’s categorical grammar is influenced by Husserl’s notion of Bedeutungskategorie, but the way it is used as a method of thorough thought originates in Twardowski 7. Ajdukiewicz, who was married to one of Twardowski’s daughters, is probably the student closest in style to Twardowski, being an excellent teacher with Klemens Szaniawski and Jerzy Pelc among his pupils . Twardowski’s most famous students, Jan Łukasiewicz and Stanisław Leśniewski, have specialised in mathematical logic. In 1902, Łukasie‑ wicz obtained his doctorate with Twardowski on induction and deduction . Leśniewski, born near Moscow, came to Lwów in 1910 after having studied abroad, mainly in Germany, and obtained his doctorate on existential prop‑ ositions under Twardowski in 1912 . Typical of mathematical logic in the Lvov‑Warsaw School, compared for example to the work of David Hilbert, is the rejection of a strict formalism: signs essentially represent something . Twardowski would be the first to remind them that symbols always symbolize something . Cf . (Twardowski 1921, p . 262) . In 1930, held his public lecture on the semantic definition of truth not only in Warsaw, but also in Lwów . Tarski, who got his education in Warsaw and wrote his doctorate

5 On these women philosophers in the Lvov‑Warsaw School, see (Woleński 1889) and (Pak‑ szys 1998) . The bibliographical references show that most of the women published in Polish, but that Hosiasson published in Mind, Journal of Symbolic Logic and Synthese, and Koko‑ szyńska in Erkenntnis . 6 “In Kotarbiński’s opinion, a particularly important task for the professional philosopher consists in pinpointing and eliminating philosophical questions which are wrongly posed and result in confusion . … It must be assumed that on that issue he shared Twardowski’s standpo‑ int, i e. ,. he assumed the possibility of identifying genuine and well posed problems directly from the semiotic properties of their formulations” (Woleński 1989, p. 63). Kotarbiński adds to this method of philosophical analysis a method of logical analysis based on Leśniewski’s formal system . 7 Cf. (Ajdukiewicz 1978, p. x). (Brożek 2011, p. 225) gives the following quote from Aj‑ dukiewicz, in her translation: Twardowski errichtete keine philosophische Schule, die sich an charakteristischen Aussagen erkennen ließe. Dafür war das eine andere Schule, die man nicht als philosophische bezeichnen kann, weil sie eine Schule im weiteren Sinn war: eine Schule für gediegenes Denken [thorough thought, MS]. In 1922, Leśniewski introduced the term “kategorie semantyczne” as a translation of Husserl’s term, using the notion for formal languages, as an alternative to Russell’s hierarchy of types . Ajdukiewicz’ notion can also be used in natural languages, as linguists have shown . Cf . (Haas 1989) . 14 Maria van der Schaar under Leśniewski, hoped to get the newly raised chair in mathematical logic in Lwów . Twardowski and Ajdukiewicz acted for Tarski, but the mathema‑ ticians promoted Leon Chwistek, who got the chair . Eventually, Tarski got a position in the United States, and through him the Lvov‑Warsaw School influenced the development of modern analytic philosophy . Chwistek nev‑ er studied under Twardowski, and one can discern an important difference between Chwistek’s view on semantics and that of Twardowski’s students . As Woleński writes: “Chwistek reduced semantics to a formal theory of expressions, logicians influenced by Twardowski looked upon referential relations as inherently involved in the use of language” (Woleński 2009, p. 52). Tarski’s views on semantics, influenced by Leśniewski’s opposition to formalism, made it possible that a semantics could be developed for for‑ mal systems (see chapter VI) . There is certainly a striking difference between the method used by Twardowski and the one used by his students . There is no sign that Twardows‑ ki read Frege, and he thus missed the important developments in logic . Twardowski’s early logic is mainly shaped by the logic of Brentano, which was a non‑Aristotelian logic, though still a term logic . There are manuscripts available of Twardowski’s courses on logic in Vienna and the earlier years in Lwów . Regrettably, a course on logic on the reformation of tradition‑ al logic concerning Bolzano, Brentano, Boole and Schröder disappeared during the Second World War. Cf. (Dąmbska 1978, p. 123). According to Izydora Dąmbska, the course was given several times, and was attended by Leśniewski and Kotarbiński.8 Frege was important for Leśniewski and Łukasiewicz. Łukasiewicz adapted Frege’s term “truth‑value” to his own purpose in his work on probability. Cf. (Łukasiewicz 1913; see chapter VI). In 1904, he read Russell’s Principles of Mathematics (cf . Simons 2014), including the chapter on the paradox of classes and the appendix on Frege . Leśniewski considered Frege’s Grundgesetze der Arithmetik the most valu‑ able contribution to deductive method since the time of the Greeks . Cf . (Woleński 1989, p. 318). Twardowski drew the attention of his students to the writings of Brentano, whose later ontology is most prominent in Kotarbiński’s work. Brentano’s thesis that the judgement is the bearer of truth and falsity has had an influ‑ ence on the thesis that meaningful sentences are the proper bearers of truth and falsity, which is defended by the majority of Twardowski’s students, and by Tarski, as well . Cf . (Rojszczak 2005) . At the same time, Twardowski drew the attention of his students to Bolzano’s Wissenschaftslehre . Bolzano’s influence played, for example, an important role in Łukasiewicz early work on probability. Cf. (Künne 2003a). Łukasiewicz and Leśniewski were espe‑

8 [O]n a le droit de supposer que c’est ici que la logique mathématique en Pologne a pris son essor (Dąmbska 1978, p. 123). Introduction 15 cially concerned with the status of the law of excluded middle and the law of contradiction in relation to the concept of truth, which topic may directly be traced back to Twardowski’s work on truth from 1900, and Bolzano’s defence of the absoluteness of truth . Because of these different kinds of influences, a different kind of analytic philosophy could emerge in Poland, combining the Brentanist tradition with Bolzano’s logical objectivism . Apart from his method, Twardowski’s defence of the absolute notion of truth was one of the forming influences on the Lvov‑Warsaw School, and it has played an important role in the discussion on truth, time, and the law of excluded middle around 1913 . Twardowski himself participated in the discussion, and it will therefore be the main topic of the last chapter (chapter VI) . The philosophical importance of Twardowski I hope to show by presenting the way he uses his method of philosophical grammar, dealt with in chapter II, in answering the question what the object of thought is (chapter III); what the bearer of truth and falsity is, and what meaning is (chapter IV); and what the difference is between knowledge and prejudice (chapter V) . Especially in the last mentioned chapter I hope to show how Twardowski’s educational ideal of a critical mind was a central element of the Lvov‑Warsaw School . 2.

QUESTIONS OF METHOD. FROM DESCRIPTIVE PSYCHOLOGY TO PHILOSOPHICAL GRAMMAR

1. Descriptive Psychology

Twardowski perceived his study from 1894 on the distinction between con‑ tent and object Zur Lehre vom Inhalt und Gegenstand der Vorstellungen to be a “psychological investigation,” using the phrase as sub‑title . In his early years in Lwów he is still willing to defend a form of methodological “psychologism” in the sense in which it is opposed to metaphysicism . Cf . (Twardowski 1897a) . How is it possible that someone who sees himself as a psychologist has had such an impact on a school that is famous for its mathematical logic? At first sight, one may think that this psychological attitude towards philosophical problems has exerted only a negative influ‑ ence . All Twardowski’s students advocated some form of anti‑psychologism in the sense that psychological questions are strictly separated from logi‑ cal questions . There is also another sense, though, in which Twardowski’s psychological way of dealing with philosophical problems is of importance to understand the later developments . It is especially Twardowski’s meth‑ od, in which linguistic as well as psychological distinctions are equally used, which must have convinced his students that philosophy can be done in a scientific manner . This psychological approach to philosophical problems Twardowski inherited from Brentano . The main work Brentano had published at the time is called Psychology from an Empirical Point of View . Psychology, according to Brentano, is not to make any metaphysical presuppositions; it is not a science of the soul, but the science of mental phenomena . Phenom‑ ena are not opposed to noumena, or things in themselves; they need not be phenomena of something out there . Phenomena are always appearing to someone . There may be mental or physical phenomena: examples of the former are judgings, acts of loving, and acts of presentation; examples of Questions of Method 17 the latter are the tone one hears, or the colour one sees . Physical phenomena are thus not events independent of someone who experiences them . They form the object of the natural sciences, just as the mental phenomena form the object of psychology . For Brentano, psychology consists of a part that describes and classifies the different kinds of mental phenomena, later to be called descriptive psy‑ chology, and a genetic psychology that formulates the laws concerning the causal relations between the mental phenomena . Descriptive psychology is an analysing description of our mental phenomena .1 It is prior in the order of explanation to genetic psychology, because the latter makes use of the concepts developed in the descriptive part, and because the exact laws of descriptive psychology show what necessary relations there are between the different phenomena . For example, acts of loving are necessarily dependent upon an act of presentation that provides the object of love . Empirical psychology is so‑called for its method . It is the method of inner perception that Brentano understands to be empirical in the first place . A mental phenomenon such as a sensing of red is the object of a special act of perception, called inner perception . This act of perception is contained in the act of sensation itself; it is a dependent aspect of the sensation . The inner perception of the sensation of red is an inner awareness: one has a presen‑ tation of the act of sensation, and judges that it exists . The inner perception is not the product of attention; it comes automatically with every mental phenomenon (Brentano 1874, p .41) . Every mental phenomenon is thus ac‑ companied by an act of inner perception as part of this phenomenon, and is thereby essentially complex . When we use the method of inner perception to obtain an exact law, a form of ideal intuition (ideale Anschauung, p . 1) is needed . In order to understand what classes of mental phenomena there are, and how they depend upon each other, we need to think of the varia‑ tions that might or might not be possible . Although Brentano assumes that inner perception is infallible, we can see in the moment of ideal intuition, and in the description one gives of the object of inner perception that the possibility of error is not excluded . Psychology needs to answer the question what its object, the mental phe‑ nomenon, is . According to Brentano, one has to start with clear examples (p . 11) . Hearing a tone, thinking of a triangle, judging that God exists and loving one’s neighbour are examples of such phenomena . A first characteristic of these phenomena is that each is either an act of presentation or has such an act as its foundation (p .112) . There is no judgement or fear without an act of presenting that provides the object for one’s judgement or fear . Second, all and only mental phenomena are the object of inner perception (p . 128) .

1 Ich verstehe darunter eine analysierende Beschreibung unserer Phänomene (Brentano 1982, p . 129) . 18 Maria van der Schaar

Primarily (p .137), though, mental phenomena are characterised by “the intentional (or mental) inexistence of an object” (p . 124) . They are all related to a content, directed to an object, as Brentano explains, not distin‑ guishing here between content and object . The content may be a physical phenomenon, such as the red I see, or the dagger that appears to me; it may be a concept or term, when I think of the characteristic (Merkmal) of being a man or of being mortal; or it may be a mental phenomenon, as in the case of inner perception . Brentano explicitly denies that the content as such is something real, for it does not enter into relations of cause and effect on its own . Only the mental phenomenon, of which the content forms a dependent part, is real . Physical phenomena such as the red I see have intentional, but no real existence (p . 129) . The appearing dagger has no real existence; it is but a dagger of the mind . Only the object of inner perception, the mental phenomenon itself, has both intentional and real existence (p . 137) . On one interpretation, Brentano acknowledges two forms of existence: intentional and real existence . One may also say, though, that intentional existence is not a form of existence at all, as Twardowski will put it, when he is speak‑ ing about the object of an act . This is problematic, though, if the physical phenomenon, the content of the act, is said to have intentional existence, as Brentano does . For there must be something in the act which makes a presentation of red different from a presentation of blue; this is precisely why Brentano distinguishes between act and content or object . The point Brentano seems to make is not that the content does not exist, but rather that there does not exist a dagger if we imagine a dagger . It is thus the dagger which has merely intentional existence . The same applies to a presentation of red . If one has a presentation of red, this does not imply that there exists something red inside or outside the mind . The red has merely intentional existence . For Brentano, the judgement that something red exists is always false . I prefer to put it this way: Brentano excludes all questions of exis‑ tence from the analysis of presentations and their intentionality; questions of existence arise only at the level of judgement . Brentano’s account of intentionality says nothing more than that in every presentation something is presented, in judgement something is judged, and in loving something is loved . We can say that these mental acts are characterised by their inten‑ tionality, as long as we understand that intentionality is not a mysterious relation between the mind and the world, for that would imply a metaphysical claim, and such claims do not have a place in descriptive psychology . On the contrary, these psychological analyses have metaphysical consequences, as we will see below . Finally, mental phenomena are characterised by the fact that they appear as a unity to the one who perceives them (p . 137) . Brentano speaks here of a multitude of perceived mental phenomena at a certain moment in time, Questions of Method 19 not about the unity of consciousness through time . The way different mental acts and their contents enter into relation with other acts and contents is described by Brentano in terms of different kinds of parts . For example, the red I perceive can be distinguished from the act of perceiving only by the mind, and is therefore a dependent part . Brentano develops a mereology for the mental in order to account of the unity and differences of our conscious‑ ness at a certain moment . For the early Brentano descriptive psychology aims at the last mental parts from which we may obtain any complex of mental appearances, thus giving a foundation to a characteristica univer‑ salis in Leibniz’ sense . Cf . (Brentano 1895, p .34) . Descriptive psychology thus aims at a universal language of thought that is foundational to logic . Brentano had an important reason for restricting his account of inten‑ tionality to the immanent object or content of the act . In the descriptive analysis of mental phenomena it is irrelevant whether the external object exists; in the analysis of the mental act one cannot determine anything about the existence of the external object . It is true that Brentano himself makes a distinction between content and object in his logic lectures, but the distinction does not play a prominent role in Brentano’s thought as far as the idea of intentionality is concerned . In these lectures, the distinction is made rather for semantical reasons 2. With the question what intentionality is, and what judgement is, we have reached questions that are of importance to philosophy . By asserting that these questions belong to psychology, Brentano is making a claim about the more theoretical parts of philosophy . The questions what existence is, what truth and goodness are, and what the judgemental content is, also belong to descriptive psychology . These questions belong to descriptive psychology, because these concepts have their origin in a reflection upon our acts of loving and acts of judgement (Brentano 1874, p . 89) . How do metaphysical questions and the more practical disciplines like logic or ethics relate to descriptive psychology? For Brentano, a metaphysi‑ cal question such as the question of the validity of the ontological argument for the existence of God cannot be answered without understanding what existence is, and what judgement is . These questions presuppose the anal‑ ysis of the basic concepts of descriptive psychology . Descriptive psychology gives a foundation to the universal language of logic free from the ambiguities and grammatical structure of natural lan‑ guage . Brentano does not depart from the traditional idea of a term logic; the logical change he proposes is essential, nonetheless . The central problem of Aristotelian syllogistics is, according to Brentano, that it takes grammat‑

2 The content of a presentation is the meaning (Bedeutung) of a name . The external, existing object of the presentation is named by the name (Brentano 1870, 13 018). . Cf . (Chrudzimski 2001, ch .10), (Rollinger 2009, p . 7), and (Betti 2013) . 20 Maria van der Schaar ical distinctions at face value . It is for this reason that it understands every judgement to consist of a subject and a predicate . According to Brentano, judgement is not to be explained in terms of predication, or as a combi‑ nation of two terms, a subject and predicate . By making a distinction be‑ tween the act of judgement and its content, Brentano is able to see that the distinguishing mark of judgement is not to be found in the combination of subject and predicate; it is to be found on the act side . The act of judgement is a special way in which we are directed to a content, and is in this sense distinct from the act of presentation . The content of judgement is simply the content of the underlying presentation, and every presented content has the right form to be judged . This means that one can affirm the content human being, one can affirm red, and one can affirm this tree . What can affirmation or judgement mean on this account? For Brentano, judging is not acknowledging the truth of a proposition or Gedanke, as Frege would say . Platonic entities, such as Bolzano’s Sätze an sich or Frege’s Gedanke, do not play a role in his logic . Because he does not acknowledge propositions, no account of propositional negation is given . Instead, Brentano has an act of denial besides the act of affirmation . Premises and conclusions in logic are not abstract propositions, but judge‑ ments, and the laws of contradiction and excluded middle are formulated in terms of judgements . Because Brentano does not acknowledge propositions, judgements are for him the proper bearers of truth and falsity . Judgement, according to Brentano, is either the acknowledgement or the rejection of the existence of a presented content . Affirming red is ac‑ knowledging the existence of something red . Existence is thus not a predi‑ cate, and can be obtained only by reflection upon the judgemental act . The word “is” has no meaning on its own (Brentano 1874, p . 57) . The + sign and the – sign are used by Brentano as signs that the content is affirmed, respectively denied, and is in this sense comparable to our assertion sign . (A B) + presents the judgement that some A are B, that an object exists that has both the A and B marks . The idea that existential judgements cannot be analysed in terms of sub‑ ject and predicate can already be found in Hume .3 Brentano refers, though, to Herbart and Trendelenburg, rather than to Hume, probably because he is opposed to Hume’s account of judgement or belief as vivid idea . Herbart acknowledges a special judgemental form for existential judgements . Exam‑ ples of existential judgements are “There are people” (Es gibt Menschen), but also “It rains” and “It lightens” (Herbart 1813, §63) . The topic was

3 “[I]t is far from being true, that in every judgment, which we form, we unite two different ideas; since in that proposition God is, or indeed any other, which regards existence, the idea of existence is no distinct idea … we can thus form a proposition, which contains only one idea” (Hume 1739, bk .1, pt . 2, p . 67, note) . Questions of Method 21 soon taken up by linguists (see the next section) . It may be argued that the idea goes back to Kant’s idea that existence is not a real predicate, and that existential judgements form a special class of judgements in which there is no synthesis of subject and predicate, but a mere positing . Cf . (Martin 2006, p . 52) . Because judgements are for Brentano the proper bearers of truth and falsity, the notion of truth, just like existence, can be obtained only by re‑ flection upon the act of judgement . By claiming that truth properly belongs to judgements alone, Brentano does not have a timeless bearer of truth, a topic that will become of central importance to Twardowski and his stu‑ dents . Brentano claims that the same judgement may change from true to false: whereas the judgement that it is raining is now true, it may be false in an hour, when it has stopped raining (Brentano 1889b, §55) . We thus see in what sense descriptive psychology is foundational to logic: it shows what judgement is; it gives the basis concepts to logic, just as it does for ethics and metaphysics, truth, goodness, existence and object; and it provides the basic elements for a universal language of thought . Brentano’s program forms the background of Twardowski’s early writings . Twardowski’s notions of phenomenon, intentionality and judgement, and his method are not identical with those of Brentano, as we will see below and in the coming chapters, but the latter form a starting point for his thinking . The book on the distinction between content and object is not very explicit on the method it uses . There is a paper, though, on the relation between psychology, physiology and philosophy from 1897 that is of interest for the question how Twardowski understands his philosophical method, in which psychology plays a central role . Psychology is not a part of physiology, Twardowski argues . Each has a different object: respectively, the mental and the physiological phenomena . The latter are in space and time, the former are not; the physiological phenomena are the object of outer perception alone, whereas mental phenomena are the object of inner perception (Twardowski 1897a, pp . 43, 44) . One has to make a distinction between observation and the inner perception of one’s own mental state . When one observes a tree, one focuses on the tree by means of an act of attention . There is no such observation of our mental acts the moment we have them, but there is a form of inner, direct acquaintance of these acts . We know a few things about our own mental activities, and this implies that they must be accessible to in‑ ner experience, as the outer senses do not reveal anything about our mental activities, at least not directly . It is precisely for this reason that empirical psychology cannot be reduced to a part of physiology, which uses only the method of observation and outer perception . Twardowski does not agree with Brentano that there is a strict distinction between empirical and experimental psychology, or between descriptive and 22 Maria van der Schaar genetic psychology, as Brentano calls it . The method of inner perception needs to be supplemented by results of, for example, Wundt’s experiments, which have the advantage of being repeatable and being accessible not only to the agent who has the perceptions (pp .49‑53) 4. Twardowski him‑ self lectured on psychophysics in 1897/1898, and on optical illusions in 1898/1899 . Cf . (Jadczak 1998, p . 46) . Twardowski did his first experiments in experimental psychology in a physics laboratory in 1896/97 . Apparently this was not approved by the authorities of the university, as Twardowski writes to Meinong, this happened silently (natürlich ganz im Stillen; letter to Meinong, July 25, 1897. I thank Wioletta Miśkiewicz for the reference). The method of inner perception needs to be supplemented by outer perception of our mental phenomena, and the physiological phenomena that express them . Although Twardowski praises Wundt’s method, it is clear that he is critical of the idea that psychological questions can be reduced to physio‑ logical questions to be answered in a laboratory . Twardowski’s position is fully within the Brentano framework, as the method of inner perception is prior to that of outer experience . If we study the mental life of animals and children we have to rely on the method of outer experience, but in order to give meaning to the results of these experiments we have to make use of inner perception . For, we take the physical phenomena to be expressions of mental phenomena, and for this we need to assume that there is an analogy between ourselves and children and animals (p .55, note) . This answer to behaviourism is still of value today, and Twardowski’s view on method here is not unlike the method that the biologist Frans de Waal has developed in his research on chimpanzees and bonobo’s . While separating psychology from physiology, Twardowski affirms the relation of psychology to philosophy . This thesis stands in sharp contrast to the claim of psychologists at the time that psychology should separate itself from philosophy . According to T . Ribot, psychology must separate itself from philosophy, in order to become a science, just as the other branches of science have separated themselves from philosophy 5. Ribot has a point, Twardowski answers, only if philosophy is understood as based on meta‑ physical axioms . Twardowski endorses the Brentanian thesis that neither psychology nor philosophy should be based on metaphysical presuppositions . Both psychology and philosophy should start with an exact description and classification of mental phenomena, of particular psychological facts . Meta‑

4 Twardowski not only spent some time in Wundt’s psychological laboratory in Berlin in 1892 as was noted above, but founded the official laboratory in Lwów in 1907 . Cf . (Smith 1994, p . 156), and (Jadczak 1998, p . 44), who describes Twardowski’s interest in Meinong’s psychological laboratory in Graz, as documented in a letter to Meinong (Twardowski 1897b) . 5 Twardowski refers to T . Ribot, “Enquête sur les idées générales,” Revue Philosophique 32 (1891) . Questions of Method 23 physics comes in only at the end of our philosophical inquiries . Philosophy should be based on experience, not on metaphysics . It is in this context that Twardowski contrasts metaphysicism, a philosophy based on metaphysical axioms, with ‘psychologism’, a philosophy based on psychology (p. 57): an exact description of psychological phenomena should be the beginning of philosophy . Twardowski thus adheres to a methodological psychologism . Philosophy should not start with such abstract questions as what the essence of truth is, or what the attributes of God are . Instead, it should investigate the difference between a judgement that is universally regarded as true and a judgement that is not so regarded, or it should start with the investigation how the concept of God arises in our mind . Furthermore, not only the meth‑ od, but also the object of philosophy is provided by psychology: without the inner perception of our judgements and emotions, logic, epistemology and ethics would be impossible (p . 59) . Philosophy is a group of sciences comprising psychology, logic, epistemology, ethics and metaphysics, where psychology comes first, and metaphysics comes last (p . 60) . Twardowski’s attitude to metaphysics is not negative: it properly studies the concepts of change, substance and causality . The point is rather that a philosopher should not study these concepts in abstraction from our experience . The positive attitude towards metaphysics in the Lvov‑Warsaw School as compared to the attitude of the can thus be traced back to Twardowski’s idea of a non‑dogmatic metaphysics . The early Twardowski endorses a form of psychologism in the sense that psychology is to provide the basic concepts of philosophy, that the method of inner perception plays a central role, and that the question what a concept means is not clearly separated from the question concerning the concept’s origin . Does Twardowski’s “psychologism” as presented above imply a psychol‑ ogism in the sense that philosophical questions are reduced to psychological questions? I do not think so (in the next chapter we will see, though, that Twardowski gives a psychologistic account of meaning) . Twardowski’s point is rather negative: we should not start with speculative abstractions, and psychological facts may be a good starting point for philosophical prob‑ lems . This is not all there is to Twardowski’s philosophical method, for as we will see in the next section the psychological method is supplemented by a rigorous method of linguistic analysis, which makes Twardowski’s method less vulnerable to psychologism than it looks at first sight . For Twardowski, psychology is a part of philosophy, but it certainly is not all there is to philosophy . An important part of philosophy consists of valua‑ tions concerning truth and falsity and good and evil, and logic and ethics are to provide rules for such valuations . To conclude, philosophy should not start with metaphysical axioms, but rather with psychological investigations of mental phenomena, free from 24 Maria van der Schaar metaphysical presuppositions . Eventually, we will ask metaphysical ques‑ tions, but these are to be answered “from the bottom up,” starting with the results of inner and outer experience (p . 63) . The method of exact descrip‑ tion, the opposition to speculative metaphysics, and the detailed analysis of particular philosophical problems must have inspired Twardowski’s students, and must have given them the idea that it is possible to do philosophy in a scientific way .

2. A Philosophical Grammar

Twardowski propagated an openness towards the sciences, whether it con‑ cerns psychology, linguistics, mathematics or the natural sciences, philosophy is to take into account the results of these sciences . Apart from psychology, linguistics is for Twardowski the most important science to take into con‑ sideration . In 1912 Twardowski publishes a paper called “On Actions and Products . Some Remarks from the Borderline of Psychology, Grammar and Logic,” and Twardowski’s philosophical method may be characterised as involving all three disciplines . Logical analysis has to combine psychological analysis with grammatical distinctions 6. In the nineteenth century, linguistics and logic are generally understood as being independent of each other . The Port Royal logic and its grammar developed within the Cartesian tradition had many reprints in the eigh‑ teenth century . It expressed the idea that a universal grammar is possible, and that universal grammar and logic may form an inspiration for each other . Wilhelm von Humboldt (1767‑1835) had an eye for the differences between languages, and the idea of a universal grammar became unpopular . For Humboldt, there is no universal language of thought; it is rather the uniqueness of each natural language which determines how people think . The linguist H . Steinthal (1823‑1899) claims that language should be studied independently of logic; linguistics is rather a member of the psychological sciences (Steinthal 1855, pp .141 -142) . On the side of the logicians, the independence of logic in relation to linguistics is also asserted by that time . According to Brentano, logical distinctions have to be clearly distinguished from grammatical distinctions . The grammatical thesis that every judgement contains a subject and a predicate does not have a logical counterpart, and many sentences have to be given a thorough analysis in order to reveal their logical structure . For example, universal positive sentences such as “Horses are mammals” are, from a logical point of view, to be analysed as negative existentials, for one denies that there exists a horse that is not

6 As Barry Smith (1989, p . 182) already noted, the title recalls that of Marty’s “Subjectless Sentences: On the Relation of Grammar to Logic and Psychology” (Marty 1884) . Questions of Method 25 a mammal, if one judges that “horses are animals ”. For Brentano, a gram‑ matical distinction is not a prima facie sign of a logical distinction . Logic is rather dependent on psychological analysis, that is, on an analysis of the way we think and judge . At the end of the nineteenth century, we see a rapprochement between linguists and logicians . In 1883, the slavist Franz Miklosich gives an interesting analysis of impersonalia, sentences such as “It rains,” which he calls “subjectless sentences ”. These sentences are called “subjectless” by him, because they contain only a predicate, not a subject . Miklosich refers to Brentano’s theory of judgement as an inspiration for his linguistic views .7 The same year Brentano publishes a review of the book, in which he endorses Miklosich’s thesis, although he is reluctant to speak about “subjectless sen‑ tences .” The term suggests that these sentences do have a predicate . Using the “predicate” terminology presupposes that one primarily thinks of sen‑ tences in terms of subject and predicate, and one should abandon this way of thinking (Brentano 1883, p . 190) . Although Brentano’s main idea is to treat logic separately from grammar, he considers Miklosich’s linguistic thesis to be a confirmation of his logical thesis that the subject‑predicate structure is not at all essential to judgements . The thesis that logical analysis should be distinguished from linguistic analysis is also endorsed by Brentano’s students . Thought is not to be defined as inner speech, Marty argues (Marty 1884, p .6), but this does not mean that linguistic analysis is irrelevant to logic . One may put the newly considered relation between logic and lin‑ guistics this way: logical analysis is to be identified with semantic analysis, and semantic analysis is to be distinguished from grammatical analysis . This tension between grammar, on the one hand, and logic and seman‑ tics, on the other hand, makes it necessary to study the grammatical surface structure of the sentence, in order to understand what the logico‑semantic deep structure of the intimated judgement is . According to Brentano’s logic, every judgement is an acknowledgement or a rejection of existence, as we have seen above . Existence thus belongs to the act of judgement, and is not a simple concept on the same level as the concept of red or circle . The con‑ cept of existence can only be obtained by reflection upon the act of judge‑ ment, for A exists means A can be correctly affirmed (Marty 1894, p . 201) . Marty elucidates this point by borrowing the term “innere Sprachform” from Humboldt . For Humboldt, the inner linguistic form is a property of a language as a whole, which expresses the world view (Weltanschauung) of the people using the language . Each language has its own inner linguistic form; there is no universal language of thought, as we have seen . According to Humboldt, the different names in Sanskrit for elephant, “the one with

7 Miklosich’s work Subjectlose Sätze (1883) is a second edition of Die Verba impersonalia im Slavischen, which appeared in a journal in 1865 . 26 Maria van der Schaar two teeth” and “the one who drinks twice,” express different concepts . It shows that the world view of those who spoke Sanskrit is different from the one expressed in modern European languages . In his critique of Humboldt, Marty gives a completely new meaning to the term innere Sprachform . For Marty, the different names for elephant all express the same concept; the only difference involved is a difference in inner linguistic form . Each of the names for elephant are associated with different presentations of the elephant thus appealing to our fantasy and aesthetic pleasure . These presentations are, according to Marty, nothing but auxiliary presentations, expressed by the inner linguistic form of the sentence, and do not form part of the concept of elephant (Marty 1893, p . 115) . The inner linguistic form is thus evoked by the surface grammar of the sentence, but does not correspond to anything in the deep structure or meaning of the phrase . Be‑ sides such figurative linguistic forms, there are also inner linguistic forms that are produced by linguistic constructions . Take the sentence “A tiger exists ”. On a surface level, “existence” is presented as a predicate (Marty 1894, pp . 201, 225) . The deep structure, though, of the sentence does not involve either a subject or a predicate, for there is only one idea involved, the concept of tiger . The thesis that every judgement consists of a subject and predicate is based on a confusion of the inner form of a sentence and the meaning of the sentence, as Marty says (Marty 1893, pp . 123‑124) . Mar‑ ty distinguishes between a logical and a psychological view . The logician abstracts from the differences in inner linguistic form, while psychology is needed to explain the inner linguistic form . For, one can explain the dif‑ ferent inner linguistic forms only by means of the psychological laws of association and imagination, and such an explanation has no relation to logic (Marty 1893, p . 125, note) . The form of psychology Marty is speaking about is genetic psychology, and in this sense Marty does not deny Brentano’s thesis that logic is to be based on descriptive psychology . We see in Marty a linguistic approach to logic at the expense of the relation between logic and psychology . Compared, though, to Russell’s early conception of logic as philosophical grammar (Russell 1903), in which a linguistic distinction is a prima facie sign of a logical distinction, Brentano and Marty clearly distinguish between logical and linguistic analysis . Their conception of the relation between logic and grammar is in this sense more in agreement with Frege’s Begriffsschrift . It is therefore of interest what Marty has to say on Frege’s Begriffs‑ schrift . First, Marty shares Frege’s thesis that the distinction between sub‑ ject and predicate is not a logical one (Marty 1884, p . 56) . Second, Marty endorses to a certain extent Frege’s distinction between judgeable content and act of judgement . There is an agreement insofar as Marty, like Bren‑ tano, distinguishes between the “matter” or content of the judgement and Questions of Method 27 the act of judgement . Both for Frege, and for Brentano and Marty, the act of judgement cannot be explained as a special case of an act of thought or presentation . The act of judgement is sui generis . Marty criticises, though, Frege’s characterisation in the Begriffsschrift of the judgeable content as a combination of presentations (Vorstellungsverbindung; p . 57) . Cf . (Frege 1879, §2) . Apparently, Frege still uses the traditional terminology, although he will not repeat this way of speaking about the judgeable content in later writings . The point Frege wants to make is that the content needs to form a unity in order to be able to be judged, and the horizontal is a sign that such a unity is present . For Marty, though, the basic judgemental form is not in need of such a sign, for a simple concept is already a judgeable con‑ tent . The concept of a house can form the matter of the judgement, and is therefore a judgeable content in these early writings of Marty; in fact, the Fregean distinction between judgeable and non‑judgeable contents is ab‑ sent .8 Finally, Marty criticises Frege’s thesis that negation belongs to the judgeable content, rather than to the act of judgement . If one claims that the matter of the judgement can be a simple concept, there is no place for propositional negation on the side of the judgeable content . Propositional negation can become part of the judeable content, only if one acknowledges a that‑structure as the fundamental structure of the judgeable content: white snow cannot be negated in this sense; one can only negate that snow is white . Brentano’s non‑propositional account of judgement can only account for negation by the acknowledgement of an act of denial 9. At first sight, for Twardowski, the relation between logic and language seems to be closer than in the writings of Brentano and Marty . According to Twardowski, although speaking and thinking are not completely parallel, there is an analogy between mental phenomena and the linguistic forms that express them (Twardowski 1894a, p . 10) . A name is the expression (Ausdruck) of an act of presentation; sentences that are used to make an assertion express acts of judgement; and sentences that are used to make requests, questions or orders are the expressions of feelings and acts of the will (p . 11) . Names are what the scholastics called “categorematic” terms; they are not yet the expression of a complete judgement, feeling or willing, but have an independent meaning, nonetheless, because the categorematic term is the expression of a complete act of presentation . Following Marty, Twardowski distinguishes three functions of a name or categorematic term: (1) it intimates, on its own, an act of presentation occurring in the speaker;

8 In his later writings, Marty does acknowledge a special judgeable content . 9 In his early writings, Brentano acknowledges term negation (non‑white) besides the act of denial, but later he reduces term negation to denial . 28 Maria van der Schaar

(2) it arouses a mental content, the meaning of the term, in the person addressed; (3) it designates an object; the term “the sun” is a name of the sun, Twardowski says with J .S . Mill (Twardowski 1894b, pp . 8‑10) . Is it possible that the same mental content can be in the speaker and in the hearer as well? If the content is mental, it seems to depend on a particular act of presentation . But, if it is possible that speaker and hearer have the same content, the content cannot be dependent upon the particular act of presentation of either the speaker or the hearer . In the next chapter I will deal with these questions, and the question whether Twardowski’s semantics is psychologistic needs to be raised there as well . Given the analogy be‑ tween speech and thought, the three functions of a name run parallel to the three‑fold distinction between act, content and object . For, the first function concerns the act, the second concerns the content, and the third concerns the object of the act . Basically, the reason for this analogy is captured by the first function of the name, to manifest a speaker’s act of presentation . Twardowski, though, fully acknowledged that grammar cannot be a straightforward guide to logic . Semantic ambiguities may arise because a sentence does not give full expression to the act of judgement . In 1902, in a paper on truth, Twardowski argues, for example, that an assertion by means of the utterance of the sentence “It rains” is an incomplete expression of the act of judgement . If we do not clearly distinguish between sentence and judgement, we may think that truth is relative, because the sentence seems to change its truth‑value, depending on whether it rains, or not (see chapter VI) . Not making the distinction between sentence and judgement confronts us with philosophical problems that disappear as soon as we notice the distinction . In his work on the distinction between content and object Twardowski argues that philosophers have often confused content and object, because the phrase “presented object” is ambiguous (see be‑ low) . Philosophical problems can be disentangled only if these semantic ambiguities are acknowledged . We have already seen in Brentano and Marty that the syntactic structures of language may be misleading, as far as logic is concerned . They are in need of a form of logical analysis not unlike Russell’s paradigmatic analysis of definite descriptions in 1905 . In 1894, Twardowski gives an analysis of the word “nothing” that can be understood as a form of logical analysis in this sense . In order to show that there are no objectless presentations, and that thinking of nothing does not form a refutation of this thesis, Twardowski argues that the word “nothing” is not a name, not a categorematic term, and therefore does not intimate an act of presentation . “Nothing” is rather a syncategorematic expression, that is, an expression that does not have an Questions of Method 29 independent meaning . On Twardowski’s account this means that the term does not intimate an act of presentation on its own, and that there is no content of an act of presentation that can function as meaning of the term, without invoking the meaning of other terms . An explanation of the meaning of “nothing” can only be given in a certain context . A full analysis of the term can only be given if we understand the term as part of a negative sentence (Twardowski 1894a, p . 23; see III 3). . As “nothing” is not a categorematic term, neither “Nothing exists” nor “Nothing does not exist” makes sense; nor does it make sense to say “There is nothing .” In this way we can refute the solipsist, Twardowski says (p . 35) . A grammatical distinction cannot be treated as prima facie evidence of a philosophical distinction 10. As far as semantic ambiguities are concerned, Twardowski uses Mar‑ ty’s notion of inner linguistic form to argue that we may express the same thought by means of different linguistic forms . According to Twardowski, the expression “land without mountains” has the same meaning as the term “flat land”, and the corresponding presentations have the same content (der Inhalt… ist ein und dasselbe; p . 99) . The difference between the two expres‑ sions consists merely in a difference in inner linguistic form (p .97) . Just as the etymological origin of a word (the Etymon) is no longer part of the meaning of a word, so the inner linguistic form is not a part of the meaning . The inner linguistic form is not a linguistic entity, but a presentation that is connected to the phrase “land without mountains,” mediating the intended meaning . The inner linguistic form is a presentation that functions as a sign of the intended content flat land . For this reason, Twardowski says that the presentation of a land without mountains is an indirect presentation (see below) . The phrase “land without mountains” arouses in the hearer a pre‑ sentation of a judgement in which mountains of this land are denied (p . 98) . This presented judgement is an auxiliary presentation, by means of which the hearer is able to determine the intended meaning of the name, which is identical to the meaning of the term “flat land .” Twardowski uses this point to argue against Bolzano’s denial of the thesis that there corresponds to every part of a presented content a part of the object of presentation . According to Bolzano, a presentation of a land without mountains presents an object that does not have mountains as its parts, whereas the content does have the content of mountain as a part (Bolzano 1837, §63) . On Twardowski’s analysis, mountain is not a part of the intended content or meaning of the name, and he is thus able to defend the traditional thesis that the material parts of the content correspond to parts of the object . Twardowski also makes a distinction between direct and indirect presenta‑ tions . Direct or intuitive (anschauliche; p .108) presentations, as Twardowski

10 Twardowski’s philosophical grammar is thus more critical towards grammatical distinctions than Russell is in what he calls a “philosophical grammar .” Cf . (Russell 1903, p . 6) . 30 Maria van der Schaar calls them, we may have, for example, of colours and numerals . In every presentation, an intuitive presentation is involved . Following in this respect, Twardowski says that indirect presentations present their object by means of a relation to some other object, which is apprehend‑ ed in a direct way . We have no direct presentation of Socrates’ father as Sophronikos, but we have an indirect presentation of him as the father of Socrates (Twardowski 1894a, p . 95) . Presentations containing a negation or incompatible characteristics are all indirect, just as presentations of an object by means of a sign . According to Twardowski, all indirect presen‑ tations are presented by means of an auxiliary presentation . I come back to Twardowski’s concept of auxiliary and indirect presentations in chapter III . 3, the last sub‑section, where I deal with Twardowski’s account of the presentation of general objects . Marty gives an interesting solution to the problem whether the parts of the content correspond to parts of the object . He rightly makes the point that Twardowski does not clearly distinguish between indirect presentations and presentations by means of auxiliary presentations, that is, by means of an inner linguistic form . According to Marty, the distinction between direct and indirect presentations, also called the dstinction between proper and improper presentations, concerns the content of the presentations, and thereby the meaning of the corresponding names . Proper presentations present the object by means of intuitive presentations of the object; the object is directly given in intuition . These proper presentations are essen‑ tial for a universal language of thought . For, in “the content of our proper thoughs of universals and relations lie the building blocks also for all our improper presenting,” (Marty 1892, p . 292) . Symbolic presentations are a special kind of improper or indirect presentations . In arithmetic nearly all our presentations are symbolic (Marty 1884, pp . 13-14), for we present the numbers by the representative function of numerals . But, not all our thought can be symbolic or blind in this way . The simple concepts need to present themselves as what they are . And, with respect to judgements, some judgements need to be insightful . Apparently, insightful judgements are based upon a direct or proper presentation of the content . Without such a foundation, arithmetic and all science would be nothing but empty words (leerer Wortkram) . According to Marty, the presentation of a mountain is a real part of the presentation of a land without mountains, and not merely an auxiliary presentation, as Twardowski thinks . The presentation of a land without mountains is an indirect presentation of a flat land . This means that, according to Marty, the terms “land without mountains” and “flat land” do not have the same meaning . The difference between the two terms is not merely a psychological difference given in terms of association and imagination, as would be the case if there were merely a difference in inner linguistic Questions of Method 31 form . Marty’s answer to the problem is that for proper presentations alone one may correctly assert that to every part of the content belongs a part of the object (Marty 1894, p . 219, note) . I think Marty’s critique is right . The difference between the concept of a land without mountains and a flat land does not merely concern our psychological associations . These concepts have a different syntactic structure; they are not isomorphic, as Carnap would say, and can therefore not be identical in content . Marty’s distinction between direct and indirect presentations goes back to Leibniz’ traditional distinction between intuitive and symbolic presentations . In his Reflections on Knowledge, Truth and Ideas, Leibniz distinguishes be‑ tween intuitive and symbolic cognition (the Latin term cognitio is a broad notion, equivalent to “apprehension” rather than to “knowledge”) . According to Leibniz, it is not always possible to intuit the whole nature of a thing, in which case we use signs instead, for brevity’s sake, knowing that “to give the explanation is in our power .” Of a distinct and primitive notion, that is, a notion that is not composite, we can only have intuitive cognition, but in algebra and arithmetic we are often in need of symbolic cognition . When the notion is highly complex, we are unable to think all the notions that are part of it, and use the words that represent these notions instead (Leibniz 1684, p . 5) . The distinction is of importance for Leibniz’ conception of a univer‑ sal language of thought, a truly philosophical language, in which the ideas are reduced to a kind of “alphabet of human thought” that would amount to a Cabala of mystical vocables, or to the lingua characteristica of the magi (Leibniz undated, p . 12) . When we use this language, we apprehend the ideas of the primitive signs intuitively, whereas the more complex ideas need to be apprehended in a symbolic way . By means of this philosophical grammar we are able to avoid the vague and uncertain meanings of words, and replace them by determinate characters (Leibniz undated, p . 14) . Both Bolzano and Brentano have used the idea of symbolic presentation in their writings (cf . (Brentano 1895, p . 34) and (Bolzano 1837, §90)), and the idea plays an important role in Husserl’s early writings . For Twardowski, symbolic presentations are a special kind of indirect or improper presentations (Twardowski 1894a, p .75) . If we think of the number 1000, we may think of it as being related to a sign, the numeral “1000,” of which we do have an intuitive presentation . As every presenta‑ tion, a symbolic presentation contains an intuitive presentation, not of the intended object, but of the sign that represents the object . Although a blind man cannot have a direct presentation of red, he can think of the colour red by means of thinking of a sound that functions as a sign of the colour . For Twardowski, the distinction between direct and indirect presentations is fundamental to his philosophical grammar . All presentations are founded on direct presentations . Without such a direct presentation, our presentations 32 Maria van der Schaar will be abstract and indeterminate, and, when expressed in words, will give rise to an ambiguity in language . Without symbolic presentations we would never be able to express ourselves clearly concerning objects that cannot be given intuitively, as in arithmetic . The distinction between direct and indirect presentations is thus part of a universal language in Leibniz’ sense, where primitive signs correspond to intuitive ideas . The idea of a universal language of thought is essential to Twardowski’s philosophical grammar . Linguistics also plays a more direct role in Twardowski’s writings . The semantic distinction between attributive and modifying terms, and the gram‑ matical distinction between the external and the internal object of a sentence are of importance in Twardowski’s explanation of philosophical concepts and distinctions . In order to get an idea of the importance of Twardowski’s philosophical grammar for philosophy today, I will deal with these distinc‑ tions in separate sections more from a systematic than from a historical point of view . How Twardowski uses these distinctions in order to solve certain philosophical problems will be dealt with in the next chapters .

3. The Grammatical Distinction Between Internal and External Object

The distinction between actions and products forms one of the important elements of Twardowski’s philosophy . As will become clear in chapter IV, the distinction is used by Twardowski to disambiguate philosophical notions like judgement and knowledge . The distinction is central to Twardowski’s argument against psychologism in the paper “Actions and Products; Some Remarks from the Borderline of Psychology, Grammar and Logic,” of which a first version appeared in German in 1911 . Typical of his method of philosophical grammar, Twardowski elucidates the distinction by means of a grammatical distinction . There is a grammatical difference between the two sentences: (1) Hans is reading a letter . (2) Hans is writing a letter . Grammarians use the term “direct object” to denote the grammatical func‑ tion of a phrase that is called the accusative in Latin and German . Although “a letter” is in both sentences the direct object, the accusative, there is a dif‑ ference between the two occurrences . In (1), “a letter” is the accusative of the external object of the sentence, that is, the indicated object or person makes up the scope of the activity indicated by the verb, or, as Twardowski puts it, the object (or person) exists before the activity that is directed to it . In (2), “a letter” is the resultant accusative, that is, the object indicated Questions of Method 33 is brought about by the activity of writing . Twardowski consulted Brug‑ mann’s Griechische Grammatik (cf . (Twardowski 1912a, p .171)), where the distinction is also explained in semantic terms 11. Twardowski makes a further distinction for case (2): if the indicated object is enduring, one speaks of the resultant accusative in a strict sense; if the indicated object is non‑enduring, one speaks of the internal object of the sentence . Because the grammatical distinction is given a semantical explanation, it can be used to elucidate the distinction between action (Funktion) and product (Gebilde), a terminology that Twardowski took from Carl Stumpf 12. If the distinction is applied to non‑linguistic entities, and thus made into a philosophical distinction, Twardowski uses the terms “result” or “inner object” (inneres Objekt), using them for both enduring and non‑enduring products . In both cases, the product is to be distinguished from the external object, which exists before the act is directed to the object . An interesting variant of the resultant accusative in the grammatical sense we may find in (3) Hans is singing a song for them . The direct object “a song” is called the cognate object .13 In such cases the noun and the verb are semantically and morphologically related, they are cognate . Here and in similar cases, like “to dream a dream”, the verb is standardly intransitive, but is made transitive for stylistic reasons . This kind of figure of speech is called figura etymologica .14 It is explained as the junction of two words into one expression, in which the second word repeats the stem of the first . Usually, the direct object repeats the stem of the verb . The rhetorical function of the figure is that of emphasis . German and Dutch, for example, use the figure in the saying selbst sein Grab gra‑ ben, zijn eigen graf graven . According to Twardowski, the linguistic distinction between a verb in‑ dicating an action and the direct object indicating the product of that action

11 Im Akkusativ des affizierten (oder ‘äussern’) Objektes steht … eine (konkrete) Sache oder Person, die in ganzer Ausdehnung das Ziel, den Wirkungsbereich einer Verbalhandlung (oder des Agens einer solchen) bildet … Kommt eine Sache … erst durch die Verbalhandlung zu‑ stande, spricht man vom effizierten Objekt oder Akkusativ des Ergebnisses (diese beiden werden auch als ‘Akkusativ des Inhalts’ zusammengefasst), z.B. ‘eine Münze schlagen’, d.h., sie schlagend herstellen . Eduard Schwyzer’s Griechische Grammatik auf der Grundlage von Karl Brugmanns Griechischer Grammatik, München: C .H . Beck, 1959, zweiter Band, p . 71 . 12 See chapter IV . The German term Funktion is better suited than the English term “action” to cover, besides physical actions, both active and passive psychical processes . 13 Compare A Comprehensive Grammar of the English Language, Longman, 1985 . 14 See (Twardowski 1912a, p . 160) . Cf . Historisches Wörterbuch der Rhetorik, G . Ueding (ed ). (Darmstadt: Wissenschaftliche Buchgesellschaft, 1996) . The term “etymon” is used in the meaning of stem . 34 Maria van der Schaar need not correspond to an ontological distinction . When I say that I dreamt a fascinating dream, I do not presuppose that there is a dream besides the function of dreaming . On the one hand, it seems clear that there is not some extra entity, say the dream, besides the activity of dreaming; on the other hand, we can say something of the dream, that we cannot say of the activity of dreaming, and vice versa . Although there may not be an ontological dis‑ tinction, there is always a semantical, or, as Twardowski says, a “logical” distinction corresponding to the linguistic distinction . In the sentence “they kept on making noise”, the topic is the activity of noise‑making, whereas in “the noise made at your party was too loud” the static aspect is at issue . There is thus a logical distinction between the “dynamic” aspect and the “phenomenal” or “static” aspect of the same activity, that is, there is a log‑ ical distinction between action and product, although there may not always be an ontological counterpart . Philosophically, the distinction between action and product becomes relevant, when it is used to disambiguate terms such as “presentation”, “cog‑ nition”, “assertion” and “judgement” (Twardowski 1912b, p . 111ff) .15 Each of these terms may denote an act, or the product that arises in the act . It is important to disambiguate these terms, because, according to Twardowski, the activity of presenting, conceiving or judging is part of a psychology of thought, whereas logic is concerned with judgements and concepts as products . Confounding the two leads to psychologism in logic . Whereas psychology is interested in the causal relations between acts of judgement, logical validity is a relation that holds between judgement products . The distinction is thus explicitly introduced to overcome the problem of psy‑ chologism without assuming that there are abstract propositions, such as Bolzano’s Sätze an sich (See chapter IV) .

15 The distinction between act and product is acknowledged in the scholastic tradition, espe‑ cially the Thomistic one; compare Jacques Maritain’s Petite Logique: La distinction essentielle entre l’acte de l’esprit (jugement) et l’ouvrage logique construit par lui (proposition ou énon‑ ciation) (1933, p . 5) . It is to be noted that the notion of proposition is not to be identified with the notion of proposition used in modern analytic philosophy: the former has a declarative structure, whereas the Russellian proposition has the structure of a that‑clause . In Aquinas the product is called the terminus (ad quem) . As in Twardowski, the terminus of an act of knowing, the “inner word” or inner object, is distinguished from the external object of knowledge . Questions of Method 35

4. Modifying Terms16

Introduction Twardowski wrote a small paper “On the logic of adjectives”, published in Polish in 1927, in which he proposes a classification for adjectives like “counterfeit,” “purported,” “true” and “actual .” Terms like “fake,” “pure,” “mere,” “true,” “real,” “actual,” “authentic,” “apparent,” “proper,” “genuine” have peculiar logical properties, and they are of interest to the linguist and logician alike . These terms behave in a systematic way, and are therefore not on a par with irony, or other forms of non‑literal meaning . There is something enigmatic about adjectives like “fake,” “false,” “mere,” “true” and “real ”. Some of these terms behave strangely in inferences: a fake Rembrandt is not a Rembrandt . Others seem to be redundant: there is no difference between gold and true gold . We do not know how to give a conceptual analysis of complex concepts like false friend or fake gun . None of our methods of analysis seems to be able to cope with terms like “real” and “fake ”. The method of conceptual analysis as we know it from Moore’s early writings is a whole‑part analysis and cannot help us here . Conceptual analysis may also be conceived along Fregean lines, as analysis of the content of a judgement . The content of the judgement John is married to Sue may be analysed into John and married to Sue, into Sue and John is married to, or into John and Sue, in that order, and being married to, depending on the kind of inference one needs to draw . On Frege’s account, we obtain three different concepts as a result of the three possibilities to analyse the judgeable content . We may also conceive of logical analysis along Russell’s analysis of definite descriptions . The linguistic sentence of which the definite description forms a part is reconstructed with a logical aim . In natural language, the definite description seems to have an independent meaning; on a deeper level, though, we can only give a logical analysis of the phrase when the linguistic context is taken into account . We are in need of a contextual explanation of some of our terms, and this general point is not irrelevant when we try to give a logical analysis of terms such as “fake” and “false .” None of these meth‑ ods of analysis, though, is able to capture the relation between the concepts gun and fake gun, between wine and pure wine, or between gold and real gold . The concept real is not simply a part of the concept real gold, nor is the concept friend part of the concept false friend in any straightforward sense of the term “part .”

16 This section is an improvement of my paper (Schaar 2014a), which was offered to Jan Woleński as part of a Festschrift . The paper is used with kind permission of Palgrave Mac‑ millan . 36 Maria van der Schaar

The topic of modifying terms, and of other non‑attributive terms, is generally neglected by logicians, although it has raised the interest of some linguists, as we will see below . Twardowski’s classification and explanation of non‑attributive terms is unique at the time, and is still of value today . In the rest of this section I take Twardowski’s explanations as a starting point in order to develop a full classification of non‑attributive terms . A new form of logical analysis is needed, one that is more sensitive to the linguistic con‑ text in which these adjectives are used . These terms are not only of impor‑ tance to the logician . (Twardowski 1894a, §4, pp .12ff) uses the distinction between attributive and modifying terms to explain that philosophers have used the phrase “presented object” in an ambiguous way, thus showing the need to make a distinction between the content and the object of an act of presentation . In the rest of the section I will not give a precise analysis of the three‑page paper that Twardowski has written on non‑attributive terms, as the paper is an example of clarity . Instead, I will give a systematic treat‑ ment of the topic, taking Twardowski’s classification in 1927 as a starting point, ending with a historical account of the distinction .

The Distinction between Attributive and Non‑Attributive Adjectives Attributive or determining adjectives like “blue” and “round” are standardly used to attribute a property to the object denoted by the noun phrase to which the adjective belongs . This means that the concept of being a blue ball is a complex consisting of the concepts being blue and being a ball . This is not the case for non-attributive terms, like “fake” and “apparent”, which stand for a function from concepts to concepts. If the term “fake” stands before the term “pistol” and we understand the meaning of these terms, we create from the concept pistol a new concept: fake pistol . Formally, what non‑attributive terms refer to may be understood as functions from classes to classes . Because many of these terms create intensional contexts, it is preferable to understand them as standing for functions from properties to properties . If we have grasped the non‑attributive meaning of “false,” we are able to understand new uses of the term, as in “false wall ”. This does not mean, though, that false walls, false Rembrandt’s and false friends have a property in common . For, a false Rembrandt is called false in a different sense than a false friend is . In contrast to attributive terms, non‑attributive terms are not able to split from the nouns to which they belong, or so it seems . The term “red” in “He is wearing a red jacket” may split from the term “jacket,” as in “He is wearing a jacket that is red ”. In contrast, we cannot say “This is gold that is real” or “This is knowing that is botched .” The linguist Barbara Par‑ tee has recently argued that splitting is possible for terms like “fake” and Questions of Method 37

“false,” at least in some languages, and that these terms are therefore to be classified as attributive . As I am going to argue that these terms are not at‑ tributive, I should have a different criterion to distinguish attributive from non‑attributive terms, at least, if it is true that these terms do sometimes split from their noun . At first sight, non‑attributive terms do not seem to form a homogenous class, for the criterion to distinguish them from attributive terms appears to be disjunctive . As a first suggestion, one might say that a term is non‑attributive if either of the following two inferences is invalid, where A is an adjective: (I) a is (an) A N . Therefore, a is (an) N . (II) a is (an) A N . Therefore, a is A* .17 The invalidity of (I) suffices to classify modifying terms like “fake” and “false” as non‑attributive, whereas the invalidity of (II) would make it possi‑ ble to classify terms like “real,” “mere” and “good” as non‑attributive . This criterion needs improvement, though, for the second inference is often not simply invalid; for quite some adjectives, it is rather that “a is A”, “this is pure”, does not make sense because these adjectives are non‑predicative . Can we improve our criterion? Is it possible to formulate a positive, simple cri‑ terion to distinguish attributive terms from non‑attributive terms? I think so . A term A is attributive precisely if inference (III) is valid . (III) a is (an) A N . Therefore, a is A* & a is (an) N . If a term does not have this logical property, it is non‑attributive . It may be the case that the inference is not valid, because the conclusion is not grammatical .

The Distinction between Relative Terms and Other Non‑attributive Terms Relative terms are a special case of non‑attributive terms . They are not used to attribute a property to an object; they rather qualify the property attributed . A term like “good”, as in “good pianist,” qualifies the property or concept denoted by the noun phrase, being a pianist, thus creating a new concept, being a good pianist . Compositional analysis cannot elucidate this complex concept, for a good shot is “good” in a different sense than a good pianist is “good ”. Someone may be a good pianist, but a bad leader of the

17 A and A* are not of the same syntactic and semantic category, for A* (“good”) does not, like A (“good” in “good pianist”), stand for a function from properties to properties, but simply for a property, at least, if a is A* is grammatical and makes sense . 38 Maria van der Schaar orchestra . Relative terms are qua terms: someone who is a good pianist is good as a (qua) pianist . Someone may be a good pianist and a man, but not a good man . For relative terms, the following inference is invalid, where R is a relative term, a is the subject term and N and N’ are nouns or noun phrases: a is an A N & a is an N’ . Therefore, a is an A N’ . And inference (II) is also invalid for relative terms: (II) a is an A N . Therefore, a is A* .18 It is for this reason that inference (III) is invalid for these terms, and that they are non‑attributive . There is an agreement between relative terms and terms like “false” and “fake .” Both inferences just mentioned are invalid for terms like “false” and “fake .” It is true that we sometimes say: “This is (a) fake,” but such a sen‑ tence is elliptical, precisely because it may be a fake pistol, but a real toy . There is also an important difference between relative terms and terms like “false” and “fake .” (1) For relative terms the following inference is valid, whereas it is in‑ valid for “false” and “fake”, as we will see below: (I) a is an A N . Therefore, a is an N .

For, a good pianist is a pianist, just as a big mouse is a mouse, but false gold is not gold . There is another, related difference (2) between the group consisting of terms like “fake,” “real” and “pure,” on the one hand, and terms like “good” and “great,” on the other hand . “He is a good pianist” gives an answer to the question: “To which group of pianists does he belong, the good or the bad ones?” . The good pianists, if there are any, form part of the extension of pianists . Just as small mice form part of the extension of mice, and we may thus speak of two kinds of mice, the small and the big ones . But, there are not two kinds of guns, fake guns and other guns, nor two kinds of gold, gold and real gold . This is precisely because terms like “fake” and “real” neither simply qualify the property denoted by the noun phrase, nor are they attributes of the object denoted by the subject term . The question is sometimes raised what kind of adjective the term “dead” is . Because there are not two kinds of man, the dead and the living, “dead” seems not to be

18 See footnote above . The relative term “big” is thus a different word than the absolute term “big .” Questions of Method 39 a relative term . As Aristotle remarks, it is false to call a dead man a man (De Interpretatione, chapter xi, especially 21a18‑33) . As Brentano saw, “dead” is a modifying term (see the last section below) . There is a third way (3) to distinguish the relative terms from other non‑attributive terms . It makes sense to say “He is good,” “She is skilful,” or “That is a big one” (speaking about a mouse) . This is caused by the fact that these terms qualify the subject qua so and so . Thus, I can say of him qua pianist that he is good . So, if the context is one in which young pianists are contesting for a price, one simply says “he is good (but she deserves the price) .” In contrast, it is difficult to give meaning to phrases like “This is real” or “This is pure,” and these phrases might even be called ungram‑ matical . And, although there may be a context in which one can say “This is fake,” one cannot say “It is fake” qua gun, or “false” qua gold .

Apart from the Relative Terms, There Are Two Kinds of Non‑Attributive, Non‑Modifying Terms: Restrictive and Restorative Terms Among the non‑modifying terms one may distinguish between the terms “mere” and “pure”, on the one hand, and terms like “real,” “proper” and “authentic”, on the other hand . The latter terms have a counterpart: “real” and “fake;” “true” and “false;” “proper” and “improper;” “authentic” and “inauthentic ”. When it is suggested that someone is a false friend, the an‑ swer might be “No, he is a real friend” . To understand the meaning of “real friend,” one needs to understand the meaning of “not being a real friend,” that is, of “being a false friend ”. It is the negative use that wears the trou‑ sers, as Austin says (Austin 1962, p . 70) . We first have to understand what it is for something not to be real gold, that is, to be a material that looks like gold but does not have the chemical properties that would make it into a piece of gold . Only against this background does the phrase “real gold” get a meaning . Saying that this is not a real Rembrandt, is not denying that it is a painting, nor is one generally claiming that it is a forgery . The question is rather whether it is painted by Rembrandt, or by one of his pupils . Only against the background of this question is the assertion that it is, or that it is not, a real Rembrandt given sense . The question might also have been whether it is a forgery or not, and then asserting that it is a real Rembrandt has a different meaning . Of course, the term “true” can also be used in an ironic sense, as T .S . Eliot uses the term “the True Church” in his poem “The Hippopotamus,” but even here the name “True Church” is given a meaning only in contrast to the way the name “hippopotamus” is used in the poem . The hippopotamus stands for a non‑institutionalised religion of flesh and blood, whereas the True Church is a church of dead dogma’s . In the man‑ uscript on the theory of judgement, Twardowski explains these terms as “adjectives which retain the meaning of nouns” (Twardowski 1902‑1903, 40 Maria van der Schaar p . 171) . Later, Twardowski says that terms like “real,” “true” and “actual” may restore the change in meaning that was caused by such terms as “fake,” “false” and “former” (Twardowski 1927, p .142) . It is for this reason that these terms come in pairs: “true” and “false;” “actual” and “former ”. Bor‑ rowing Twardowski’s terminology, we may call them restorative terms . The terms “mere” and “pure” do not have the same complexity as the restorative terms . They seem to contain a negation: mere belief is belief and nothing more, just as a mere child is a child and can therefore not be treated as an adult . “Pure wine” is wine that is not mixed with anything of lesser value; “pure wine” is wine and nothing less . The difference between “mere” and “pure” is not very strict, for the German translation of “mere nonsense” is “reiner Unsinn,” and we also can say “pure nonsense .” These terms do not function the way negation does, for “mere” and “pure” cannot be iterated, and it is not negated that a is an N . For, pure wine is certainly wine, just as mere nonsense is definitely nonsense . Terms like “pure” and “mere” restrict the application of the noun N: the object we talk about is nothing more or less than N . For this reason I call “mere” and “pure” re‑ strictive terms . It seems that a language is in need of only two restrictive terms: one corresponding to “mere”, and another corresponding to “pure” . Because it does not make sense to say that something is real, actual, pure, or mere, inference (III) is not valid in the case of restorative and restrictive terms, and these terms are thus non‑attributive . They differ from relative terms insofar as they are not qua terms: we do not say that something is pure qua water .

The Distinction between Modifying and Non‑Modifying Terms Within the group of non‑attributive terms, the modifying terms like “fake”, “botched,” “potential” and “false” may be distinguished from the non‑mod‑ ifying terms like “real,” “true,” “mere,” “pure” and the relative terms . For non‑modifying terms A, inference (I) is valid: a is (an) A N . Therefore, a is (an) N . It does not follow that a is A, for “a is A” does not make sense, and is not well‑formed, at least, in the case of non‑relative, non‑modifying, non‑at‑ tributive terms . “He is real,” “He is mere” and “She is true” do not make sense . But, it does follow that she is a lady, if she is a true lady . A is a modifying term, precisely when inference I is invalid: a is (an) A N . Therefore, a is (an) N . Questions of Method 41

From the premise that this is a fake gun it does not follow that this is a gun .

There Are Two Kinds of Modifying Terms: Privative and Modal Terms Modal terms like “potential” in “potential candidate” or “alleged” in “al‑ leged murderer” are clearly modifying in the sense that an alleged murderer need not be a murderer . From the fact that he is the alleged murderer, we can neither draw the conclusion that he is not a murderer, nor that he is . The intensional context makes it impossible to make either inference here, and such terms may be called modal modifying terms . The terms “putative,” “questionable” and “disputed” also belong to this class 19. Privative terms are distinguished from the modal terms by the validity of the following inference, where A is a privative term, (IV) a is (an) A N . Therefore, a is not (an) N . The term A may be an adjective, like “botched” or “false,” an adjectivally used noun, like “toy,” or a prefix, like “non” or “in ”. The validity of the inference implies that calling someone a false friend means that one is committed to assert that he is not a friend, although one has not asserted yet that he is not a friend .

There are Two Kinds of Privative Terms: Pure and Non‑Pure Terms There are quite some prefixes that are purely privative, think of “non,” “in,” and “un,” or the Greek alpha privative, and, perhaps, “ex .” These terms do not simply negate of a the quality denoted by N; they rather indicate that a has the positive contrary of that quality . We call something “nonsense,” not simply because it is lacking sense, but because it is expected to make sense, while not fulfilling this expectation . The purely privative terms should not be understood as negation .20 Like negation, they are not categormatic terms . But, whereas negation is a propositional connective, denoting a function from truth‑values to truth‑values, the purely privative term is non‑attributive, and therefore denotes a function from concepts to concepts .21 In the case of purely privative terms, the following inference is valid:

19 Benjamin Schnieder (2007, p . 533) has called these terms “partly modifying,” because some apparent idiots are idiots, while others are not . 20 Aristotle clearly saw the point in chapter x of the Categories: “Names and verbs that are indefinite (and thereby opposite), such as ‘not‑man’ and ‘not‑just’, might be thought to be negations without a name and a verb . But they are not . For a negation must always be true or false; but one who says ‘not‑man’ – without adding anything else – has no more said something true or false (indeed rather less so) than one who says ‘man’.” (Cat. 20a31‑36) . 21 Roberto Poli’s remark that “the modifying expression is of a syncategorematic type, i .e . that it behaves like a connective” (Poli 1993, p . 44), should be qualified: the modifying expres‑ 42 Maria van der Schaar

a is (an) A N . Therefore, a is not (an) N & a has the contrary of the quality denot‑ ed by N . The Aristotelian notion of privation is the ontological counterpart to the semantic notion of a purely privative term . When a man is called “blind”, he is asserted to lack a capacity he is to have according to his nature . The man is said to be deprived of a certain habitus, and the predicate can be constructed by means of an α‑privativum, a purely privative term. The non‑purely privative terms form an important and typical group of modifying terms, and may therefore be called modifying in the strict sense . If I call something a false N, I am committed to assert that it is not an N, and that it has the appearance of an N . A fake pistol has the appearance of a pistol, but misses the essential ingredient: it cannot be used to kill . In the case of modifying terms in the strict sense, the following infer‑ ence is valid:

a is (an) A N . Therefore, a is not (an) N & a has the appearance of (an) N . Is it essential that the appearance is deceptive, as in the case of mock‑Geor‑ gian windows, a false Vermeer, or a fake pistol? If one were to give an affirmative answer, one would exclude terms that are naturally classified as modifying in the strict sense, such as “toy”, and one would exclude the modifying uses of “mock” or “dummy” that do not have the deceptive as‑ pect, as in “mock examination,” or “dummy shells for the cannon ”. If we do not demand that the appearance is deceptive, the terms “presented”, “imagined,” “fictional,” “painted,” “stone” and “dead” can be considered as modifying terms in the strict sense, too, at least, in one of their meanings . In contrast to the term “fake,” these words can be used both as modifying and as attributive terms . The phrase “a painted landscape” is ambiguous, as Twardowski has shown (see chapter III) . It may be used to talk about a land‑ scape that is painted, in which case “painted” is used attributively . We use the term attributively to talk about a landscape with a mill and a river near Amsterdam that hasn’t changed since Rembrandt painted it . Or, it may be used to talk about a painting that has a landscape as subject . In the latter sense “painted” is used as a strictly modifying term: a painted landscape is not a landscape, just as a stone lion is not a lion, although a stone bridge is a bridge . The landscape is a painted one, that is, not a true landscape (sie ist keine wahrhafte Landschaft (Twardowski 1894a, p .13)) . Constitutive material modifiers, as used in “stone lion” and “wooden horse” are modi‑ sion does not behave like the logical connectives of propositional logic; it denotes a different function, and it is not iterable . Questions of Method 43 fying terms . “Rocking” in “rocking horse” is not a modifying term, because there are no similar constructions with “rocking .” The logician has to treat words like “rocking horse” and “teddy bear” as one word, for there are no other constructions in which “rocking” and “teddy” is used in this way . Ac‑ cording to Brentano and Twardowski, “former” is also a modifying term, whose restorative counterpart is “actual” (Twardowski 1927, p . 141) . I am not sure, though, whether it is a modal or a privative term . If it is true that a former senator is not a senator, “former” is privative . If the former senator is elected again, we no longer speak of a former senator .

Questions of Method Can we now broaden the method of logical analysis so that the relation be‑ tween concepts such as friend, false friend, and true friend becomes clear? One can start with a classification that can be given for different kinds of adjectives by considering their logical properties, that is, by considering the way they behave in inferences, as we have done above . Second, one may raise the question how a concept such as wooden horse is formed . Stan‑ dard concepts like red jacket are obtained by adding to the concept jacket the concept red . The concept wooden horse can be constructed by starting with the concept horse, of which a part is deleted, namely living being; the next step is to substitute for the deleted part a new concept, here made of wood . A concept like nonsense is formed by applying the function non to the concept sense, thereby obtaining a concept which is the opposite or the contrary of the concept sense . The concept false gold is constructed by applying the function false to the concept gold, thus obtaining the concept not gold, and adding the concept having the appearance of being gold, and, perhaps, pretending to be gold in order to obtain the more complex concept . The concept of true friend turns out to be a more complicated one: first, the function non is applied to he concept false friend; the concept true friend is thus identical with the concept non – false friend. Second, the concept false friend is obtained by applying the function false to the concept friend, ob‑ taining the concept not being a friend, while adding the concept appearing to be a friend, and, perhaps, pretending to be a friend. Although the concept false friend does not have that of friend as its part, it cannot be understood without understanding what a friend is . The concept friend has thus a logi‑ cal priority over the concept false friend, and the latter is logically prior to the concept of true friend .

Partee’s Proposal and Answer to Partee Recently, Barbara Partee has proposed the thesis that there are no privative or modifying terms, apart from the modal adjectives . Cf . (Partee 2010) . 44 Maria van der Schaar

Like standard attributive terms, and like relative terms, privative terms are subsective, she argues . In our terminology, inference I would be valid for privative terms, according to Partee . She gives three arguments for this original thesis . (1) Because one can sensibly say “Is that gun real or fake?”, the term “gun” must include both real and fake guns among its extension . (2) Unlike modal adjectives, and like attributive adjectives, modifying terms can split in some languages (This is a gun that is fake), and in some languages they can even precede the noun, as in Polish Fałszywe znaleźliśmy banknoty (False we‑found banknotes) . Finally, (3) this gives us a possibility to explain in what sense terms like “real” and “true” have meaning, and to show that they are not merely redundant or tautologous . For, real guns form now a sub‑class of all guns in the appropriate context . What can we answer to Partee? It is true that there is an important agree‑ ment between relative terms like “good” and “skilful,” and modifying terms, but, I think, for a different reason than Partee believes . Their agreement consists in the fact that in both cases the inference a is (an) A N . Therefore, a is A*, and a is (an) N . is invalid . Relative terms may therefore be classified as non‑attributive . Furthermore, there is an agreement between relative and privative terms: in the appropriate context, we can say “This is fake,” and “This one is false” (speaking about teeth), just as we can say “He is good” and “She is skilful ”. There are also three differences between terms like “good” and “skilful,” on the one hand, and modifying terms, on the other hand, as we have seen in the sub‑section on the distinction between relative terms and other non‑attributive terms . With respect to Partee’s first point (1), one can answer that the fact that “Is this Rembrandt real or false?” makes sense, does not imply that “false” is subsective, and thus an attributive term . It does not imply that the term “(being a) Rembrandt” is used here as a general term including both true and false Rembrandt’s . The general term is used in a deviant way for what people call a “Rembrandt”; in that sense we may perhaps say “That Rem‑ brandt is false ”. But, there are not two kinds of Rembrandt, the true and the false ones, just as there are not two kinds of gun, the real and the fake ones . In this sense there is a crucial difference between privative terms and subsective or attributive terms . Regarding argument (2), although we can sensibly say “Zeus is a god who is fictitious” instead of “Zeus is a fictitious god,” this does not mean that there are two kinds of god: fictitious and real ones . Possibility of split‑ Questions of Method 45 ting is apparently not a reliable indication of attributive terms . Instead, one should ask oneself whether it makes sense to say that the extension of the term can be divided in two sets by means of the adjective . For all non‑at‑ tribute adjectives A that are non‑relative, one can say that A N do not form a special kind of N . Just as a false Rembrandt is not a kind of Rembrandt, a mere child is not a special kind of child, a potential terrorist is not a kind of terrorist, and real gold is not a kind of gold . Regarding (3), I agree with Partee that “real” in “real Rembrandt” is not redundant, but not for the reason that real N may form a subclass of N in the appropriate context . The phrase is given a meaning dependent upon a counter suggestion, in this case, that the painting is not a real Rembrandt, that it is a false one . “Real” has thus a restorative function . My conclusion is that for privative terms inference (IV) is valid, which implies that they are not subsective, that is, that they are not attributive: (IV) a is (an) A N . Therefore, a is not (an) N . One may put the different kinds of non‑attributive terms in a schema:

A History of Modifying Terms Earlier than Twardowski, Brentano’s older student Anton Marty has written not only about modifying terms, but also about restorative terms . According to (Marty 1884, pp . 52-53), terms like “real,” “actual,” and “true” are able to restore the original meaning of the head noun, when the meaning has been modified by terms such as “painted” in “painted lion,” or by terms like “past” and “future ”. 22 Marty points to the fact that we sometimes can delete the modifying term, but still use the noun in its modified meaning, for

22 Benjamin Schnieder rightly points out that Marty’s explanation of modifying terms is prob‑ lematic . Modifying terms do not literally modify the meaning of the qualified general term, “such that ‘fish’ should have different meanings in ‘depicted fish’ and ‘blue fish’” (Schnieder 2007, p . 532) . The same critique applies to Twardowski, who claims: Modificierend ist eine Bestimmung dann, wenn sie die ursprüngliche Bedeutung des Namens, bei welchem sie steht, 46 Maria van der Schaar example, when we refer to a painting of a landscape by calling it “a land‑ scape ”. Philosophical interest in these modifying terms in the nineteenth century originates with Brentano and Bolzano 23. In the first part of the Wissenschaftslehre, there are at least four places where Bolzano makes philosophical use of modifying terms (Bolzano 1837, §§19, 23, 29 and 59), although Bolzano does not call them that way . In sec‑ tion 19, an interesting quote from the fifteenth century Florentine Savonarola is given: “[Just as] dead man has the form and the likeness of a man, but is not a man” ([Sicut] homo mortuus habet figuram et similitudinem hominis, non tamen est homo) . The explanation Savonarola gives is of interest, for it is true that if A is a modifying term and N the noun phrase, we can draw the conclusion that the object is not an N, and also that the object has the appearance or form of an N . Bolzano’s explanation of phrases in which modifying adjectives occur may be called syntactic . If the adjective is not a modifying term, as in “golden candle,” the head noun of the phrase indi‑ cates the head‑presentation of the complex presentation that is meant by the phrase . This means that a golden candle is a kind of candle . In cases where the adjective is a modifying term, it is not the head noun that indicates the head presentation; for example, in “painted fish” the head noun “fish” does not indicate the head presentation, for a painted fish is not a kind of fish . It is rather a kind of painting, which means that “painted” indicates the head presentation (Bolzano 1837, I, §59) . This distinction can be used to clarify misleading terminology in philosophy . According to Bolzano, a possible thought, cogitatio possibilis, is not a kind of thought, it is rather a kind of possibility (§23) . Just as an apparent truth is not a kind of truth, so a rela‑ tive truth is not a special kind of truth, either (§29) . There are, according to Bolzano, also phrases where neither of the terms can be understood as indi‑ cating the head presentation . In such cases we have an improper nomination (uneigentliche Benennung; §182) . Bolzano gives the example of “a formal truth ”. A formal truth is, for Bolzano, neither a kind of truth nor a kind of form . The point is that there are not two kinds of truth, the formal truths and the real truths . What philosophers call a formal truth is a proposition that does not contain a contradiction, but this need not be a truth at all . These philosophers take the term “truth” too broad in extension (§29, pp . 138-139) . Brentano mentions the modifying terms in his logic lectures from the early seventies (Brentano 1870, EL80‑13 063[4]). . When dealing with dif‑ ferent kinds of ambiguities, he says that an adjective in the composition of an adjective and a substantive is normally determining, but may also vollständig ändert (Twardowski 1894a, p . 13) . In the main text below (III 1). I have given an interpretation of Twardowski that would not be vulnerable to Schnieder’s critique . 23 Markus Stepanians and Arianna Betti have already pointed to the Bolzano side of the topic . Cf . (Stepanians 1998, p . 23) and (Betti 2006, p . 59) . Questions of Method 47 be modifying, as in falsches Geld and gedachter Taler . This is not a true form of ambiguity, Brentano says, but rather an ambiguous form (äquiv‑ oker Form) . The passage from the manuscript consists merely of notes and does not contain full sentences, so one has to give an interpretation . I assume that Brentano is pointing to a syntactic ambiguity: the combina‑ tion of adjective and substantive is given a different meaning depending on the question whether the adjective is a modifying or a non‑modifying term . So, there is a syntactic ambiguity in the phrase “painted landscape,” because the adjective may be understood to be determining, or the noun may be understood as qualifying what is referred to by the adjective: the thing that is painted is a landscape . Understanding modifying adjectives in the context of syntactic ambiguity shows a certain similarity with Bolza‑ no’s analysis, although Brentano does not use Bolzano’s concept of head presentation . Brentano’s first publication on the topic can be found in his Psychologie vom empirischen Standpunkt (1874, pp . 60‑63) . In a long note Brentano quotes from the “brouillon” of a letter he has sent to J S. . Mill, discussing Mill’s thesis that the copula does not necessarily include exis‑ tence . The correspondence dates from shortly before the publication of the Psychology, as Mill’s answer is from February 1873 . According to Mill, the sentence A centaur is a fiction of the poets does not include an affir‑ mation of existence, for it cannot be implied that a centaur, being a fiction of the poets, exists (Mill 1843, Bk . I, Ch . IV, §1) . Brentano answers that the sentence does contain an affirmation of existence, although not of the existence of a centaur, for the sentence has the same meaning as “There is a poetic fiction, in which the upper part of a human body and the trunk of a horse are unified in thought into a living being ”. So, the main phrase is: “there is a poetic fiction .” Again, this is not unlike Bolzano’s analysis, who equally would say that the sentence “There is a fictitious centaur” contains an assertion about a certain kind of fiction . There is no affinity, though, in terminology between Brentano and Bolzano . The logical role of a term like “fictitious centaur” can be explained, when we understand that not all adjectives add something to the concept expressed by the term they belong to . Equally, the term “dead” in the sentence “a man is dead” is a modifying term, for a dead man is not a man at all (Brentano 1874, II, p . 62, note) . In our terminology, inference (I) is invalid, and inference (IV) is valid for these terms . Brentano understands that terms in predicative position may also function as modifying terms . He uses here Savonarola’s example, but there is no reason to presume that he got the example from Bolzano, for we also find it in Aristotle’s De Interpretatione (21a18), and it was well known in Scholastic literature . Linguistic distinctions should not be confused with logical distinctions, Brentano ends the long note . That is, the sentence “A man is dead” should not be treated as having categorical form; the logical form 48 Maria van der Schaar of such a sentence is “There is a dead man ”. Brentano’s early analysis of modifying terms is thus given in logical terms with regard to the structure of the sentences in which these terms occur, and thus has some agreement with the account I have given above . In 1914, Brentano comes back to the modifying terms . Adding the word “past,” “painted,” “thought,” or “ap‑ parent” to a noun, such as “king,” does not enrich, but rather modifies the meaning, deleting the original meaning, and substituting another meaning instead . A past king as such is no more a king than a beggar is (Brentano 1914, p . 46) . The 1914 explanation is clearly different from the one given by Bolzano, as it is now given purely in semantic terms . Although the se‑ mantic explanation is published only in 1914, it is reasonable to assume that Brentano had used it in his lectures, and that Marty and Twardowski knew this explanation, perhaps because they have attended these lectures, or have discussed the topic with Brentano in private . When Twardowski makes use of the distinction between attributive and modifying terms to explain the ambiguity of the term “presented object,” he refers in this context only to Brentano, and the explanation is given in semantic terms . According to Twardowski, an adjunctive clause is modify‑ ing, when it changes the original meaning of the name, to which the clause belongs, and substitutes a new meaning (Twardowski 1894a, §4, p . 13; for a critique of this position, see footnote 19 above) . Marty introduces the idea of modifying and restorative terms in the context of Mill’s example and the Brentanian thesis that all judgements are existential in form . Like the later Brentano, he gives a semantic explanation of modifying terms: they “modify” the meaning of the noun phrase (Marty 1884, p . 50) . The source for Marty and Twardowski thus seems to be Brentano rather than Bolzano . It is most likely that Bolzano and Brentano were familiar with this special function of terms like “dead” and “in thought” because Aristotle writes about it at several places, and because it can be found, for example, in the writings of Ockham and Cajetan . Aristotle’s Categories opens with the ambiguity or homonymy of certain names . The example he gives relates to the fact that in Greek the same name is used for a painting and for a man . Because a different definition is given for the two kinds of objects, Aristotle says, this is an example of homonymy . The example is not related to modifying terms, for there is a real ambiguity in the Greek language . Aristotle comes back, though, to the topic of homonymy in a more interesting context in De Anima (412b17ff): “The eye is matter for sight, and if this fails it is no longer an eye, except homonymously, just like an eye in stone or a painted eye” (Hamlyn 1993, p . 10) . If the word “eye” refers to an eye that can no longer be used to see, we cannot use the word “eye” anymore in the same sense, and the same holds in case the term is followed or preceded by modifying phrases like “in stone” and “painted ”. The most important passage can Questions of Method 49 be found, though, in De Interpretatione (21a18ff, the end of chapter 11) . Paraphrasing the first part of the passage, we may say that if something is a white man, that he is white, and that he is a man . This does not hold for all terms, though . When the adjective contains something that is opposite to what is contained in the noun, a contradiction follows, and the conclusion is always false . For example, when we call a dead man a man, this is false, because the adjective dead contains the meaning of not living, whereas the noun man contains the meaning of living . In terms of the validity scheme and our concept of modifying terms, Aristotle would say that “dead” is a modifying term, for if it is a dead man, it follows that it is not a man . But there are also cases, Aristotle adds, in which the conclusion may be either true or false, that is, in which case it is invalid that if it is an A N, it is an N, as in the example of Homer is a poet, in which a poet qualifies the verb is, which means that the conclusion that Homer exists cannot be drawn . Aris‑ totle ends the passage with an interesting example . If something is thought, or exists in thought, it does not follow that it is, or that it exists 24. In the terminology introduced above, “(in) thought” behaves like a non‑attributive term . The meaning of “existence” is qualified by the modifying term “in thought .” There is no way, though, in standard formal logic to capture this behaviour of “existence” in terms of the invalidity of inference schema I, for “existence” does not belong to the category of nouns N . So, we have to assume either that there is a meaning of “existence” in which it is a first‑or‑ der predicate, or we have to assume that something else is going on here . We will see in the next chapter that a modifying sense of “existence” plays a role in Twardowski’s explanation of intentionality . Although Twardowski had learnt from Brentano that existence is not a logical predicate, he seems to imply that existence is in some cases a first‑order predicate .

24 Weidemann translates the example as follows: Von etwas Nichtseiendem kann aber nicht deshalb, weil es ein vermeintliches (Seiendes) ist, wahrheitsgemäss ausgesagt werden, es sei ein Seiendes. Denn die (blosse) Meinung, (es sei), hat man von ihm ja nicht etwa deshalb, weil es (seiend) wäre, sondern gerade deshalb, weil es nicht(seiend) ist (Weidemann 1994, pp . 24-25) . 3.

CONTENT AND OBJECT. FROM PSYCHOLOGY TO METAPHYSICS

1. The Distinction between Content and Object

Twardowski gives in the first section of his treatise on the content and object of presentations a quote from Höfler’s Logic .1 Höfler, in reaction to the Kantian thesis that we can never know anything about the things in themselves, argues that the word “object” is ambiguous . On the one hand, it may indicate the object to which the act is directed, and which exists in itself (an sich), or, as he also puts it, which is assumed to be independent of the act; on the other hand, it may indicate the object existing “in” us, the more or less approximating “picture” of the real object . This “quasi‑picture” or sign is the content of the act . This content is also called the “immanent or intentional object” of the act (Höfler 1890, §6, p . 7) . The distinction as it is meant in logic and psychology should be independent of metaphysical claims about being as such (das an sich Seiende; idem) . Although Höfler does not say more about the distinction, all of Höfler’s theses are endorsed by Twardowski . Both for Höfler and Twardowski, the distinction has to be made by every philosopher, although each philosopher may give a different account of the ontological status of the act’s content and object . The distinc‑ tion is philosophically neutral, and is, as a psychological distinction, prior to any philosophical theory . Although Twardowski starts with a psychological account of the distinction, he is also interested in metaphysical questions, and in the philosophical arguments for it . The question what the ontolog‑ ical status of the object of our acts is, is not neglected by Twardowski . At first sight it may seem that contradictory claims are made concerning the ontological status of the object . On the one hand, the object is said to be

1 It is especially the first edition of the Logik that is relevant here . The book is written with the co‑operation of Meinong . Content and Object 51 a phenomenon, a counterpart to the act of thinking; on the other hand, ob‑ jects are claimed to be independent of the act of thought . These two claims are made, though, from two different points of view . If one separates the psychological from the metaphysical point of view, it is possible to combine these seemingly contradictory theses, as we will see below . Twardowski introduces the distinction between act, content and object in analogy with the three functions of a “name,” as we have seen in II 2. . According to Twardowski, philosophers have often neglected the distinction between content and object of presentations, partly because one may use the phrase “presented object,” or “something presented,” for both the object and the content of the act . Twardowski disambiguates the phrase by means of the linguistic distinction between attributive and modifying adjectives introduced in the former chapter . According to Twardowski: [A] determination is called attributive or determining if it completes, enlarges … the meaning of the expression to which it is attached . A determination is modifying if it completely changes the original meaning of the name to which it is attached (Twardowski 1894b, p . 11) . This does not mean that, according to Twardowski, the meaning of the expression type changes . It is rather that he is speaking of the meaning of a particular occurrence of a term, that is, the meaning of the term in this particular context . The meaning of the expression in the context of a mod‑ ifying term is different from the meaning it has in more standard contexts . As Twardowski follows Brentano’s existential account of judgement, one may give the following analysis of the distinction: If A is an attributive term, and B is a noun, the inference from (A B) + to (B) + is valid . Given that “German” is an attributive term, one can infer that there is a pistol from the premise that there is a German pistol . A German pistol is a certain kind of pistol . If A is a modifying term, the inference from (A B) + to (B) + is invalid . “[I]f one says “dead man,” one uses a modifying adjective, since a dead man is not a man” (p . 11) . Some words can be used attributively in one context, and modifying in another, as the word “false” in, respectively, “false judgement” and “false gold .” A false judgement is a kind of judgement, but false gold is not a kind of gold . Adjectives like “presented” and “painted” can be used in both ways . One may speak of a “painted landscape” in two senses: one may speak about a landscape near Amsterdam that is painted . Here “painted” is used attribu‑ tively .2 One may also speak of a “painted landscape” when speaking about

2 Der Zusatz ‘gemalt’, in diesem Sinne zum Worte ‘Landschaft’ hinzugefügt, modificiert die Bedeutung des Wortes ‘Landschaft’ keineswegs; er ist ein wahrhaft determinierender Zusatz, welcher angibt, dass die Landschaft in einer bestimmten Relation zu einem Bilde steht, in einer Relation, welche die Landschaft ebensowenig aufhören macht, eine Landschaft zu sein, als… (Twardowski 1894a, p . 14) . 52 Maria van der Schaar a painting, that is, “painted” is used as modifying term .3 From the judgement (painted landscape) +, one is not allowed to draw the conclusion (landscape) +, when “painted” is used as a modifying term . The example is not chosen by accident, as one may characterize the activity of presenting as a kind of mental picturing (eine Art geistigen Abbildens) . Cf . (Twardowski 1894a, p . 14) . The object is presented by means of a content, which functions as a kind of picture of the object . Just as the picture is for the painter a means (ein Mittel; p . 17) to present the landscape, so the presented content is a kind of picture by means of which the object is presented (p . 18) . In the analogy, we substitute “presented” for “painted;” “presented” may be used both attributively and modifyingly; we substitute “object” for “land‑ scape;” and we substitute “content” for “picture ”. The phrase “presented object” may refer the object of presentation, in which case “presented” is used attributively, or it may be used to refer to the content of presentation . In the latter case, “presented” is used as a modifying term: the presented object is thus not a certain kind of object . A point to be remembered when we will compare Twardowski’s account with the traditional representational account of ideas . The presented object is a content, just as the painted landscape is a picture . The presented object is thus identical with the presented content, where “presented” in the first occurrence is used in a modifying sense, whereas it is used in its attributive sense in “presented content” (pp . 14-15) . Twardowski starts the section with another, related analogy between painting and presenting . As a preparation for the elucidation of the distinc‑ tion between content and object, he argues that we can speak about a picture and about a content in two ways . In the phrase “painted picture” (gemaltes Bild), the term “painted” is used as an attributive term: a painted picture is indeed a picture . In this sense the painted picture, the result of the activity of painting, is identical with the painted landscape . In the phrase “painted landscape,” as it is used here, “painted” is a modifying term: a painted land‑ scape is not a landscape; it is a picture (“die gemalte Landschaft ist eben keine Landschaft,” p . 13) . In a similar way, we can say that the content of a presentation and the presented object are one and the same (p . 15) . The term “presented” is used here as a modifying term, for a presented object is not an object, but a content 4.

3 Sagt man dagegen von der Landschaft, sie sei gemalt, so erscheint die Bestimmung ‘gemalt’ als eine modificierende, denn die gemalte Landschaft ist eben keine Landschaft, sondern … ein Bild (Twardowski 1894a, pp . 13-14) . 4 So, in this passage Twardowski does not speak about the distinction between content and object . I therefore do not follow Fréchette’s reading: the analogy is here not between a picture and a presented content, on the one hand, and a painted picture and an object of presentation, on the other hand . Cf . (Fréchette 2010, pp . 231-232) . That analogy is introduced only later on in the text . In my interpretation of the passage, the identity of the painted picture and the painted landscape is not difficult to understand; in both cases, we speak about the painting . Content and Object 53

The distinction between content and object can be used to argue that Kerry’s distinction between the “presented as such” (Vorgestellter als sol‑ cher, p .19) and the “presented plain and simple” (Twardowski 1894b, p .17; Vorgestellter schlechtweg) is not clear . The phrase “presented as such” is still ambiguous . Both the object and the content of an act of presentation can be understood as “presented:” in the first case, the term is used in the attributive sense: the object is presented; in the latter case, it is used in its modifying sense: the presented object is not an object, but a content (p .20) . The distinction is immediately relevant for Twardowski’s interpretation of the Brentanian account of judgement: what is affirmed or denied is the object, not the content of an act of presentation (pp . 15-16) . When we say that an object is or is not (ein Gegenstand sei oder sei nicht; p . 16), we are talking about the presented object as object, not about the presented content . Sometimes, though quite reluctantly, Twardowski calls the act’s content an “intentional object,” to be contrasted with the real object, apparently following an existing terminology that was already noted by Höfler: “It can only be confusing when one takes at times the intentional object – thus the content – at other times, the real object as the object of a presentation .”5 The term “intentional” in the phrase “intentional object” has to be understood in its modifying sense, and the term “real” (wirklich) used by Twardowski in the same sentence is a restorative term, that is, a term that restores the original meaning of the noun after it has been modified in a prior context (see II .4) . In contrast to Höfler, Twardowski presents arguments for the distinction between content and object, which is for him a real distinction, and not merely a conceptual distinction . For these arguments, Twardowski refers to Benno Kerry, another student of Brentano, who must have stimulated Twardowski to read Bolzano’s Wissenschaftslehre in relation to the distinc‑ tion, and the question whether there are objectless presentations, as Kerry often refers to Bolzano in these contexts .6 Twardowski gives four possible arguments, of which he takes one to be based on a mistake . The first two arguments are formulated independently of any linguistic distinctions: (1) the content always exists as part of the act, while the object may not exist . When I think of a golden mountain, the content of my thought exists, but a golden mountain does not exist; (2) the object may have properties that the content does not have . The golden mountain is made of gold, but the

5 Cf . (Twardowski 1894b, p . 38) . Wie es also nur verwirrend sein kann, wenn man als Object einer Vorstellung bald den intentionalen Gegenstand derselben, also ihren Inhalt, bald ihren wirklichen Gegenstand bezeichnet (Twardowski 1894a, p . 40) . 6 According to Twardowski, the writings of Benno Kerry drew his attention to Bolzano’s Wissenschaftslehre (Twardowski 1926, p . 24) . In his Habilitationsschrift, Twardowski refers to Bolzano, Zimmermann and Kerry as those who have made the distinction between content and object (Twardowski 1894a, p . 17) . For a comparison of Twardowski’s distinction with the one in Bolzano and Zimmermann . See (Fréchette 2010, pp . 239ff .) . 54 Maria van der Schaar content of my thought is not made of gold . In the explanation of the third argument (3), Twardowski makes use of two names for Salzburg: “the city located at the site of the Roman Juvavum” and “the birthplace of Mozart .” “The two names have a different meaning, but they both designate the same thing” (1894b, p . 29) . In Brentano’s logic lectures, the distinction is also made from a semantical point of view 7. The presentations expressed by these names are called equivalent presentations (Wechselvorstellungen), a term Twardowski borrows from Bolzano . Equivalent presentations with contents A and B present the same object now as A, the next moment as B (p . 82) . Because of the analogy between names and presentations, equiva‑ lent presentations can now be defined as “presentations in which a different content, but through which the same object, is presented” (idem) . We may also have equivalent presentations of objects that do not exist . I come back to this topic of intentional identity in the section 3 . A fourth (4) argument, proposed by Kerry, would be that the content may be the same while the object differs . This is said to happen in the case of general presentations: the presentation of a triangle in general would have the same content each time a triangle is presented, although the particular triangles, which would be the objects of a general presentation, may differ from case to case . Be‑ cause Twardowski gives a unique account of general presentations, he un‑ derstands the argument from general presentations as flawed . The general presentation of the triangle as such has one object: a general object (see the sub‑section on the general object at the end of section 3) . Twardowski’s arguments have had some influence on Meinong, as Meinong will equally assert that we may present something that does not exist .8 Part of the second argument, that the golden mountain is made of gold, although it does not exist, comes back in Meinong’s principle of the independence of so‑being from being: das Prinzip der Unabhängigkeit des Soseins vom Sein (Meinong 1904, p . 489) . For the object to have properties, it need not exist .

7 The content of a presentation is the meaning (Bedeutung) of a name . The external, existing object of the presentation is named by the name (Brentano 1870, 13 018);. cf . Chrudzimski (2001, ch .1) and Rollinger (2009, p . 7) . Sohn der Phänerete and der Weiseste der Athener name the same object, but have a different meaning: Sie nennen unter Vermittlung verschiedener Bedeutung (Brentano 1870, 13 . 019 [5]) . Like Bolzano, Brentano acknowledges objectless presentations: Es gibt gegenstandslose Vorstellungen, according to Rollinger (2009, p . 17, note 37) . Das zweite ist das, was der Name nennt. Von ihm sagen wir, es komme der Name ihm zu. Es ist das, was, wenn es existiert, äußerer Gegenstand der Vorstellung ist (Brentano 1870, 13 018. [4]) . Cf . also Höfler’s Logik, §6 and §17 4. . Twardowski’s denial of objectless presentations is thus unique at the time, and definitely before Meinong defended the idea . 8 Nichts ist gewöhnlicher, als etwas vorzustellen oder über etwas zu urteilen, was nicht exis‑ tiert (Meinong 1899, p . 382) . Content and Object 55

The distinction between content and object has also been made by a British reader of Franz Brentano, G .F . Stout, who was both a philosopher and a psychologist, and the teacher of G .E . Moore and Bertrand Russell in 1894/1895 . Stout’s distinction shows a great similarity with the one presented by Twardowski . Although Stout has made the distinction before 1894, it can be shown that he wrote the anonymous review of Twardowski’s book for Mind published the same year, and he will use all four arguments Twardowski has presented in his work on Analytic Psychology (1896) . Stout considers, like Twardowski, the content to be a psychological entity, dependent upon the mental act, but, as far as semantics is concerned, he draws a different conclusion from this thesis: he considers the content to be too psychologi‑ cal to play a role in semantics . For Stout, the basic semantic relation is that between a term and the object designated . It is this position that has made it possible for Moore and Russell to defend a Gegenstandstheorie before having read Meinong . There is no sign of a theory of objects in Meinong before 1899, and Moore already defend a version of a theory of objects in his second dissertation of 1898, of which the important parts are published in Mind in 1899 . Twardowski has thus played an important, though indirect role in the development of Moore and Russell towards a direct realism in their early years 9.

2. The Content of the Act

For Twardowski, the content has two roles or functions: (1) it is the content of an act of presentation; and (2) the meaning of a name . Starting with the first role, the content of the act is that through which the object of the act is presented (p . 16); it has a mediating function (p . 28), directing the act to this object rather than another . The same object may be presented in dif‑ ferent ways, that is, by means of different contents, even if the object does not exist: “Admittedly, a circle in the strict geometric sense does not exist anywhere . Yet one can conceive of it in different ways (Doch kann man ihn auf die verschiedenste Weise vorstellen; cf . (Twardowski 1894a, p .32)), be it as a line of constant curvature, be it … as a line whose points are at the same distance from a given point” (Twardowski 1894b, pp . 29-30) . Twardowski is thus able to account for what Geach has called intentional identity, and I come back to this in section 3, when Husserl’s reaction to Twardowski’s account of intentionality is given . The content has also two characteristics: (1) it exists only as part of the act, and (2) cannot be considered as a thing

9 See my work on Stout: (Schaar 1996) and (Schaar 2013) . 56 Maria van der Schaar or res, that is, as something that stands in relations of cause and effect . The content is in this sense existent, but not real . “Someone who utters a name intends to awaken in the listener the same mental content which appears in himself” (Twardowski 1894b, p . 9) . In this passage, Twardowski assumes that one can speak of the same mental content in the speaker and the listener . At the same time, though, Twardowski argues that the content is dependent upon the act of presentation, and in this sense there cannot be an identical content for different acts . In 1894, the content is understood as a dependent moment of the act, and the content of two different acts can thus never be identical . In a footnote, Twardowski gives an elucidation of how the meaning of a name can be understood as general . The meaning of a term is the mental content that the name aims to awaken in the hearer, falls er zugleich die Fähigkeit hat, dieses Ziel in der Regel zu erreichen (Twardowski 1894a, p . 11, note), that is, on condition that the name has the capacity to evoke this content as a rule in whoever hears the name . Through this condition, accidental aims to arouse a certain content in the hearer do not belong to the meaning of the name, and it seems that the idea of a rule gives some generality to the name . The content that a particular name may evoke differs, though, from hearer to hearer: the name “Socrates” may evoke a picture of Socrates in one person, or simply a picture of the word “Socrates” in another . The name has its logical function independent of the immanent, particular content of these acts . The name may have its logical function without arousing a similar psychological content in who‑ ever hears and understands the name . In this period, Twardowski does not make a distinction between the logical and the psychological content, and it is no surprise that Husserl criticises Twardowski’s semantics precisely on this point . Twardowski confuses, according to Husserl, the real, psycho‑ logical content with the ideal, logical content (Husserl 1896, p . 350, note) . Cf . (Hickerson 2007, p . 62) . As Twardowski identifies the meaning of the name with the content of the act, there is no way he can prevent a form of psychologism as far as the content is concerned . The semantics presented in 1894 is therefore psychologistic .10 And it is precisely for this reason that Twardowski develops a new semantics after 1900 . In 1894, the content of the act is an immanent aspect of the act which differs from presentation to presentation, and is in this sense an irrepeatable entity . Twardowski as‑ sumes that the content has a representative or logical function making the act an act of this object as an A, but such a representative function goes

10 Jens Cavallin (1997, p . 88) claims that, for Twardowski, contents are “objective ”. As far as I have been able to determine, the term is used by Twardowski only in a footnote, where Bolzano’s position is described: Bolzano gebraucht statt des Ausdruckes ‘Inhalt einer Vor‑ stellung’ die Bezeichnung ‘objective’ Vorstellung, ‘Vorstellung an sich’ (Twardowski 1894a, p . 17, note) . Content and Object 57 beyond the particular psychological content . The logical content thus has to be a repeatable meaning, but Twardowski is not able to account for this universal aspect by the purely particular and psychological account of the content he has given . He cannot account for the fact that you and I may both think of this object as an A, that is, that we may think of it by means of the same logical content . And neither can he account for the fact that, although I have in mind only a picture of the words “lime tree,” I am thinking of the green lime tree I have seen this morning . The idea of a rule mentioned above, which is only presented by Twardowski in a footnote, can be used to develop a notion of logical content, and Twardowski will do so after 1900 . According to R . Grossmann, the best translation of the German term Vor‑ stellung is not “presentation,” but “idea .” This translation shows the connection of the account of Vorstellung given by Brentano and Twardowski with the empiricist notion of idea . If one systematically translates the term Vorstellung with the term “idea” there is a danger, though, that one blurs a distinction central to the writings of Brentano and Twardowski . In the empiricist tradition, the act/object ambiguity is often neglected, and the term “idea” may either stand for the act or for the object, or for both at the same time . Grossmann proposes, more specifically, to use “idea” for the content of the act: I shall speak of an idea, the act of having an idea, and the object of an idea where Twardowski speaks of the content of a presentation, the act of presentation, and the object of a presentation, respectively (Grossmann 1977, p . viii) . The problem of translating Inhalt (“content”) as “idea” is that it seems in this way that Twardowski defends the empiricist representational theory of thought . According to John Locke, the object of thought is always an idea, and the idea is a representation of the external object in the mind . Twardowski is not committed, though, to this form of representationalism . For Twardowski, the act is directed to the object, not to the idea or content . It is precisely because the content does not function as object of the act, that Twardowski has to say that all acts need both a content and an object, for there is always something to which our acts are directed, which is distinguished from the act’s content . Twardowski’s account of thought differs from that of the traditional repre‑ sentational theories we find in the empirical tradition insofar as Twardowski accounts for the intentionality of thought . Twardowski’s account of inten‑ tionality also differs in a crucial moment from Brentano’s . For, the object of thought and perception is something that is understood to be independent of this particular act of thinking . His account of intentionality thus has an aspect that transcends the particular act of presentation: the immanent content is not the target; it merely has a mediating function . It is precisely for this reason that Twardowski’s account of intentionality is of importance for the development of phenomenology in Husserl’s writings (see section 3 on Husserl’s reaction) . 58 Maria van der Schaar

According to Twardowski, the content is not to be understood as a picture of the object (p .64) . He rejects such a primitive form of psychology . There is a sense, though, in which Twardowski can be understood as standing in the tradition that claims that the content is a quasi‑picture or sign of the object, and thereby represents the object in the mind . Such a sign need not have a photographic resemblance with its object . There is thus a way in which Twardowski’s account can be called a representational account of thought, as the interesting paper by Ryan Hickerson (2005) shows, although it should be noted that the content is not a substitute for the object, for the act is directed to the object, not to its content . Like the representationalist, Twardowski points to the fact that content and object of the act may agree in the structure they exhibit . There may be an isomorphic relation between content and object . If the object is presented as simple, the relation between content and object is a primitive one that cannot be further explained . If the object is presented as complex, there is, according to Twardowski, a rela‑ tion of isomorphism between the content and some aspects of the object . “[T]here is an analogy between the composition of the parts of the object and the composition of the constituents of the content” (Twardowski 1894a, p .65) . This analogy is quite strong, not in the sense that to every part of the object belongs a part in the content, for our presentations are always inade‑ quate, but in the sense that every part of a presentation is the presentation of a part of the object . And this means that to every part of the content belongs a part of the object, as Twardowski claims in opposition to Bolzano (p . 88) (see also chapter II .2, and section 3 below, the sub‑section on mereology) . Until now we have distinguished four functions to be fulfilled: (1) being the content of an act, and thereby having a mediating role; (2) being the meaning of a name; (3) being what the act is about, that to which the act is directed; (4) being what is named by a name . What kind of entity fulfils these functions? According to Twardowski, the third and the fourth function are fulfilled by the object of the act . What func‑ tions as content of presentation is called the “content” . The content fulfils both the function of mediator between act and object, and the function of meaning of a name, according to Twardowski, that is, the content fulfils the first and the second function . The content is a particular dependent aspect of an act of presentation, and not an abstract entity or Platonic idea . It exists here and now as part of the act, and it has no independent existence . The distinction between roles and the kind of entity that fulfils these roles is relevant for the question whether the content as entity can ever fulfil the role of being the object of an act . In an act of reflection it is possible, according to Twardowski, to think of the content of another act, where this content has Content and Object 59 the function of being the object of the reflective act . Although the content of an act can never exist independently, it can be considered in abstraction from the act to which it belongs, and thus become the object of a new act .

3. The Object of the Act

From Psychology to Metaphysics Twardowski is faithful to Brentano’s thesis that all presentations are directed to an object . For Twardowski, every act of presentation has an object; every name designates something . This goes directly against Bolzano’s position in the Wissenschaftslehre, where it is argued that there are objectless presen‑ tations (gegenstandlose Vorstellungen; cf . (Bolzano 1837, §67)) . Bolzano gives as examples of objectless presentations the presentation of nothing, of a round square and of a golden mountain . To start with the last two examples, according to Twardowski, the incompatible properties of being round and being square are attributed to the object; they cannot be attributed to the content, for the content exists, and therefore cannot have any incompatible properties . So, there is an object, although it does not exist . The word “nothing,” Twardowski says, is a syncategorematic term (see II 2):. the word on its own does not express an act of presentation, or any other act . In standard cases, where “non” precedes a categorematic term, we get a new categorematic term, such as “non‑Greek .” This makes sense only if a superordinate concept, human being, is divided into concepts such as Greek and non‑Greek . The latter is thus a categorematic term . If there is no superordinate concept, as is the case with the most general concept some‑ thing, we cannot form a new categorematic term by putting “non” in front of “something,” thereby obtaining the term “nothing ”. “Nothing” is thus not a categorematic term . And for this reason, it does not express an act of presentation (cf . II .2); and no objectless presentation is thus involved . The not too clever Cyclops Polyphemus didn’t understand that “nobody” cannot be a name, when the cunning Odysseus said that his name is “nobody .” When he shouted that nobody was trying to kill him, none of the other Cyclopes came to help him . “Nothing” has no meaning in isolation: “One can say neither that “Nothing” exists nor “Nothing” does not exist” (Twardowski 1894b, p . 32); “ “Nothing is eternal” means “There is not something which is eternal” ” (p . 20) . Assuming that a judgement is made, the logical form is: (eternal); the term “nothing” has disappeared in the logical analysis of the sentence . What are these non‑existing objects like golden mountains and round squares? At first sight, it seems that for Twardowski the object of presenta‑ 60 Maria van der Schaar tion is nothing but a counterpart to the mental act: “We have called object that entity which is presented through a presentation, judged in a judgement, and desired or detested in an emotion” (p .32) . The meaning of “object” coincides in this sense with the meaning of the expression “phenomenon” or “appearance” (p . 33) . The object is not a Ding an sich, Twardowski adds . He does not want to make any metaphysical assumptions at this stage of his analysis: “[W]hatever [the object] may be, it is … the object of these acts, in contrast to us and our activity of conceiving” (p . 33) . The act “relates to an object which is presumed to be independent of thinking” (p . 7, M .S .; the German reads: als unabhängig vom Denken angenommenen Gegenstand) . A psychological analysis disregards “the real, possible, or impossible exis‑ tence of objects and their parts” (p . 49) . This does not mean, though, that Twardowski claims that the object is dependent upon the act of presentation . At a certain moment in Content and Object, Twardowski changes his psychological approach to a metaphysical one (p . 34, italics M .S .): Everything which is, is an object of a possible presentation; everything which is, is something . And here, therefore, is the point where the psychological discussion of the difference between content and object of presentations turns into metaphysics . In the first sentence Twardowski claims that the object is independent of the act, adding that some objects exist, while others do not . Even what can never exist is an object (p . 35) . Twardowski’s theory thus implies that existence is a property of objects, although Twardowski never explicitly says so, for this would go against the Brentanian theory of judgement . What is non‑real, such as a lack of something, is an object, too . Everything that is presented as an object, is presented as a unified whole, and is thereby an object . Independent of the act of presentation, every object is a unity, and sets itself off against all others (p . 86) . Having such a unity is a defining criterion for being an object . The non‑existing objects have what the Scholastics called an “objective, intentional existence” (p .23) . This is not to be taken in the sense that these objects exist in a special way, that they subsist, as Russell would say . The term “intentional” has here its modifying function: intentional existence is not existence at all . If an object exists as presented object, this does not imply that it exists, for “as presented object” modifies the meaning of “ex‑ ists” (p .25) . So, some objects exist, while other do not . It is to be noted that Twardowski does not speak here about the content of presentation, which exists due to the fact that it is a dependent part of an existing act (p . 24) . Twardowski’s notion of intentional existence differs from Brentano’s notion of intentional existence, at least, on one possible interpretation of his writ‑ ings . On that interpretation, all contents have intentional inexistence insofar Content and Object 61 as they exist in the mind, or in the act . When I think of a stone, my thought is directed to a content, which exists, in the mind, though not as a stone . It is thus clear that Twardowski develops a theory of objects, a Gegen‑ standstheorie, in 1894, before Meinong . Metaphysics is “the science of objects in general,” investigating the laws which objects in general obey (p . 36) . The object of metaphysics is thus not unlike the object of logic for Frege and Russell; for them, logic is the most universal science . The logical laws concern all objects . As Russell says: “Logic is concerned with the real world just as truly as zoology, though with its more abstract and general features” (Russell 1919, p . 169) . There is one striking difference between the scope of logic in Frege’s writings, on the one hand, and the scope of metaphysics in Twardowski . Frege’s logic is not concerned with possible or impossible objects; logic does not apply to fiction, Frege would say . From a metaphysical point of view, it seems that for Twardowski the objects are understood as independent of any mental act . Everything which is, is an object of a possible presentation, that is, objects are independent of any actual act of presenting . As far as his theory of objects is concerned, Twardowski is not a psychologist . The object is independent of the mental act . Still, the foundation for this metaphysics is a psychological investigation: “Everything which is in the widest sense “something” is called “object,” first of all in regard to a subject, but then also regardless of this relationship” (p . 37) . Psychology is in this sense a heuristic means for metaphysics, and is thus in a genetic sense prior to the latter . Psychology is also providing the basic concepts for metaphysics, and Twardowski often makes use of inner perception in his explanations . In this sense one may say that a form of methodological psychologism is defended . A full theory of objects and their parts, a mereology, is developed by Twardowski in the later sections of Content and Object (see the sub‑section on mereology below) . Already in 1894, we thus see a move from psychology to ontology, without cutting the umbilical cord .

Husserl’s Reaction to Twardowski’s Account of Intentionality Husserl had a high esteem of Twardowski’s work on content and object, but also formulates some severe criticism . Twardowski gets into trouble “in his speaking of “a presenting activity moving in a double direction,” in his completely overlooking the meaning in an ideal sense, in his psychologis‑ tically spiriting away evident differences between meanings by recurring to different etyma,11 in the way he treats the doctrine of “intentional existence” and the doctrine of universal objects” (Husserl 1901, p . 528) .

11 Daß er [Twardowski] Bedeutung und direkt‑anschauliche Vorstellung des bedeuteten Gegen‑ standes identifiziert, während ihm der fundamentale und logisch allein maßgebliche Begriff 62 Maria van der Schaar

There are four important places in which Husserl gives an explicit re‑ action to Twardowski . There is a posthumously published manuscript on intentional objects written in the summer of 1894, Intentionale Gegenstände (Husserl 1894), in which Husserl develops his own account of intentionality in reaction to Twardowski . Husserl wrote a review of Twardowski’s trea‑ tise on the distinction between content and object of presentation, which is published posthumously (Husserl 1896) . And Husserl gives two reactions to Twardowski in the Logical Investigations: one reaction is given in the context of a critique of the picture theory; the other one is a critique of Twardowski’s account of general objects, and will be dealt with in the sec‑ tion on general objects below . From a historical point of view, Husserl’s earliest reaction to Twardowski in the manuscript on intentional objects is the most important one, for it can be shown that Husserl’s concept of intentionality is primarily developed in reaction to Twardowski, rather than in reaction to Brentano . Cf . (Schuhmann 1991b, p . 138) and (Husserl 1902, p . 144) . The manuscript on intentional objects has been part of a bigger one called “Vorstellung and Gegenstand .” The lost part probably dealt with ideal contents or meanings, as the preserved manuscript starts with a summarizing remark that to every presentation belongs a meaning‑content (Bedeutungsgehalt; cf . (Husserl 1894, p . 142)) . The part on intentional objects opens with two theses that seem to con‑ tradict each other . On the one hand, every presentation presents something (jede Vorstellung [stellt] einen Gegenstand vor, (idem)); there is thus an object for every presentation . On the other hand, there does not correspond an object to every presentation (nicht jeder Vorstellung [entspricht] ein Gegenstand), for the presentation of a present emperor of France or the pre‑ sentation of a round square do not correspond to an object . For the second thesis Husserl refers to Bolzano’s thesis that there are objectless presenta‑ tions . Husserl will criticise the first thesis, which may refer to Brentano, although Husserl does not mention him . When he gives a critique of the first thesis, it is rather a theory like Twardowski’s he has in mind, for his critique concerns an account of intentionality that acknowledges both a content and an object of thought . The tension, Husserl adds, also applies to sentences . Every sentence presents a state of affairs, which does not exist or subsist in case the sentence is false, and yet there does not correspond to every sentence a state of affairs . The first thesis preludes Twardowski’s position in 1897, as we will see later . In the first place, Husserl criticises a picture theory of presentation as it is believed by the masses . He does not mention Twardowski here, because Twardowski denies his contents to be simple pictures . The theory presumes der Bedeutung ganz entgeht. Daher verfällt er darauf, Bestandstücke der Bedeutung (‘ohne Berge’) als ‘Hilfsvorstellungen nach der Art der Etyma’ zu faßen (Husserl 1901, p . 305) . Content and Object 63 to account for both theses . On the one hand, every presentation presents an object, that is, a picture; on the other hand, not every presentation corresponds with an object, that is, there is not always an object of which the picture is supposed to be a picture (p . 144) . The first problem with the picture theory is that it suggests that we may form a picture, although there is no object . Because we can easily form a picture of what does not exist, the picture theory lies at the basis of the thesis that not every presentation corresponds to an object (p . 143) . Second, there are presentations of art, literature and science whose object is so abstract that we cannot form any picture of them; and, what kind of picture are we supposed to make of a round square? The picture theory is absolutely not confirmed by experience . Third, the the‑ ory cannot account for the fact that it is the same object that is presented, whether it exists or not: “The same Berlin, that I present, also exists, and the same town would no longer exist, were it punished by God’s judgement as had happened to Sodom and Gomorrah ”. 12 We are never interested in the picture: “The object itself we present, about this we judge, to this we are directed in joy and sorrow, wish and wanting ”. 13 The problem of the theory is that the “two kinds” of presentations are directed to “two kinds” of objects, pictures and the intended objects, and there is therefore only an apparent solution of the problem . Finally, the picture is not automatically a picture of something . In order for the content to function as a picture or representation of the ob‑ ject, the content must already possess a certain pointing beyond itself, an Über‑sich‑Hinausweisen, that makes it into a picture or representation, and by which it is distinguished from a mere psychological content of the act (idem) . Does this argument apply also to Twardowski’s quasi‑picture theory, in which contents are quasi‑pictures or signs of the represented object? To put it differently, is Husserl’s critique applicable to any representation theo‑ ry? In the Logical Investigations, Husserl repeats the critique of the picture theory, and adds that an account of intentionality in terms of quasi‑pictures or signs is as vulnerable to this critique as any other representation theory . The picture theory says that we are able to think of a certain object, because the act of thinking has an immanent content that shows a resemblance with the object . According to Husserl, a resemblance between content and object of presentation does not make the content a picture of the object . In order for the one to be a picture of the other, we need someone who interprets the content as a picture of the object (Husserl 1901, p .436) . Being a picture

12 Dasselbe Berlin, das ich vorstelle, existiert auch, und dasselbe würde nicht mehr existieren, bräche ein Strafgericht ein wie bei Sodom und Gomorrha (Husserl 1894) . Cf . (Husserl 1896, p . 353, note) . 13 Den Gegenstand selbst stellen wir vor, über ihn urteilen wir, auf ihn beziehen sich Freude und Trauer, Wunsch und Wille (Husserl 1894, p . 144) . 64 Maria van der Schaar is not a characteristic of the content by means of which the content may function as representation of the object . Being a picture is rather a charac‑ teristic that can obtain only on the foundation of an act, but an account of such an act has to be given in terms of intentionality, the concept one hopes to explain . We have thus a circle in our explanation . This critique equally holds for any representation theory of intentionality, whether the content is a picture or a sign . If the content is understood as a sign of the object, one can say that being a sign is not a characteristic of the content independent of a mental act in which something is understood as a sign, and one needs to give an account of intentionality for such an act, without using the idea of sign or representation (Husserl 1901, pp . 437-438) . It is in this sense that Husserl’s critique of the representation theory is supposed to apply to Twardowski’s quasi‑picture account . Husserl’s critique would be correct if Twardowski were to assume that reference to an object, intentionality, is a property of the content, but Twardowski does not say this . For Twardowski, it is the act which directs itself beyond the content to the object . Each act of presenting or judging is characterised by a special intentional relation to the object (Was Vorstellungen und Urteile von einander scheidet … ist die besondere Art der intentionalen Beziehung auf den Gegenstand; cf . (Twardowski 1894a, p . 5)) . On Twardow‑ ski’s account, it seems that the mental act takes the content to be a content of the object . I can think of the ball by means of a colour or a shape that is not at all the colour or the shape of the ball . It is precisely for this reason that the possibility of error arises . I take the colour and shape to be a sign of the ball, but there is not an internal relation between the object and the content, independent of the act, that makes the one a sign of the other . On this interpretation of Twardowski, Husserl’s critique just mentioned does not apply . It is also true, though, that Twardowski’s formulations suggest that there is a special relation between the content and the object that is to account for the intentional relation . In the section on the relation between object and content of presentation, Twardowski argues that there is an analogy between the structure of the content and that of the object of a presentation . The ma‑ terial elements of a content of presentation relate (beziehen sich auf; p . 69) partly to material and partly to formal elements of the corresponding object . In his reaction to the picture theory, Twardowski asserts that there must be a certain relation between content and object, in virtue of which an object belongs to a certain content, and in virtue of which a certain content is the content of this and no other object .14 It is for this reason that Twardowski’s

14 Nun muss gewiss eine Beziehung zwischen dem Inhalt und dem Gegenstand bestehen, kraft deren ein Gegenstand eben zu diesem bestimmten Inhalte gehört und ein Inhalt eben der ei‑ nem bestimmten – und keinem anderen – Gegenstande entsprechende Inhalt ist (Twardowski 1894a, p . 67) . Content and Object 65 theory is a quasi‑picture theory, as we have seen in the former section, and Husserl’s criticism correctly applies to these passages . It is the former point, though, this pointing beyond itself as an aspect of the act of thinking that is central to Twardowski’s account of intentionality . And this idea has been crucial to the development of Husserl’s account of intentionality . Where for Brentano, the intentionality of mental acts con‑ sists in the fact that they have a content that is immanent to the act, for Twardowski the intentionality of mental acts consists in a directedness of the act towards an object that transcends the act . Twardowski does not use the term “intentionality” for this pointing beyond, as he sticks to Brentano’s terminology in which the so called intentional object is immanent to the act . Twardowski uses the term “intentional object” to refer to the content of the act, not to the act’s object . Cf . (Twardowski 1894a, p . 40); (Twar‑ dowski 1894b, p . 37 and p . 38)15 Although Twardowski is a clearer writer than Husserl, Husserl has created the more convincing terminology for the problem of what we now call intentionality . For Husserl, intentionality is that character of the act that makes the act directed to the intended object . In an intentional experience, the object to which the act is directed is the intended or intentional object, which may or may not exist (Husserl 1901, LU V, §11, p .387) . Jupiter is not presented in a different way than Bismarck, Husserl says, and Twardowski agrees on this point, but for Twardowski this implies that there is an object Jupiter independently of the mental act, which may become the object of different presentations of Jupiter, whereas for Husserl Jupiter can neither be found in the act of presentation, nor outside the mind: er ist überhaupt nicht (idem) . Husserl’s next critique in Intentionale Gegenstände is explicitly directed to Twardowski, and concerns Twardowski’s idea of intentional existence . According to Twardowski, every act has an object; in case we think of a round square, there is an object, although it has only intentional existence . For Twardowski, the term “intentional” has here its modifying meaning; it modifies the meaning of “existence” in such a way that the round square does not exist . Twardowski refers in this context to Brentano, and does not take him to argue for two modes of existence (Twardowski 1894a, p . 25) . There is one important difference, though, between Twardowski and Brentano . For Brentano, the content has intentional existence, whereas for Twardowski,

15 The secondary literature on Twardowski often takes over Husserl’s terminology that has been so successful in the twentieth century, and thus uses the term “intentional object” for the object to which the act is directed when giving an interpretation of Twardowski . Compare (Jacquette 1991, p. 183). Rightly so, for the term ‘the intentional object’ as we understand it refers to the object, not to the content of the act . Grossmann (1977) uses the term “intention” for the object to which we are directed . This need not be a problem as long as the reader knows that Twardowski himself uses the phrase “intentional object” for the immanent object or content, following Höfler’s terminology in the Logik . 66 Maria van der Schaar the object has intentional existence . For Twardowski, the content really exists as part of an existing act . For him, intentional existence is not a form of existence; it characterises the object of presentation insofar as the ob‑ ject does not exist, but is the object of a mental act, nonetheless . Existence and intentional existence function here as first order predicates, although Twardowski does not expicitly endorse this thesis, for it goes against Bren‑ tano’s theory of judgement . I see no other way to give an interpretation to Twardowski’s thesis that “some objects have existence in addition to their objecthood” (Twardowski 1894b, p .35) . Twardowski is thus able to give a solution to the apparent contradiction (der scheinbare Widerstreit, p . 145) mentioned above: every presentation has an object, although the object may not exist truly (wahrhaft) . According to Husserl, Twardowski’s terminolog‑ ical move does not explain the problem, for it is unclear what “intentional existence” means, and what wahrhaft means (Husserl 1894, p .145) . It is true that Twardowski’s move seems to be a terminological one, although certainly an interesting one . Husserl is right, though, that Twardowski ac‑ knowledges both intentional and truly existing objects . As for Brentano, for Husserl, presentations are neutral as far as the existence or non‑existence of their objects is concerned; the existence of an object becomes relevant only at the level of judgement 16. According to Husserl, there is on Twardowski’s account, as in the picture theory, a doubling of objects . Twardowski acknowledges for every mental act both a content and an object, but this is not the theory that Husserl at‑ tributes to him . Husserl assumes that on Twardowski’s account, whether we think of Jupiter or of Bismarck, the object of thought is an object that is immanent to the act: [M]an glaubt doch eine Lösung zu haben … daß jeder Vorstellung ein immanenter Gegenstand, aber nicht jeder ein wahrer Gegenstand zugeteilt wird (p . 146) . This is Twardowski’s position insofar as he considers the content of the act to be an immanent object, but the immanent object is not the object to which the act is directed, that is, the object that may have merely intentional existence in case we think of Jupiter . For Twardowski, the content or im‑ manent object of the act always exists as part of the act, whether we think of Jupiter or Bismarck . Husserl’s point that the immanent object exists as much as the act, is thus endorsed by Twardowski . In the next paragraph, Husserl again assumes that for Twardowski the object of a presentation that has merely intentional existence is an immanent object:

16 For Husserl, the notion of modification becomes relevant only on the level of judgements . If one makes the judgement that “Pegasus has wings” the judgement is a modified one; such a judgement can be true only under the assumption that Greek mythology were true . Content and Object 67

Man bleibt also dabei, den vorgestellten Gegenstand als ein der Vorstellung im ei‑ gentlichen Sinne Einwohnendes zu fassen (p . 146) . Twardowski’s distinction between content and object of the act makes it possible to say that the presented object, whether we think of Bismarck or Jupiter, is not identical with what exists in the act as a content of the act, and Husserl thus has not understood Twardowski’s position . As in the case of the picture theory, Husserl argues against Twardowski that the Ber‑ lin that I present and the Berlin that is judged to exist is the same Berlin (p . 146) . The Berlin one thinks of is the existing Berlin, Husserl says . If Twardowski were to endorse the attributed position, the criticism would be appropriate . But Twardowski is here on Husserl’s side: the Berlin that I think of is the existing Berlin, according to Twardowski . Compared to Brentano, Twardowski is already making the important step: the objects to which we are directed in thought transcend the act of thinking . But, in con‑ trast to Husserl, Twardowski does not use the term “intentional object” for this act‑transcending object . He uses that term for the immanent content of the act, which makes it understandable that Husserl interpreted Twardowski the way he did . Karl Schuhmann’s interpretation of Twardowski’s theory is probably influenced by Husserl’s incorrect reading of the text . According to Schuhmann, when Twardowski speaks about the meaning of the phrase “presented object,” where “presented” is taken in its modifying sense, he refers to an object that is not a real object, and therefore has merely inten‑ tional existence (Schuhmann 1993, p .45) 17. For Twardowski, though, the “presented object” has here its modified sense refers to the content of the act . This content does not have intentional existence; it exists as much as the act whose part it is . For Husserl, the problem how we can think of non‑existent objects can be solved only in terms of the judgements we make about such objects . It does not make sense to make a division between existing and non‑existing objects, Husserl argues . We may think of a lion behind a tree, which may or may not exist, but this does not mean that there are two kinds of lions, the existing and the non‑existing ones (Husserl 1894, p . 149) . According to Husserl, such a division is not a division in two kinds of objects, but rather a division in two kinds of functions that presentations may have . There are presentations which may become a foundation for judgements in which the existence of the presented object is rightly acknowledged . And there are presentations which may become a foundation for judgements in which the existence of the presented object is rightly denied (idem) . It is true that

17 Cf . also: “Twardowski had considered the immanent object to have a merely intentional existence, i .e . no true existence at all” (Schuhmann 1993, p . 50) . Twardowski would fully agree with Husserl’s thesis that what is immanent has genuine existence . Rollinger (1996, pp . 119, 120) sees that Husserl does not give a correct interpretation of Twardowski . 68 Maria van der Schaar

Twardowski endorses the thesis that there are two kinds of objects, those that exist and those that merely have intentional existence, but he does not say that there are two kinds of existence, for intentional existence is not a certain kind of existence . For Twardowski, Jupiter is an object, although he does not exist . For Husserl, we can only speak of the independence of the object of an act of presentation on the basis of judgements in which the existence of such an object is affirmed . It is only a small step to Husserl’s later position in which objects are constituted in judgements of existence and identity, a position far apart from Twardowski’s . Twardowski’s theory is clearly able to give an account of what Peter Geach has called intentional identity (Geach 1972, pp .146ff) . How can we refer to an object introduced earlier in the conversation although this object may not exist? How is it possible to refer to the girl that comes back in our dreams again and again, whereas she does not exist? According to Twardowski, there is a girl, although she does not exist . For Husserl, iden‑ tity is constituted in identity judgements that presuppose the correctness of the acknowledgement of the existence of the object . This means that one cannot account for intentional identity if the object does not exist . How can Husserl explain that we think of Jupiter in different ways although there are no Greek Gods? For Husserl, intentional identity is possible only under the assumption that, for example, Greek mythology is true (Husserl 1894, p . 151) . Even in the case of geometry, we make assumptions about space . It is only upon reflection that we obtain the right form for these judgements, and understand that these are judgements only in a modified sense of the term insofar as they are only “judgements under an assumption” . Husserl’s account is not to be taken in the sense that there are different possible worlds: the world of geometry, that of Greek myth, and the actual world . For Husserl, there is only one world, and the judgements that directly relate to this world do not show any assumptions in their proper form (p .159; see footnote 16, above) . Husserl’s other important critique of Twardowski concerns the fact that Twardowski identifies the content of a presentation with the meaning of a name . The two cannot be identified, for the meaning of two “names” may be the same, although the immanent content of the act may be different, and vice versa (Husserl 1896, p . 349, note) . This critique is dealt with above, in section 2 .

Metaphysics and Mereology A general theory of objects is an important part of Twardowski’s work on the distinction between content and object of presentations . Although Twardowski starts with an account of the object as counterpart to the act Content and Object 69 of presentation, he soon takes up the metaphysical questions what kind of objects there are, what their parts are, what their unity is, and whether the structure of a content may differ from that of the corresponding object .18 It will turn out that Twardowski defends a particularistic metaphysics: there are no entities like Platonic forms or Aristotelian essences . An important question is therefore how we can have general thoughts, how we can think of a triangle in general . This topic will be addressed in the next sub‑section . Carl Stumpf had developed a theory of wholes and parts in his work on the psychological origin of the presentation of space, and in his work on the perception of tones . Stumpf makes a distinction between independent parts (selbständige Inhalte) and dependent parts (Teilinhalte; cf . (Stumpf 1873, p . 109)), which distinction is going to play an important role in Twardowski’s account of objects . For Stumpf, though, the distinction applies to the con‑ tents of our consciousness alone . Twardowski is the first Brentano student who applies such a theory of wholes and parts to objects in general, due to the clear distinction between content and object of consciousness in his work . To see Twardowski’s importance here, we may compare his position with that of Meinong at that time . In 1894, Meinong’s distinction between founded (dependent) and founding (independent) parts is still a distinction between contents alone . As soon as Twardowski proposes a theory of objects from a metaphysical point of view, he relates his theory of objects to the Scholastic tradition . Objects are in the first place characterised by their objecthood (Gegenständ‑ lichkeit; cf . (Twardowski 1894a, p . 37)): they are what can be presented, what can be the object of thought; they are something (in irgend einem Sinne “etwas”; p . 38);19 second, they form a summum genus, a highest genus . There is no higher genus than being an object, than being something . Objects are potential objects of judgements and emotions, and are thereby, respective‑ ly, true, verum, and good, bonum, not in the sense that they are objects of correct existential judgements, and objects of correct emotions; it means that objects are judgeable, and emotions are evaluable . All objects can be

18 Metaphysics as a general theory of objects turned out to be a recurrent theme in Polish philosophy . Even Tarski did not have any problems with this conception of metaphysics: “For some people metaphysics is a general theory of objects (ontology) – a discipline which is to be developed in a purely empirical way, and which differs from other empirical sciences only by its generality … I think that in any case metaphysics in this conception is not objectionable to anybody” (Tarski 1944, p . 363) . Cf . also Twardowski’s empirical conception of metaphysics explained in II.1. Leśniewski uses the term “ontology” for his logic of names, “because he thought that his logic of names formulates ‘general principles of being’” (Woleński 1989, p. 153) . 19 First, Twardowski seems to explain objecthood as the property of being presented, but a few lines later it becomes clear that what can be presented (was … vorgestellt werden kann (Twardowski 1894a, p . 37)) is a more precise formulation . 70 Maria van der Schaar understood to be one, or unum, insofar as they are objects of thought, and insofar as they are presented as a unified whole . We have seen above that this characteristic of objects is defining; it is precisely because the object is a unified whole that it can become an object of presentation (see further below) . Objects are everything that is thinkable: they are characterised by their cognoscibilitas, where a round square is as much thinkable as any possible object . The summum genus contains “all categories of what is conceivable” (Twardowski 1894b, p . 34) . Or, to make the same point from a linguistic point of view, “everything which is designated [by a noun] is an object” (idem) . Finally, the Latin term “ens” applies to all objects in the sense that “entity” is the most general term synonymous with Twardowski’s use of the term “object .” Objects include real and non‑real entities . Twardowski does not give an explanation of the reality of an object (Realität; (Twardowski 1894a, p .36)), but he gives some examples of non‑real objects: possibilities, absence, a lack of something – they may be called conceptual objects (Begriffsgegenstand; idem), in contrast to real objects . Such conceptual objects may exist as well as the real objects . There are existing and non‑existing objects, in the sense that Odysseus is an object, whether he has ever existed or not . An object may thus be existent but not real . If it is true that John has no money, his lack of money is an existent, though not a real object . Objects may be events or actions, like a storm or a murder . And parts, properties and relations of objects are objects, too . Even the content of thought, a dependent aspect of a particular act of thinking, can be understood as an object, for we can think of it in a reflective act . Twardowski considers objects to be wholes of parts, and a metaphysical analysis is needed in order to distinguish the different kind of parts of the object . Whereas metaphysical analysis relates to the object, psychological analysis relates to the content (p . 67) . Both forms of analysis are variants of decompositional analysis, as Michael Beaney calls it . One start with an object as a complex whole, and one obtains through analysis the different parts and relations of the complex . Psychological analysis, analysis of the content, may reach a final point, when it obtains the elements of the contents of presentation . Twardowski follows in this sense Wundt (p .44) . Metaphysical analysis cannot give a full analysis of the object, according to Twardowski, for an object consists of an infinite amount of relations to other objects . Metaphysical analysis never reaches the object in all its detail . Although we sometimes speak of simple objects, such as God, they are only simple in a loose sense of the word, for even a simple object has an infinite amount of relations to other things, which relations are constituents of the object (p . 74, see below) . Content and Object 71

A part or constituent is everything that can be distinguished in or about the object of a presentation (p . 47) . Each object has both material and formal constituents . The material constituents of a chair are its legs, its seat and its back, but also its color, shape, weight and size . The former parts are the results of “real analysis,” which analysis results in parts that may be sepa‑ rated from their object . The colour and the shape of the chair are the result of an analysis in thought (p .47) . These are called metaphysical parts . They cannot be separated from the object in reality, but they can be separated in thought by an act of abstraction (p . 55, note) . The material constituents are of different rank: windows are constituents of houses, houses are con‑ stituents of a town, and windows are therefore parts of second order of the town . One has first to distinguish the parts of first order, the windows of a house, or the houses in a town, in order to be able to distinguish the parts of second order (p . 47) . There are two kinds of formal constituents . The primary ones are formed by the relations between the parts and the object as a whole . The secondary formal parts obtain among the constituents . The primary formal constitu‑ ents include the having of the parts by the whole, the forming of at whole by the parts, relations of coexistence between the whole and its parts, and relations of succession, if the object is, for example, a motion or a year . All these relations may be terms of new relations, the latter being relations of second degree . According to Twardowski, every complex object can be understood as a function of its material parts . The formula for an object is for example f1 (P1, P2, P3), where any fn is a formal, dependent constituent, and Pn is a material constituent (p . 55) . Because there are different ways in which the constituents are combined into a complex object, f may be different for different objects . It should be noted that Twardowski’s aim is decomposi‑ tional analysis of the object, not conceptual analysis of a judgeable content . Twardowski’s notion of function thus differs from Frege’s . It is rather on a pair with Ehrenfels’ notion of Gestaltquality, or the notion of form of unity in G .F . Stout, the dependent part that unifies the parts into a whole . It is the way the material parts are contained in the whole, as Twardowski puts it (die Art ihres Enthaltenseins im Ganzen; idem) . I come back to this below . Metaphysical parts such as the red of the rose are sometimes called properties, but Twardowski prefers to use the term “property” (Eigenschaft) in a different sense . The red of the object is a metaphysical part, but being red, having red as a part, is a property or a property relation, as Twardowski (relation) sometimes says in order to distinguish it from the metaphysical part . The having of regiments and soldiers is a property of the army, although no one would call the soldiers a property of the army (p . 56) . It should be noted that for Twardowski all metaphysical parts and property relations 72 Maria van der Schaar are particular entities, and the property relations are themselves parts of the object; they are its formal constituents . For Twardowski an individual is a complex of parts, both dependent and independent parts, and material and formal parts . What makes the different parts constituents of one and the same object? Is an object more than the sum of its parts and relations? Does Twardowski acknowledge a special form of unity, a Gestaltquality, in order to account for the unity of a complex object? After having asserted that the properties of the form having P are themselves constituents of the complex object, Twardowski acknowledges that these properties are thereby had by the object, thus obtaining the having of having P as a constituent, which argument would then endlessly go on . He concludes: “Perhaps it is just this infinite nesting of primary formal constituents which contain the key to the answer to the question concerning the nature of the relation which holds the parts together in a whole” (p .56) . Such an infinite regress does not have the explanatory power that is needed . The unity of the complex object is rather presupposed by the relations of the different parts to one and the same object . Twardowski has something to say about the unity of the object at the end of his mereological investigations . Every object is one, a unified whole . This is the reason that the Scholastics called the object unum . Twardowski takes refuge in a psychological analysis to explain the point . “Everything that is presented as an object, no matter how complex it is, is presented as a unified whole” (Twardowski 1894b, p . 86) . This is immediately followed by a mereological explanation: “Its parts are united into this unified whole through property relations which have a common term on one side” (idem) . The fact that the object is one is a primitive idea that cannot be explained further, but is rather used to explain the property relations . The fact that the object is a unified whole cannot be reduced to property relations and formal constituents . By means of all these mereological notions, Twardowski is able to clar‑ ify traditional philosophical terms such as “characteristic” and “essence” . Twardowski uses the idea of property relations to explain what the essence of an object is: The totality of property relations from which one can derive, because of causal de‑ pendency, all other property relations of an object is called the essence [Wesen] of the object (p . 57) . Twardowski gives a mathematical example, where the “causal depen‑ dency” should be taken in a broad sense . The property of a triangle that is expressed by the Pythagorean theorem depends on the fact that the triangular object has three straight sides and one right angle . Because the latter is not derived from any other property, it is an essential property of the object . Such essential properties can be apprehended only if the explanatory or‑ Content and Object 73 der is determined . As we need scientific research to determine this order, Twardowski’s notion of essence is not metaphysical in the traditional sense . It can be understood as a critique of the traditional explanation of essence in terms of genus and specific difference . According to Twardowski, the term “characteristic” is ambiguously used by philosophers for both a part of the content and a part of the object of presentation . The best way to understand this notion is to say that charac‑ teristics are constituents of the object of presentations, not of their content . As for Frege (Frege 1884, §53), for Twardowski, characteristics are prop‑ erties of things . There also seems to be an important difference between the two philosophers . For Twardowski, a characteristic is not a constituent of a concept, as a concept is for Twardowski an act of presentation (see the next section), and he explicitly denies characteristics to be on the level of acts of presentations and their contents . Frege asserts in the section just mentioned of the Grundlagen der Arithmetik that characteristics constitute the concept . The difference can be explained in terms of a different inter‑ pretation of what concepts are . For Frege, concepts are objective properties or functions, whereas for Twardowski they are psychological entities . There is a sense, though, in which Twardowski relates characteristics to concepts in his explanation of characteristics: “What corresponds to the concept in the thing are the characteristics of it” (Twardowski 1894b, p . 80) . Part of a Brentanian mereology is the explanation of the concepts mutual (in)separability and one‑sided separability . For Brentano, two material con‑ stituents are mutually separable in case they “can also exist by themselves, separated from the whole whose parts they are” (p . 48) . Two constituents are one‑sided separable if and only if the one depends for its existence on the other, but not the other way round . And two constituents are mutually inseparable if and only if they depend for their existence mutually on each other . According to Twardowski, this explanation is problematic, because it presupposes that the parts exist (p . 49) . In the phenomenological approach to metaphysics that Twardowski endorses, no presuppositions of existence should be made . Meinong takes up the point in 1904, when he endorses the thesis that questions of existence do not concern the object . “Being and non being are external to the object” (Sein wie Nichtsein [ist] dem Gegenstande gleich äusserlich) . This thesis is called der Satz vom Aussersein des reinen Gegenstandes (Meinong 1904, p .494) . The pure object stands beyond being and non‑being (der reine Gegenstand stehe jenseits von Sein und Nicht‑sein; idem) . According to Twardowski, one should give an explanation of (in) separable parts in terms of conceivability, which means that the distinction concerns the constituents of contents alone 20.

20 Als Gegenstand kann Roth ganz wol durch eine Vorstellung vorgestellt werden, ohne dass gleichzeitig die Ausdehnung als Gegenstand durch dieselbe Vorstellung vorgestellt würde, 74 Maria van der Schaar

According to (Twardowski 1894a, p .65), parts are mutually separable precisely if each part can be conceived of without conceiving the other . The contents of a presentation of the individual pages and of the cover of a book are mutually separable, as they can be presented independently of each other . Parts are mutually inseparable precisely if these parts cannot be conceived independently of each other; each of the parts can be distin‑ guished by the mind alone . The constituents extended and coloured are mu‑ tually inseparable, because one cannot have a colour presentation without having a spatial presentation, and conversely . In the example, the relation of dependency is not between a particular colour presentation, such as red, and a particular spatial presentation, for the same spatial presentation may be coloured differently, but it has to have some colour . It is in this sense an example of generic dependency, although Twardowski does not put it that way . Parts are one‑sidedly separable precisely if one can be conceived without the other, but not the other way round . One cannot have a red pre‑ sentation unless one has a colour presentation . But the presentation of colour is independent of that of red .

General Objects If we speak about the triangle in general and assert that the sum of the in‑ ner angles of the triangle as such is 180o, we do not merely make an asser‑ tion about this triangle and the other triangles we have become acquainted with, for we make a universal claim . A mathematical proof concerns any triangle, and not merely the particular triangles one may think of . We take an arbitrary triangle, and give a proof for that object, and if the triangle is rightly chosen, the proof counts for any triangle . What can we say about such arbitrary objects? According to Kit Fine, the arbitrary triangle has “those properties common to the individual objects in its range” (Fine 1985, p . 5), but it has none of the properties that belong only to some triangles . So, the arbitrary triangle has the common property mentioned above, but, as Locke puts it, this triangle “must be neither Oblique, nor Rectangle, neither Equilateral, Equicrural, nor Scalenon” (Locke 1690, p . 596) . The arbitrary natural number is neither odd nor even, although it can be said to have the disjunctive property of being either odd or even, for this property is com‑ mon to all natural numbers . The arbitrary number is not identical with an individual number, for the latter must have the property of being odd, or the property of being even . So one has to acknowledge that something may be odd or even, without having either of the two properties . It is precisely und umgekehrt (Twardowski 1894a, p . 66) . As objects, red and a certain extension can be presented independently of each other; as contents, red and extension are, in a generic way, mutually dependent, though (idem) . Content and Object 75 for this reason that an arbitrary number does not belong to the category of numbers, and that an arbitrary triangle does not belong to the category of triangles . An arbitrary triangle is not a triangle, although it should be, given that all triangles share the property of being a triangle . So, we should be careful in our formulations . Claiming that the arbitrary number is odd or even amounts to the claim that any of the individual numbers falling in its range has the property of being odd or the property of being even . In the same way, saying that the arbitrary triangle is a triangle means that every object falling in its range has the property of being a triangle . Some have understood the arbitrary triangle as a triangle, although an incomplete one . On this account, the arbitrary triangle is a triangle, although a defective one . An arbitrary triangle is just another triangle, and this leads to “the absurd conclusion that one might count with arbitrary numbers or have tea with an arbitrary man ”. (Fine 1985, p .8) . The arbitrary natural number has a range consisting of all individual natural numbers, but is not itself one of these individuals . The arbitrary triangle does not belong to the category of triangles . At the same time, we seem to attribute properties that belong to individual triangles to the arbitrary triangle, when we say that the triangle has three angles, and that the sum of its inner angles is 180o . Such a sentence cannot be taken at face value, for a general object or a universal has no angles . Because the general triangle is not a triangle, we do not have an intuitive presentation of the general triangle . One may assert that there are arbitrary objects, without claiming that they exist, just as one may claim that there are numbers without claiming that they exist . One may thus have a name for such an arbitrary object . We use the names “the triangle” and “the whale,” when we assert that the triangle has three angles, or that the whale is a mammal; these names are names of arbitrary objects . If one assumes that there are arbitrary objects which do not exist, one has to give in the end an explanation of these objects in terms of objects that one takes to exist, or that one has already assumed for other reasons . Because Twardowski denies the existence of general objects (the world contains only particulars), it makes sense to compare his account with that of Locke . According to Locke, as far as ontology is concerned there are only particulars, and the senses give us only particular ideas . We can form general ideas, though, by means of an act of abstraction . If we start with a particular idea of this triangle or of a particular man, we may abstract from it the ideas of time, of place and other ideas . We thus obtain isolated ideas which can be used to represent a plurality of individuals that agree with each other: “Thus the same Colour being observed to day in Chalk or Snow, which the Mind yesterday received from Milk, it considers that Ap‑ pearance alone, makes it a representative of all of that kind” (Locke 1690, 76 Maria van der Schaar

II: p . 159) . These isolated ideas are general; they are applicable to many things . There is nothing general, though, in the world, and general ideas are merely fictions of the mind (Locke 1690, IV, p . 596) . Locke acknowledges that these ideas are “marks of our imperfection”: “an Idea wherein some parts of several different and inconsistent Ideas are put together” (idem) . From a Fregean point of view, we cannot speak about the concept of a triangle; the concept of a triangle has only a predicative function . When we assert that the triangle is a geometric figure, we assert something about the relation between the concepts being a triangle and being a geometric figure, namely that every object that falls under the concept of triangle also falls under the concept of geometric figure . We can make the same point in terms of properties: every object that has the property of being a triangle has the property of being a geometric figure . We have to pay a price for this position, for in natural language we do speak about the triangle in general, whereas Frege has no means to speak about the concept of a triangle or that of a horse . Frege’s position is not a good starting point to understand Twardowski, and it is for this reason that I started this section with Locke’s account of general ideas and the account of arbitrary objects in the work of Kit Fine . None of the more traditional positions with respect to universals is defend‑ ed by Twardowski . Twardowski is neither a logical realist, nor a nominalist or conceptualist . When we think of the triangle in general, the object of our thought is not a sum of individual triangles, for we make a universal claim . When we think of the triangle in general, we do not think of a plurality of individual triangles, as the nominalist claims; we think of one object . The general object we think of when we think of the triangle is also not a Pla‑ tonic entity . There are no Platonic entities like the triangle in general, nor is the triangle a singular species, as Husserl claims in the second of the Logical Investigations . Essential to Twardowski’s explanation of the general object is the notion of metaphysical part, the constituents referred to in the following explana‑ tion of the general object . [W]hat is presented through a general presentation is a group of constituents [meta‑ physical parts] which are common to several objects . This group of constituents is presented as a whole that belongs together; this is the object of the general presenta‑ tion (Twardowski 1894b, p . 100) . Each red object has a metaphysical part which makes it into a red object, and this red moment is an irrepeatable, dependent part of the object, which can only be apprehended by means of an act of abstraction . Due to our ability to abstract, this metaphysical part can be thought of independently of any other object . The triangular aspect of an individual can also be ap‑ Content and Object 77 prehended without any apprehension of its size, and its particular angles . The general object is a group of such particular aspects which are common to several objects (Twardowski 1894b, p . 100) . These metaphysical parts, the particular triangularity aspects or particular properties of each triangle, constitute the general object of a triangle . This group of common particular aspects can be apprehended as a unified whole, as one object, insofar as the aspects of the objects of individual presentations stand in a relation of equality to each other, and thereby form a group . Husserl argues that such a relation of equality has to be based on a relation of identity, the identity of the species, which has the individual objects as its instances, but this is not needed on Twardowski’s account . If the relation of equality would hold among the different individual triangles, we would have to explain in what sense these individuals are equal, in which case we would have to refer to an identical species (Husserl 1901, p . 118) . For Twardowski, though, the relation of equality holds between the different metaphysical parts of the triangles, each being a different, though exactly similar triangular aspect . Each of these parts already constitute the triangularity of the different tri‑ angles, and we are not in need of an explanation in what sense these aspects are equal . These general objects do not exist, according to Twardowski, but they are objects nonetheless, constituted by the different equal meta‑ physical parts . Their psychological unity is explained by the fact that these constituents are thought of as a unified whole . Their metaphysical unity is explained by the relations of similarity between the different metaphysical parts . There is thus a sense in which there are universals, for Twardowski, but ontologically they can be fully explained in terms of the metaphysical parts of individual objects, including those of possible objects . Twardowski’s position is in agreement with the position defended by the British psychologist and philosopher G F. . Stout . Stout explains universals as unities of similar abstract particulars . These are what Twardowski has called metaphysical parts: the red of this rose, or the triangularity of this triangle . The universal is nothing but a group of particular properties or characters that have a certain unity, by which we are able to understand the universal as one object (Stout 1921) . Like Twardowski, Stout acknowledg‑ es non‑existing possibilities as objects, which means that the universal is a unity of both actual and possible characters . This means that the universal or general object does not change if new objects come into existence . It is thus possible to acknowledge the universal Being a Centaur, the general object of centaur, although there are no centaurs . Although these theories of universals or general objects have a psychological origin, there is no reason to classify this account of general objects as psychologistic . Twardowski gives a linguistic and a psychological explanation why the general object is often not acknowledged . We use the same name, “the trian‑ 78 Maria van der Schaar gle,” for both the individual and the general object . In languages that have the definite article, the form “the triangle” is primarily used for the general object, whereas the individual object is designated by the substantive to‑ gether with a pronoun or demonstrative, “this triangle,” or by a substantive together with a subordinate clause, “the triangle that we see in front of us” (Twardowski 1894b, pp . 101-102) . The general object may also not be acknowledged because there is a special psychological relation between the presentation of a general ob‑ ject and the presentation of an individual object . According to Twardowski, we present the general object by means of a presentation of the individual object . Because Twardowski does not distinguish between indirect presen‑ tations and presentations by means of auxiliary presentations, he explains the presentation of general object in both terms . As we have concluded with Marty that there is a distinction between indirect presentations and presenta‑ tions by means of auxiliary presentations, we should make a decision here . Is, according to Twardowski’s theory, a presentation of a general object an indirect presentation, or is it rather a presentation by means of an auxiliary presentation . This is to be decided by the question whether the presentation we use is essential to the content of the general presentation, or not . The auxiliary presentation has only an accidental, psychological relation to the content of the presentation, whereas the indirect presentation has a concep‑ tual, a logical and semantic, relation to the intuitive content . Twardowski claims that the presentation of a particular triangle is not essential to the presentation of a general object: [O]ne can well imagine a more perfect mental organization than the human being’s which would be capable of thinking of general objects without recourse to presenta‑ tions of the corresponding individual objects (p . 104) . It is for this reason that the presentations of the individual objects do not essentially belong to the general presentation . Furthermore, if it would be an indirect presentation, we would be in need of the direct or canonical presentation of the general object, to which the indirect presentation can in principle be brought back . But we never have such a canonical presentation, and we have no idea what it would look like, on Twardowski’s account, given that we have to assume that only beings with a different form of intuition can have such a canonical presentation of the general object . We can understand now what the explanation is of the fact that we use the same term for both the general and the individual object: the word “triangle” is first used for an individual triangle, and the same name is extended to the general object, because we use a presentation of an individual triangle as auxiliary presentation for the presentation of the general object . In order for the individual presentation of a particular triangle to function as auxiliary Content and Object 79 presentation for the presentation of the triangle in general, the presentation needs to be accompanied by the presentation of a certain judgement, which concerns the particular size and the angles of the individual triangle . In such a presented judgement it is denied that the general object has the individual properties that characterize only the presented particular triangle . According to Twardowski, we do not actually make these judgements – “the judgements are only presented in the modifying sense of the word” (p . 103) . (See the next chapter for an explanation of represented judgements .) Twardowski does not say much about the content of the general pre‑ sentation, but one thing is clear: the content of a general presentation is different from the content of an individual presentation . Twardowski does not completely deny Berkeley’s thesis that in the case of a general idea of a triangle we always have an idea of an individual triangle that represents all other triangles . For Twardowski, the idea of an individual triangle has only a psychological, accidental relation to the general idea . The presentation of an individual triangle plays a role in the explanation of how it is possible for us to apprehend the general object, but it is neither part of the explanation of what the general object is, nor part of the explanation what the content of a general presentation is . Because Twardowski identifies the meaning of our singular terms with the content of the corresponding presentation, the content of the presentation of an individual triangle is not a part of the explanation of the meaning of “the triangle as such .” Contents of auxiliary presentations are not part of the meaning of “the triangle as such;” they only mediate our apprehension of the meaning . And it is for this reason that Twardowski’s account of general presentations is not psychologistic, although the way we apprehend these meanings is given in psychological terms . Twardowski’s account of generality differs in an important aspect of the one given by Locke . For Locke, there is no moment of triangularity in the particular triangle that may form the basis of a general object or idea . On both accounts, an act of abstraction is needed to account for generality, but for Locke the act of abstraction has a particular idea as its starting point, whereas for Twardowski, it is the object of a singular presentation from which we abstract the purely particular aspects . In Locke’s case the act of abstrac‑ tion results in a general idea, whereas for Twardowski the act of abstraction brings to light the property or metaphysical part, which is a dependent con‑ stituent of the particular triangle . Most important, for Twardowski, the act of abstraction needs to be supplemented by a synthesising act that brings together the different moments of triangularity of the different particular triangles . For Locke all abstract ideas are general, whereas for Twardowski there is a difference between a general presentation of the triangle and an abstract presentation of the triangularity aspect of this particular triangle . 80 Maria van der Schaar

The latter has no generality . Without the synthesising act of the different triangularity moments we do not obtain a general presentation . Twardowski’s notion of the general triangle can also be compared with the notion of an arbitrary triangle . Twardowski’s general triangle does not have the properties common to the individual objects in its range as the arbitrary triangle has it; it rather consists of these properties, that is, of the metaphysical parts of the particular triangles . In later writings, Twardowski will come back to this account of general objects in order to give an account of what Husserl has called the identical or ideal meaning (see Twardowski 1912b and the next chapter), without committing himself to more than par‑ ticulars in his ontology .

4. Images and Concepts

In 1898 Twardowski published a small book on images (anschauliche Vor‑ stellungen) and concepts in Polish . In 1903 the text of a lecture on the same topic was published in German, called “Über begriffliche Vorstellungen” . I will start with the Polish paper . Although Twardowski himself suggest “images” as the English translation of the Polish wyobrażenia, the term suggests a restriction to mental pictures, which is not what Twardowski intended . If we imagine a certain emotion, no picture is involved, whereas it is a clear example of an imagining . As the images and concepts are acts of presentation, one may also speak of imagining and conceiving . According to Twardowski, all conceiving is based on imagining . Cer‑ tain imaginings form a necessary condition for certain kinds of conceiving . For example, the fact that we have concepts for judgements, emotions and other mental phenomena implies that we have had perceptions and images of particular judgements and emotions . All imagining is complex, and can be defined as a synthesis of relatively simple parts; if we imagine a physical object, these parts are impressions or remembered impressions . There is no conscious mental construction that unifies the elements into a complex whole; psychological analysis distinguishes parts of a complex whole given to us prior to the analysis . The elements to be distinguished are in fact elements of the content of the act of imagining, which correspond to the elements of the imagined object . According to Twardowski, all imagining is “concrete insofar as it appears as a undifferentiated whole” (Twardowski 1898b, p .96) . Concreteness is to be understood as a togetherness; it is a characteristic of a complex whole of unified parts . The notion of being concrete is opposed to that of being abstract, which will be explained below . Content and Object 81

When I perceive, for example, an iron ball lying on the table in front of me, I imagine a number of features integrated into a concrete whole, but I do not simultaneously perceive all the features I imagine (p . 97) . The elements that are not given in perception, are imagined, as we do present the ball as having a back side, and a smooth structure, although we do not perceive these elements . It is for this reason that the term “to imagine” is better than “to intuit”: intuitions we have of the given alone . We may imagine some elements more vivid than others, but this does not imply that abstraction is involved . The point is rather that imaginations are “sketchy” . In opposition to Meinong, Twardowski denies that imaginings may be abstract . The point is that Twardowski has a different view on what abstraction is . For Meinong, abstraction is “bringing out certain features of an imagined object by concentrating attention on them, while neglecting other features” (p .95) . Cf . (Meinong 1889) . For Twardowski, this describes only a necessary, not a sufficient condition for abstraction . For abstraction, attention is not enough . In the essay on conceptual presentations from 1903, concepts are under‑ stood as the result of abstraction . Analytic concepts are a kind of non‑intu‑ itive presentations; they are presentation of objects set apart from a larger whole by means of analysis . They are the presentations of characteristics, properties and relations, when we ignore the objects and terms to which these properties and relations belong . For example, “a shape is brought into relief by means of the activity of abstracting” (Twardowski 1903, p .85), and the presentation of the shape is thereby an analytic concept . In order to explain what abstraction is, Twardowski introduces the idea of a represented judgement . A represented judgement is a presentation of a judgement: one thinks of the judgement, the judgement is an object of thought, but one does not pass the judgement . The concept of represented judgement will be more extensively dealt with in the next chapter, as it plays an important role in understanding the meaning of a declarative sentence . Abstraction consists of three parts: (1) an original imagination or per‑ ception as starting point; (2) an act of attention; and (3) a form of mental elimination through represented judgements . Suppose I form an abstract presentation of the shape of the desk at which I am sitting while writing these words, thereby obtaining the concept of this shape . (1) I perceive the desk with all its properties as an undistinguished unity . (2) Through an act of attention, I think of the shape of the desk . And, (3) by means of a represented judgement, I imagine that the other characteristics of the desk are not there (Twardowski 1903, p . 85) . I do not have to apprehend the other characteristics individually; these characteristics appear to consciousness in an undifferentiated manner . An important group of analytic concepts is formed by relational concepts . By means of abstraction it is possible to think 82 Maria van der Schaar of a relation without thinking of its terms . From these analytic concepts it is possible to obtain synthetic concepts, such as the presentation of a tri‑ angular shape of my desk, and negative concepts such as a mathematical point, which starts with an imagined presentation of a dot, which is thought of as having no extension at all; the represented judgement is thus a denial of extension of the dot . This is not an actual judgement, for we know that the dot must have a certain extension . It seems that synthetic concepts may either have a particular shape of the desk as its object, or a shape in gen‑ eral; they may thus be particular or general . In the 1903 paper, the concept of a shape in general is not explained; for this, the reader is referred to the former section on general objects . The notion of judgement is for Twardowski logically prior to that of con‑ cept . Although the notion of imagining is prior in the order of explanation to both the notion of concept and that of judgement, the notion of judgement is, for Twardowski, prior in the order of explanation to concepts . We need the notion of represented judgement for the explanation of concepts, and we need the notion of judgement for the explanation of represented judgement . Compared to the Fregean account of concepts as functions obtained through analysis of a judgemental content, Twardowski seems to give a psy‑ chological analysis of concepts . Frege’s account of concepts is certainly not unproblematic, as it cannot account for the fact that we may speak of the concept horse, without turning what is called a “concept” into an object . It is clear, though, what the advantage is of his logical account of concepts . It is possible to distinguish different kinds of concepts in the same judgeable content, as he puts it in the Begriffsschrift, depending on what inferences we need to make . We can only apprehend concepts by means of an analysis of the judgeable content, but this does not make the concept a result of a psy‑ chological activity . Frege is not in need of acts of attention or abstraction in his explanation of concepts . As Twardowski’s account necessarily brings in these notions, his account of concepts is psychological . By combining an account of concepts with an account of imagination, it seems impossible to escape a psychological approach . Perhaps, though, we need both a logical and a psychological account of concepts . The latter is addressing the ques‑ tion how concepts may arise in us, rather than trying to find an answer to the question what a concept is . If that is true, Twardowski may be credited for the idea that the whole, the imagined complex, is genetically prior to the part, the concept, and that from a psychological point of view the imagined complex is thus genetically prior to the concept . Twardowski introduces the idea of a logical concept as a concept having a well‑defined content (p . 89), primarily to be explained in terms of repre‑ sented judgements . The logical distinction between positive and negative concepts derives from the distinction between affirmative judgement and Content and Object 83 denial, and has its origin in the fact that represented judgements are essen‑ tially part of the explanation of concepts . Twardowski discusses the theories of contemporaries who understand concepts to be judgements or products of judgement . Although it is true that represented judgements can be under‑ stood as potential judgements in a certain sense of that term (see the next chapter), concepts are a kind of presentation, not a kind of judgement . As soon as Twardowski makes the distinction between actions and products, he understands concepts to be the products of certain acts of presentation (p . 95, note) . As the concept as mental product is isolated by means of an act of abstraction, the question arises to what extent Twardowski’s distinction between action and product will enable him to overcome psychologism . 4.

JUDGEMENT AND MEANING. ON ACTIONS AND PRODUCTS

1. The Historical Background of Twardowski’s Theory of Judgement

The reform of logic is an important aim in the nineteenth century, and a new account of judgement is to play a part in it . Traditionally, judgements are analysed into form and matter, where the matter consists of two general terms, one of them being the subject, the other the predicate . There are four different judgemental forms, each connecting the two terms in its own way: the predicate may be said of some things to which the subject term applies, or it may be said of all those things, resulting in, respectively, a particular and a general judgement; the predicate may be affirmed of the things to which the subject term applies, or it may be denied of such things, resulting in, respectively, an affirmative and a negative judgement . An affirmative judgement may thus have the same matter as a negative judgement, and a general judgement may have the same matter as a particular judgement . As the matter consists of two independent terms, the judgemental form is needed to unify the two, and to understand them as subject and predicate of the judgement . There are several problems with the traditional account of judgement . The general point is that it seems not to be able to account for the uniqueness of judgement . The notion of judgemental form cannot be used to account for the uniqueness of judgement, for the unasserted antecedent of a hypo‑ thetical judgement If no Greek is a God, then all Greek are mortal and the judgement No Greek is a god have the same judgemental form, while the antecedent is not judged . If the judgemental form cannot account for the uniqueness of judgement, how is one to account for it? Although the ques‑ tion is already raised by Hume, it is to play a central role in the reform of logic in the nineteenth century . Traditionally, ideas and judgements are classified as belonging to one class, thought, which is generally opposed to the will and emotions (Figure Judgement and Meaning 85

1) . Judgement is thus to be defined in terms of its genus thought; judgement is a thought with a special matter and form, as we have seen above .

Figure 1. Traditional classification of mental phenomena

If judgement is unique and a kind of its own, if it is sui generis, it cannot be understood, though, as a special form of thought . Thought and judgement are not to be related as genus and species; mere thought and judgement should rather be two of the basic ways in which we can be mentally related to something . A further problem for the traditional account of judgement consists in the fact that not all judgemental matter seems to be accounted for . As all judgements are supposed to have a complex matter, involving both subject and predicate, judgements with impersonal verbs seem to be excluded (cf . II 2. ). . What would the logical subject be if we judge that it rains? It can also be doubted that both a subject and a predicate are involved in existen‑ tial judgements . It is clear what the logical subject and predicate are of the judgement some tigers are living in India (tiger is the subject and living in India is the predicate), but what would be the subject and predicate of the judgement there are tigers? Brentano’s idea that all judgements are existential provides a solution to some of these problems, and Twardowski is to follow him in this . In the first place, Brentano understands that the uniqueness of judgement cannot be found in the judgemental content; the uniqueness of judgement lies in the act of judgement . The distinction between act and content is thus essen‑ tial for the step towards an improvement of the notion of judgement, and thereby of logic 1. For Brentano, every judgement is the affirmation or de‑ nial of the existence of what is presented . Everything that can be presented, 1 In his dissertation on Descartes, Twardowski points to the fact that Arnauld, by making a distinction between perceptio and idea, already makes a distinction between the act of pre‑ sentation and the content of presentation (Twardowski 1892, p . 6) . 86 Maria van der Schaar can be judged . In the judgement There are tigers, tiger is the object of the judgement, whereas acknowledging its existence is the act of judgement . When we judge that tigers exist, we do not predicate existence of tigers . We rather relate the concept or characteristic (Merkmal) of being a tiger to the world, thereby affirming the existence of things that have the charac‑ teristic of being a tiger . The object of judgement may thus be simple, and there are two kinds of judgemental acts: particular affirmations, and general denials . According to Brentano, all affirmations affirm that some A exist, where A is a concept; all denials deny that A exist, and denials are thereby general . Denying that unicorns exist is a claim that no object is a unicorn . There is a problem, though, with Brentano’s proposal, for if we put the word “If” in front of “tigers exist”, we see that we can delete the judge‑ mental force from the existential form . The possibility to merely entertain the thought that tigers exist, without judging, is not accounted for . For Bren‑ tano, existence comes in only at the level of judgement . All we do when we entertain the thought that there are tigers, on Brentano’s account, is to think of the concept (Merkmal) of being a tiger, but that is not the same as thinking that tigers exist . We see a similar problem when it is claimed that the judgemental form is P is true, where P is a proposition, for we may have this form without judgemental force, as in “If it is true that it snows, then we should stay in .” Frege’s position in the Begriffsschrift is still ambiguous, for he acknowledges, on the one hand, a special judgemental sign showing that a judgement has been made . On the other hand, he seems to claim that “… is a fact” (ist eine Thatsache; cf . (Frege 1879, §3)) may be understood as a common predicate for all judgements . The predicate, though, is not a unique characteristic of judgements, as we do say: “If it is a fact that it still snows, we have to stay in ”. A judgemental form cannot be used to capture the uniqueness of the judgement as act . When Frege wrote “Meine grundlegenden logische Einsichten” in 1915, he formulated clearly that the judgemental force cannot be represented by the word “true .”2 The importance of the judgemental force for logic is stressed by him in the same piece: the essence of logic is contained in the assertive force .3 The essence of judgement, and thereby of logic, can neither lie in the existential form nor in the form P is true . We have to distinguish the question what the judgemental form is from the question what character‑ izes the judgemental force . We may say that all judgements have the form P is true or A exists / A does not exist, depending on one’s theory, while acknowledging that judgemental force is to be explained independently

2 [D]ie Behauptung [liegt] nicht in dem Worte ‘wahr’, sondern in der behauptenden Kraft, mit der der Satz ausgesprochen wird (Frege 1915, p . 271) . 3 Dasjenige nun, was den Hinweis auf das Wesen der Logik am deutlichsten enthält, ist die behauptende Kraft, mit der ein Gedanke ausgesprochen wird (p . 272) . Judgement and Meaning 87 of this form . We have to explain the judgemental force in terms of the use we make of a sentence . In Frege’s writings the judgemental force plays an important role . In the Begriffsschrift, all premises and conclusions are pre‑ ceded by the judgemental stroke, which is a sign of judgemental force . For Frege, the sign is essential to logic, because he considers logic to be the most universal science, built upon axioms in the traditional sense of known judgements . The premises and conclusions in an ideography are, for him, judgements made, and they purport to be knowledge, although they may turn out to be incorrect, as in the case of the fifth axiom . In such a case, one has to withdraw one’s assertion and one has to delete the judgemental stroke from the ideography . Modern logicians generally agree with the point of critique Wittgenstein makes in the Tractatus . Frege’s judgemental stroke is logically meaningless; it only shows that the author, Frege, holds the judgemental content to be true (Tractatus 4 .442) . If Wittgenstein’s critique were right, there would be no point in withdrawing one’s assertion . For, the sign would merely show that Frege at that time had judged the fifth axiom to be true, and that is still a fact . I do think, though, that Frege captured an important point with the judgemental stroke . If logic is concerned with the premises and conclusions we make, it is in need of a sign for judgemental force . This need not be un‑ derstood in any subjective sense, for as soon as one detects a mistake, one understands that one has to withdraw one’s assertion, and that one has to cancel the judgemental sign . It is true, though, that this brings the element of a judging agent into the system . Any logical system that takes into ac‑ count the judging agent has to explain how it is to prevent a subjectivistic form of psychologism . For Brentano and Twardowski, the judgemental act plays a central role in logic, and they have thus to take this problem into account . One of the driving forces behind Twardowski’s developments can be understood as trying to find an answer to the critique of psychologism that was so eloquently put forward in Husserl’s Prolegomena of the Logical Investigations . Another problem for Brentano’s account of judgement is caused by the fact that it is non‑propositional . When a judgement is made, we claim some‑ thing to be true . For example, we may claim that snow is white is true, in which case there is a proposition, that snow is white, which is claimed to be true . So, there is at least a linguistic reason for acknowledging propo‑ sitions . Brentano’s analysis of the judgement “Snow is white” seems to be artificial . For him, judging that snow is white is denying the existence of non‑white snow . Is it true that all judgements are affirmations or denials of existence, as Brentano holds? What existence claim is involved when we judge that 7 + 5 = 12? Generalising the problem, we may say that Brentano does not distinguish between a judgeable and a non‑judgeable content . For 88 Maria van der Schaar him, everything that can be presented, can be judged: a house, a man, and a unicorn, they can all be judged . If Brentano answers that a house can be judged, because its existence can be acknowledged, his account seems to make the act side of the judgement too complex, involving both the judge‑ mental force and the existential form . Furthermore, because Brentano does not acknowledge propositions, he is not able to account for propositional negation . Not all forms of negation can be explained in terms of an act of denial, for negations may appear where no judgemental act is involved, as in questions, supposals and hypothetical judgements . The ambiguity in Brentano’s account of intentionality (the fact that our mental acts are related to something that is called both a “content” and an “object”), has an effect on the question what the object of judgement is . If we acknowledge the existence of tigers, the existence of the concept tiger is acknowledged . The act of presentation provides us with an object, the concept or characteristic of being tiger, whose existence is acknowledged as soon as the judgement is made . If we acknowledge the existence of Socrates something else seems to be involved . For Brentano, judgements concerning a specific object form an important class . In the case of inner perception, one acknowledges the existence of this individual act of seeing, hearing or feeling . These judgements do not have the form There are tigers, for they are singular judgements . We acknowledge the existence of this individual act . The act of presentation that is the foundation for this judgement seems to provide us with the judgemental object itself, at least in case we have a presentation of our own mental acts . The judgement would thus have the form “This act exists .” The judgemental form of “This act exists” seems to be different, though, from the judgemental form “Tigers exist” . One may say that the judgemental content is here an individual concept this seeing of red, whose existence is acknowledged, but his cannot be Brentano’s posi‑ tion, for he is committed to the fact that the object itself, this seeing of red, is presented . It is precisely for this reason that inner perception is infalli‑ ble . In the judgement “Tiger exist”, we say something about the concept of tiger, that it is not empty, that there are in the world objects falling under the concept tiger . In the judgement “This act exists”, we acknowledge the existence of an individual object, this act . Brentano does not seem to be able to account for this logical distinction, which a modern logician would call the distinction between a second and a first order predicate of existence, although he is not alone in this, as Frege simply denies that there is a first order predicate of existence . Apart from this criticism, we may notice some important points of Bren‑ tano’s theory of judgement, which have made it possible to see the unique‑ ness of the judgemental act . Because he does not give an explanation of judgement in terms of subject and predicate, Brentano is able to see that the Judgement and Meaning 89 uniqueness of judgement has to lie somewhere else . By making a distinc‑ tion between act and content, it is possible for Brentano to understand that presentation and judgement are distinguished by a different mode in which one may be related to the same content 4. The act of thinking of something is an act of presentation, which is a necessary foundation for the act of judgement . This means that the act of judgement itself is not a special case of thinking, but of a unique kind, and that we are in need of a new classifi‑ cation of mental phenomena . For Brentano, there are three classes of such phenomena, each on the same conceptual level: presentations, judgements and emotional attitudes (figure 2) . Because judgement thus forms a kind of its own, we cannot give a definition of it . It is a unique way in which we are directed to an object 5. Philosophers sometimes say that judgement involves thought: on the traditional account this means that judgement is a special kind of thought; for Brentano, this can only mean that the act of judgement is founded on an act of thought or presentation 6. It is crucial to Brentano’s classification that there is no class of mental phenomena in be‑ tween presentations and judgements, and Twardowski follows him in this . What Meinong calls “assumptions” do not form a separate category for them, and must therefore fall under the category of presentations .

Figure 2. Brentano’s classification of mental phenomena / mental acts:

4 Hume defends a similar combination of theses: the distinction between subject and predicate is for him not essential to judgement or belief; beliefs are essentially existential in character; and the uniqueness of the judgement or belief is to be found in “the manner, in which we conceive [the object]” (Hume 1739, 1 .3 .7, p . 66) . Cf . II .1 . I come back to the ambiguous use of the term “belief” below . In Hume there is also an ambiguity in his use of the term “idea”, which may stand for what is believed, or for a distinct mode of conceiving, an ambiguity that is overcome by Brentano’s account of intentionality, where a clear distinction is made between the act of presentation and the presented content . 5 It is for this reason that Brentano’s student F . Hillebrand speaks of the “idiogenetic” account of judgement, the judgement is ίδιογενής: “peculiar in kind” (Liddell & Scott). Cf. (Twardow‑ ski 1894a), p . 28, and (1907), pp . 99-101 . 6 Philosophers sometimes oscillate between the two pictures, as we can see for example in (Crane 2013, p . 6), where it is argued that judgement is a special kind of thought (the traditional picture), because it depends on thought (Brentano’s picture) . 90 Maria van der Schaar

Twardowski’s classification of mental phenomena is a variant of Brentano’s threefold classification, although he acknowledges manifestations of the will as a separate group besides feelings (Twardowski 1898a, p .69) . Twardowski brings images, including sensory impressions, and concepts, including ideas, under the category of presentation, and perception and inference under the category of judgement (idem) . Crucial to Twardowski is that judgement and presentation are not to be taken under one category of thought . According to him, the term “thought” is ambiguous: it may stand for the mere thought of someone, in which case we have a presentation, or it may refer to an act of judgement . When one says: “I think that he is a trustworthy person,” we make a judgement (p .70) . Judgement is sui generis, Twardowski will argue, and in 1898 he understands the unique distinguishing mark of judgement to be the element of conviction (Polish: przekonanie; idem, pp . 67-68) . He thus confuses the act of judgement with conviction or belief (see below), as so many modern philosophers still do . Twardowski does not realize here that conviction has degrees and is a state of the mind rather than an act . In the later paper on “Actions and Products” Twardowski understands that the question of the essence of judgement concerns the act of judgement, an all or nothing affair in analogy with the act of assertion . Brentano’s distinction between act and content has been of importance for overcoming psychologism in logic . His students may now argue that logic is concerned with an objective content, not with a subjective act . If one makes a further distinction between content and object, as Twardowski has done, one may argue that the objectivity of science and logic is founded on the side of the object, as Russell and Meinong have done, or that it is founded on the side of the content, as we see it in Frege and Husserl . Where does Twardowski stand in 1894, and where does he stand in his later writings? If one argues with Russell that the content is a psychological notion, irrelevant to logic, the objectivity of logic is to be founded on the object . For Russell, the object of judgement is a complex consisting of the things we judge about . It has characteristics of both propositions and states of af‑ fairs; the two notions are not distinguished . The proposition is truthbearer and truemaker at the same time . For Meinong, as for Russell, the Objektive or proposition is the object of judgement, both a truemaker and a bearer of truth . Meinong’s objective differs, though, from Russell’s proposition as complex whole, as the objective is an ideal entity . The object we judge about is for Meinong not a part of the complex whole or proposition, as Russell claimed . For Meinong, it has a foundational relation to the objective, as the latter is dependent on the object we judge about . According to Meinong, the objective as ideal entity can only be apprehended by an intellectual, synthe‑ sising act such as judging . It is precisely for this reason that Meinong was in need of a new notion of intellectual act, in which the objective may be Judgement and Meaning 91 apprehended without being judged: the act of assumption . Because Russell’s proposition is nothing but a complex object, no special intellectual act is required to apprehend it . For Russell, the proposition can be apprehended by a mere act of presentation . The logical realism of Russell and Meinong differs in an important sense from the logical realism defended by Bolzano, Frege and Husserl . For the latter, the objectivity of logic, semantics and science is founded not on the object of judgement, but on the judgemental content . For Bolzano and Frege, the parts of the bearer of truth (der Satz an sich, der Gedanke) are Vorstellungen an sich or senses, not the objects we judge about . Husserl, who acknowledges both an object and a content of judgement, understands the first as a state of affairs (Sachverhalt), while the latter is an ideal mean‑ ing . The difference between the two types of theory also shows a difference in semantics . The objectivity of meaning may be accounted for in terms of the objectivity of what is referred to, or in terms of the objectivity of sens‑ es, or ideal meanings, as in Frege, Bolzano or Husserl . Soon after 1894, Twardowski will deviate from Brentano’s non‑propositional account of judgement . Is Twardowski’s account of judgement and his semantics on the side of Meinong and Russell, or is it rather on the side of Husserl and Frege?

2. Some Conceptual Distinctions

In order to understand what judgement is, it is necessary to distinguish be‑ tween the notion of act of judgement and that of belief . Belief is a potentiality: we have beliefs while we are asleep, but we do not make judgements when we are sleeping . If we explain the act of judgement as an occurrent belief, as is sometimes done in modern analytic philosophy, the activity is explained in terms of a potentiality, the potentiality to make the relevant judgement . According to the Aristotelian order of explanation, though, a potentiality has to be explained in terms of its actualisation . And this seems to be the proper order of explanation, for we cannot understand what a potentiality is, without understanding what counts as an actualisation . This means that we have to explain belief in terms of the act of judgement . The Aristotelian order of explanation is also in accordance with a pragmatic point of view, in which dispositional notions such as beliefs are to be explained in terms of their actualisations, here the act of judgement, or if one prefers a lin‑ guistic turn, the act of assertion . When Brentano follows the Anglo‑Saxon tradition in their use of the term “belief” (Brentano 1874, II .48), there is a danger to neglect the distinction between the act of judgement and the state of belief . Brentano, though, prefers the term Urteil to the term Glaube . Although the term Urteil has its own ambiguities, as we will see below, its 92 Maria van der Schaar primary meaning is that of act of judgement . It thus seems that Brentano is asking the right question: What is the act of judgement? If the question were: “What is belief?”, we would have to disambig‑ uate the term “belief;” besides its meaning of capacity to judge, it may also mean conviction, faith and opinion . In the explanation of judgement it is especially of importance to distinguish the act of judgement from the state of conviction . Whereas the former is, like assertion, an all or nothing affair, the latter has degrees . As S .Witasek (1908, p . 79) puts it, we have to distinguish the moment of conviction from the moment of affirmation and negation . Hume’s term “belief” seems to stand for conviction, when he characterises “belief” in terms of a certain vividness of an idea, and when he understands the uniqueness of belief to consist in a certain feeling, for in both cases we may speak of degrees . Given that Brentano explains judge‑ ment as an acknowledgement of existence, it seems that he has chosen for an explanation of the act of judgement, for one cannot speak of a certain degree of acknowledging existence . Something exists or does not exist . Just as there is no third between true and false, there is no middle between being and non‑being 7. One may wonder whether another ambiguity of the term “belief” has played a role in Brentano’s account of judgement, for one may also use the term for the notion belief in . “Belief in” means trust in, as we use it in the question “Do you believe in Jesus?”, that is, do you put your trust in Him? The term “belief” is here synonymous with “faith ”. “Believing in” may also mean being ontologically committed to. Cf . (Szabó 2003) . And it is this notion that may be relevant to Brentano’s existential account of judge‑ ment . According to Zoltán Szabó, believing in is a distinctive attitude to be distinguished from believing that so and so exists . The content of believing in does not have a propositional structure, and one believes in ghosts, witch‑ es, propositions, or other abstract objects, without claiming that one puts one’s trust in these objects . Believing in implies taking what one believes as falling under a certain concept . According to Szabó, these beliefs can be correct and one should therefore be able to give the correctness condition for a belief in, for example, witches . Such a belief is correct if and only if there are witches and these women are correctly represented by the con‑ cept witch . When Hamlet says that “There are more things in heaven and earth, Horatio, Than are dreamt of in our philosophy”,8 it is implied that he believes that there are more things in heaven and earth than are dreamt

7 Es gibt auf dem Gebiet des Urteils ein Wahr und ein Falsch. Dazwischen aber gibt es kein Mittleres, so wenig als zwischen Sein und Nichtsein (Brentano 1911, pp . 154-155) . Cf . (Bren‑ tano 1889, pp . 25-26) . 8 The Tragedy of Hamlet, Prince of Denmark, in The Norton Shakespeare, base on the Ox‑ ford Edition (New York, London: Norton, 1997), p . 1687, which follows the Folio text . The Judgement and Meaning 93 of, while there is no believing in these things on Horatio’s part, as there is no concept under which he might take these things to fall” (Szabó 2003, pp . 591ff) . Does Brentano’s analysis of judgement involve the attitude of believing in? On the one hand, his account of judgement is nonproposition‑ al, and it involves an existence claim and a concept . On the other hand, for Brentano, judgement is an act, and it may not only be an act of acknowl‑ edgement, it may also be an act of denial: one may acknowledge or deny the existence of unicorns . The denial of the existence of unicorns does not seem to have a counterpart in Szabó’s notion of believing in . Not believing in ghosts is not a form of denial, but mere absence of believing in . We have thus a reason to think that Brentano does not aim to give an analysis of the attitude of believing in . If one distinguishes between the content and object of judgemental acts, as Twardowski does, we need to distinguish the question “What is the ob‑ ject of judgement?” from the question “What is the judgemental content?” . Starting with the first question, one may say that the object of an existen‑ tial judgement, “Salzburg exists” or “A tiger exists” is the object we judge about, Salzburg, respectively, an object having the characteristic of being a tiger . As Twardowski acknowledges relational judgements as a separate category besides existential judgements, we also have to ask the question what the object of relational judgements is . This object may be understood as a relation, as a relational complex or as a state of affairs . Within the Bren‑ tanist tradition, relations are generally understood as relational moments depending for their existence on the related terms . Meinong distinguishes between relations in this sense and relational complexes . According to Mei‑ nong, when there is a relation, there is also a complex, and the other way round . The complex is not simply a collection (Kollektiv) of a, b and R, but a unitary whole; the relational moment R unifies the parts a and b into a complex unity, but is not itself a constituent part (Bestandstück) of the complex (Meinong 1899, pp . 389‑391) . The latter point seems to be of im‑ portance in order to prevent a Bradleian regress: if the relational moment would be a part of the complex in the same sense in which a an b are, we would have to ask what relates the relational part to the other parts . The constituents are related to the complex, as parts are related to the whole they form part of . This means that in order for the complex to exist, the parts have to exist . This point makes the complex different from a state of affairs, at least on Meinong’s account of the notion . For Meinong, a state of affairs, or Objective, is an ideal entity, outside space and time, which may obtain or not obtain . What is the relation between a state of affairs that a is related to b and the objects a, b and the relation R? Are they parts passage from the Second Quarto is probably better known, having “your” instead of “our” . In the context above, the choice of the Oxford Edition is to be preferred . 94 Maria van der Schaar of the state of affairs? If the state of affairs is an ideal object, as Meinong argues, this seems to be excluded . If the state of affairs is understood as an ideal entity it seems indistinguishable from a proposition: both are proposi‑ tionally structured ideal entities . A subsisting state of affairs would thus be identical with a true proposition, and a non‑subsisting state of affairs would be a false proposition . And this is precisely Meinong’s position . Of course, we need not follow Meinong in this, but his position is of importance if we want to understand Twardowski’s development from Brentano’s theory of judgement towards a new account of judgement . Meinong’s position is here described not for historical reasons, for Twardowski’s account of relational judgements is given before Meinong had published his ideas on the subject . It is needed here for conceptual reasons . In order to understand Twardowski’s position we need to understand what he means by “state of affairs,” or Sachverhalt (or stan rzeczy in Polish), which he takes to be the object of a relational judgement . Is it a relation, which is dependent upon its terms, a relational complex, of which the related terms form a part, or an objective in Meinong’s sense, which is an ideal entity with a dependence relation towards its terms, or is it neither of these notions? What kind of entity could fulfil the role of being the judgemental con‑ tent? As Brentano and Twardowski claim, the content of the judgements “Tigers exist” and “Tigers do not exist” may be understood as, respectively, the existence of tigers and the non‑existence of tigers . According to Bren‑ tano, when we elucidate (erläutern) the concept of truth of the affirmative judgement through the concept existence of the object, and the truth of a negative judgement through the non‑existence of he object, we do not give a definition or explanation of truth (Brentano 1889b, §57, p . 27) . For Brentano, the existence of A has no conceptual priority over the truth of the judgement . And, the existence of A has no being independently of the act of judgement; no logical realism is involved . As Twardowski also acknowledges relational judgements, how does he account for the judgemental contents of non‑existential judgements? If one does not endorse Brentano’s account of judgement, there are two ways to understand the notion of judgemental content . One may argue that the thought that snow is white is the judgemental content . The modern notion of prop‑ osition is understood as having a that‑structure . One may also understand the judgemental content to be that snow is white is true, that is, as snow is white . The content has here the structure of a declarative sentence, that is, it contains an element corresponding to the declarative mood . This notion is traditionally also called “proposition,” which makes it understandable that the two notions are often confused . The traditional notion of proposition has the structure of a declarative sentence . The need to distinguish the two notions becomes clear if one contrasts the judgemental content in the sec‑ Judgement and Meaning 95 ond sense with the interrogative content of the question “Is snow white?” . The interrogative content is whether snow is white, or is it true that snow is white . The interrogative content can thus be understood as consisting of the proposition in the modern sense together with the meaning of the interroga‑ tive mood . In the case of an imperative “Let snow be white”, the imperative content is let snow be white, or let it be true that snow is white . In each of the three cases we see a common element, that snow is white, which is the proposition in the modern sense, and an element in which they differ: the meaning of the declarative mood; the meaning of the interrogative mood; and the meaning of the imperative mood . The proposition in the modern sense together with the meaning of the declarative mood gives the proposition in the traditional sense . The proposition in the traditional sense is also to be distinguished from the judgement (made), for it need not be accompanied by judgemental force . One may understand a judgement, without judging oneself, and one may understand a question, without asking it . What we apprehend in these cases is the judgemental content and the interrogative content, but no force is involved on our part . The distinction is of importance to understand Twardowski’s notion of judgemental content defended in the paper on actions and products from 1912 . Twardowski compares his notion to Husserl’s notion of ideal meaning in the first of the logical investigations, and, as I will argue, this notion is not to be identified with the modern notion of proposition . For Husserl, the ideal meaning is a judgemental content in the full sense of the term explained above; it is the proposition in the traditional sense . The ideal meaning (die Bedeutung in unserem idealen Sinn; cf . (Husserl 1901, p . 431, A 392) is apprehended through an idealizing abstraction upon the essence of partic‑ ular acts . The ideal meaning of an act of judgement can be apprehended by first apprehending the intentional essence of the act, the phenomenological counterpart to the ideal meaning . For Husserl, the intentional essence of a judgemental act covers both quality and matter of the act, where the quality corresponds to the fact that a judgement is made, rather than that a question is asked . If the ideal meaning is an abstraction of both these elements it must be more than the proposition in the modern sense, for it contains also an idealization of the quality of the act . The ideal meaning must therefore contain two elements: one corresponding to the matter, the proposition in the modern sense, and one corresponding to the quality of the act . Husserl thus speaks about the judgemental content in the full, traditional sense, covering both the proposition in the modern sense, and the meaning of the declarative mood, as it is explained above . That the identical meaning also involves the meaning of the declarative mood becomes clear at the end of the section, where Husserl says that the identity of “the” judgement lies in the identical meaning, which repeats itself in the different particular acts 96 Maria van der Schaar as the same judgement .9 Husserl is speaking here of “the judgement” in the ideal sense . As “the judgement” as ideal meaning is there independently of individual acts of judging, the judgement need not be actually judged . For Husserl, there are many ideal judgements that have never actually been passed; no judgemental force is therefore involved on the level of the ideal judgement . If we put it in terms of the distinction between judgemental force and judgemental form, we may say that Husserl’s ideal judgement captures the judgemental form, say, A is true, but not the judgemental force, for that is a pragmatic notion that can be acknowledged only for particular acts . In an actual judgement, the ideal judgement is instantiated as this particular act of judgement, and only the latter has judgemental force . The term “judgement” shows, like the terms “assertion,” “presentation” and “perception,” an act/product ambiguity . The term may be used for the act of judgement or for the judgement made, the product of the act . This distinction plays an important role in Twardowski’s answer to the problem of psychologism . It is not well known in modern philosophy, but at least speech act theory is in need of the distinction . The act of promising results in a promise made, which is still valid after the act has gone . The promise made seems to be an abstract object, created at a certain moment, but we may talk about it independent of the particular moment in which it was cre‑ ated . It may be perfectly clear what the promise is, and we may talk about it, without invoking the moment of promising . In reasoning the premises and conclusions are assertions or judgements made, that is, they are products of the corresponding acts . Conclusions made by the scientist on the basis of his investigations are judgements made, too . Science is not to be understood as a collection of beliefs of individual scientists . A science is a complex whole of judgements made, each of them being an abstract object created by a scientist, or a group of scientists . There would be no science without scientists who created it, but once created, the scientist may die, while we are still able to make use of his creations . The distinction between act and product may thus be used to provide an answer to psychologism, without completely neglecting the act . Brentano would not be willing to allow for such an abstract object in his ontology . And on the Brentanist account Twardowski proposes, the judgement made is to be understood as being there only as long as the judgemental act has being . The notion of judgemental product or judgement made is to be distinguished from the notion of ideal judgement, the proposition in the traditional sense, explained above, for the latter does not involve the element of judgemental force, whereas the

9 Die Identität ‘des’ Urteils oder ‘der’ Aussage liegt in der identischen Bedeutung, die sich in den mannigfaltigen Einzelakten eben als dieselbe wiederholt (Husserl 1901, p . 435, A 395, 396) . Judgement and Meaning 97 judgement product is essentially created by an act of judgement involving judgemental force . One of the central question in overcoming psychologism, and in the de‑ fence of the absoluteness of truth, is the question what the bearer of truth and falsity is . Twardowski follows Brentano’s claim that the judgement is the primary bearer of truth and falsity . Does he mean to say that the act of judgement is true or false? Is the judgement product for him the bearer of truth and falsity, or is it the judgemental content that fulfils this role?

3. A Development in Twardowski’s Early Account of Judgement

In 1894, Twardowski seems to follow Brentano closely in defending an existential theory of judgement . At the same time, Twardowski applies the distinction between content and object he has introduced for presentations to judgements . Judging is a special way in which we are intentionally re‑ lated to an object, distinguished from presenting, and there is nothing in between the two kinds of mental acts (Twardowski 1894a, p .5) . We have to distinguish three notions: the act of judgement, the content of judgement and the object of judgement . The act of judgement is an acknowledging or rejecting . In the judgement the existence of the object of judgement is ac‑ knowledged or rejected, where the existence of the object is the content of the judgement . If one affirms the existence of tigers, tigers or a particular tiger form the object of the judgement, and this object is given by an un‑ derlying act of presentation independently of the question whether a tiger exists . The existence of tigers is the content of the judgement . Is the content of judgement a proposition or a state of affairs, according to Twardowski? Stumpf called the existence of A the judged state of affairs (Sachverhalt) . For him, as for Twardowski, the content of the judgement “God exists” is the being of God or the existence of God (Stumpf 1888, p .242) . Twardowski, though, does not introduce the terminology of Sachverhalt for the content of judgement, as we will see below . In 1894, Twardowski shows an interest in relational judgements, for they seem to form a kind of their own besides existential judgements . At this time, Twardowski’s analysis of relational judgements stays within the framework of Brentano’s existential judgements with one exception . For Brentano, if the judgement with a complex whole as object is correct, the judgement which has only a part of this whole as its object must be correct, too . The existence of a complex whole entails the existence of the parts of this whole; the existence of a red table entails the existence of something red . If one understands relations as complexes, and one allows that relations may exist while the related terms do not, as Twardowski does, it seems that 98 Maria van der Schaar

Brentano’s thesis can no longer be defended . If we take the judgement the number four is greater than the number three, the judgement is correct, and the relation between the two numbers therefore exists, but this does not imply that the numbers exist (Twardowski 1894a, p . 27) . Whereas the early Meinong and Höfler understand relations to obtain between contents, which always exist as part of a mental act, Twardowski understands relations to obtain between objects . The standard relational judgement is not about contents, but about the related objects, in the example given above, the numbers themselves, “irrespectively of whether they exist or not” (idem) . Relations are for him objective and not dependent upon the mind as they would be if relations were to relate nothing but mental contents . As Twardowski puts it in the opening lecture at the Lwów University, 15th November, 1895, the whole theory of relations belongs to the domain of metaphysics: a causal relation or a relation of similarity may relate mental and/or physical parts, but the relation itself is neither mental nor physical (Twardowski 1895, p . 37) . The point is essential for Twardowski’s account of intentionality presented in 1894 . Whereas for Brentano the content to which the act is directed always exists as part of the act, for Twardowski, intentionality is a relation between a thinker and the object thought about, and this object need not exist . One may thus correctly acknowledge the re‑ lation of thinking between an agent and an object, on Twardowski’s account, while one of the terms of the relation does not exist . Twardowski is still thinking about this problem in the Logik manuscript written for the winter seminar 1894/1895 in Vienna . He considers the two judgements: (1) God exists . (2) Once there was a king . In the first case the content of the judgement is existence; what in the second? Many, like Prof . Brentano himself, think that also here existence is the content . But this can‑ not be the case . For: once there was a king = + a past king . Past = having existed, but not existing anymore . Thus the judgement seems to claim something contradictory (Twardowski 1894/1895, p . 42, note) 10. According to Brentano, a past object may be the object of a correct affir‑ mation . Because, for Brentano, every object of a correct affirmation exists, a past object may thus exist . Twardowski takes existence to be spatio‑tem‑ poral existence, and this means that a past object, for him, does not exist .

10 Im ersten Fall ist der Inhalt des Urteils die Existenz; was im zweiten? Viele, wie Prof. Bren‑ tano selbst, meinen, auch da sei Existenz der Inhalt. Aber das geht nicht. Denn: ein[st] war ein König = + einen gewesenen König. Gewesen = existiert haben, aber jetzt nicht mehr exis‑ tierend. Also scheint das Urteil etwas widersprechendes zu behaupten . An extensive analysis of the problem is given in (Betti & Van der Schaar 2004) . Judgement and Meaning 99

For Twardowski, a judgement in which a past object is acknowledged cannot be a simple judgement of existence . What is affirmed in the example just given is the subsistence of a relation of simultaneity between a king and a certain time in the past (p . 43, note; on subsistence, see below) . This re‑ lation may subsist, although the terms of the relation do not exist anymore . For Twardowski the adjunct of time does not express a determination of the content (once existing), but a determination of the object (cf . pp . 43, 76) . It is of great importance to Twardowski’s argument against the relativity of truth that determination of time and place belong to the judgemental object (see the final chapter on truth and time) . In the Logik manuscript Twardowski argues that there are two kinds of judgements: existential and relational judgements . A judgement is the affir‑ mation or denial of an object, or the affirmation or denial of a relationship (Verhältnis; p .43) . In both cases, there is something, for which Twardowski uses the term “es gibt” . In the case of an existential judgement, an object is affirmed or denied, and the content of the judgement is existence . Its form is: ± e A, where the signs “+” and “–” show, respectively, that an affirma‑ tion or a denial has been made, “e” expresses the existence of, and “A” is a schematic letter for a general term or a singular term . In the case of a rela‑ tional judgement, a relationship is affirmed or denied, and the judgemental content is subsistence (Vorhandensein, Bestehen) . Its form is ± b A, where “b” expresses the subsistence of. Here “A” is a schematic letter for a term expressing a relationship 11. For Twardowski, all categorical judgements are relational judgements, and the latter form thus an important class . It even seems that he takes all relational judgements to have the traditional categorial form: “the gen‑ eral form of sentences expressing a relation is: A has b” ((Twardowski 1894/1895, p .116); Twardowski refers in this context to (Bolzano 1837, §127)) . “Between A and B there is the relation of cause and effect” can be reformulated as “A has the quality of effecting B” (p . 113) . The object of all relational judgements is the relation of having a quality b by the object A (p . 116) . And the relation may subsist although A does not . The judge‑ ment “Two infinitely long parallel lines do not intersect” is an affirmation of a relationship which may subsist, although two infinitely long parallel lines do not exist (p . 100) . For Brentano, such a judgement is a denial of the existence of two infinitely long parallel lines that intersect . Twardowski’s analysis of such general judgements is thus essentially different from the one given by Brentano . For both, the general judgement does not involve an existence presupposition . But, in contrast to Brentano, Twardowski is able to account for this without understanding all general judgements to be

11 Cf . (Twardowski 1894/1895, p . 42) . This paragraph is taken from (Betti & Van der Schaar 2004, p . 8), with slight modifications . 100 Maria van der Schaar negations . The judgement “Once there was a king” can now be understood as a relational judgement . The content of this judgement is the subsistence of the relationship of contemporaneousness between a king and a certain period of time in the past, and the judgemental object is this relationship . On this account, one does not assert that a (past) king exists, when one makes the judgement . What is a relationship (Verhältnis), according to Twardowski? The ex‑ ample about past objects seems to show that relationships do not exist in time; it is precisely for this reason that Twardowski speaks of the subsistence of a relationship . Is a relationship an ideal entity outside space and time? Certainly, a relation is here not to be understood as a relational moment ob‑ taining between two objects, for such a moment can exist only if the terms of the relation exist . It also seems to be more than a relational complex, for as the notion is introduced above and understood by Meinong, a relational complex exists only if the relata and the relational moment exist . There is a second manuscript on logic, which contains notes from a course given a year later in Lwów . Here, Twardowski speaks of the object of the judgement “the sun rises” as the rising of the sun (Twardowski 1895/1896) . Whereas the term “the rising sun” may be understood as naming a relational complex, which exists only if the rising and the sun exist, “the rising of the sun” may be understood as denoting a state of affairs . In a letter to Meinong from 1897, Twardowski uses the term Sachverhalt for the object of judgement (Twardowski 1897b, pp . 143-144) . Although there is no reason to think that Twardowski used the term in any technical sense, his notion of Sachverhalt is clearly distinguished from the notion as it is introduced by Stumpf for the content of judgement .12 Twardowski is close to acknowledging a special object, a Sachverhalt, for the important category of relational judgements . It is to be noted that Twardowski’s Sachverhalt as object of judgement is acknowledged before Meinong developed his account of the objective at the turn of the century . Twardowski may thus have influenced Meinong in his development of the notion of the objective . If Twardowski defends the thesis that there are two kinds of judgements, the existential and the relational ones, can he still say that all judgements are existential? In a certain sense he can, for in both cases existence or sub‑ sistence is affirmed or denied . So, he could introduce a notion of existence that is general enough to cover the two cases . The important difference with Brentano’s existential account of judgement consists in the fact that Twardowski acknowledges a judgemental object, a state of affairs, which has to be ideal insofar as it may contain parts that do not exist . This posi‑

12 A more extensive defence of the thesis that Twardowski has developed a notion of state of affairs as the object of judgement is given in Betti & Van der Schaar (2004) . Judgement and Meaning 101 tion can still be found in the explanation of judgement given in the lecture notes for a course on the psychology of thought, delivered in 1908/1909 . In an unfinished manuscript on the theory of judgement, which is written for a series of lectures delivered in the academic year 1902‑1903, Twardows‑ ki follows another strategy . The question of the unique characteristic of judgement is put in a new way . There is a criterion which distinguishes judgements from all other mental activities (the first criterion): they can be described as “true” or “false” in the proper meaning of these terms . This criterion can be of use, only if we have a criterion to distinguish the proper from the improper meanings of these terms . Twardowski gives such a cri‑ terion in terms of four rules: (1) If “true” is used in its improper sense, as we use it in “true emotion,” the term “true” can be replaced with the term “genuine” without change of meaning of the phrase . (2) If “false” is used in its improper sense, as in “false friend,” “false wit‑ ness,” or “false gold,” it can be replaced by phrases such as “feigned” or “forged” without change of meaning . (3) If “true” is used in its proper sense, it can be replaced by the phrase “consistent with the truth” without change of meaning . (4) If “false” is used in its proper sense, it can be replaced by the term “erroneous” without change of meaning . Because only judgements may be mistaken, substitution rule (4) functions only for judgements or for sentences intimating a judgement . In rule (3) and (4) “true” and “false” are used as attributive terms (see II .4); attributive adjectives signify a feature of the objects referred to . Twardowski is able to use the distinction between attributive or determining and modifying adjec‑ tives, introduced in the first chapter, to elucidate a philosophical distinction again . “True” and “false” can be used in both ways . If we speak of a “false witness,” the term “false” is used in its modifying sense, for a false witness is not a witness, because this person is only pretending to testify . When we speak of a “true witness”, “true” is not a determining adjective either, for it does not complement the meaning of the word “witness” . In terms of rule (3) and (4), “true” and “false” can be substituted by, respectively, “consistent with the truth” and “erroneous” without change of meaning . If and only if these terms are used in their determining sense, we know that judgements are involved . For example, if we speak about a true sentence, and understand that the term “true” is used in its determining sense, we know that this sen‑ tence is called “true” because it intimates a true judgement . This gives us a second criterion to determine whether judgements are involved, which is clearly related to the first one: “[I]f these adjectives are defining [= deter‑ mining, attributive], they indicate that the nouns they belong to represent 102 Maria van der Schaar judgements” (Twardowski 1902‑1903, p . 172) . The sentences may “express judgements,” that is, they may function as intimations of judgements, but may also “contain either wishes, or requests, or commands, or finally, questions” (p . 179) . If a sentence containing or intimating a request or wish is called “true”, it is true insofar as it contains a true question or a true wish, and the word “true” is used in its improper, non‑attributive sense . If and only if the adjective “true” is used as an attributive term, a judgement is intimated by the sentence . Sadly, the manuscript suddenly ends here . It is clear from these passages that Twardowski cannot be accused of Austin’s descriptive falla‑ cy . For Twardowski, not all sentences can properly be called true . In 1901, he already noted that we can do different things with sentences, although he puts it in the old fashioned language of what is “expressed” by the sen‑ tence . “A sentence may express an inquiry (e .g . “When are you coming?”), an order (e g. . “Quiet!”), a wish (e g. . “I wish it were sunny tomorow”), an affirmation (e .g . “The sun is shining”), or a negation (e .g . “Lwów does not lie on the Vistula river”) . . A judgement is an intellectual action, whereas a statement [= a declarative sentence] is an external expression of that action, a judgement is a psychic, spiritual phenomenon, whereas a statement is a physical, sensory phenomenon” (Twardowski 1901, p . 119) . In 1907, Łukasiewicz publishes a paper on logic and psychology, the beginning of the anti‑psychologistic turn in Polish philosophy, so before he visited Meinong in Graz, but after he had read Husserl’s Logical Investiga‑ tions and Russell’s Principles of Mathematics . He distinguishes between a psychological principle of contradiction, there cannot exist contradictory beliefs in one person at the same time, and a logical principle of contradic‑ tion, which says that of two contradictory judgements one has to be false, preluding his work on the principle of contradictions from 1910 (see chapter VI) . Logic concerns the laws relating the truth and falsity of judgements, whereas psychology concerns mental events and processes . Terms such as “judgement” and “reasoning” are ambiguous, according to Łukasiewicz. In the context of psychology, the term “judgement” refers to a belief or an act of judging . The “judgement in the logical sense” is the fact that something exists or does not exist in such and such a way. Łukasiewicz understands logical judgements as states of affairs which are expressed in words; there are no judgements in the logical sense without language in which they are expressed (Rojszczak 2005, p. 130). Łukasiewicz’ logical judgement seems to agree with what Stumpf has called the state of affairs, the content of judgement, with one important difference . Whereas for Stumpf the state of affairs is dependent upon the act of judging, for Łukasiewicz, the logical judgement is a meaning, and as such dependent on language, but not on the judgemental act. Łukasiewicz’ state of affairs does not have the independence Meinong attributes to the objective, and it also differs from Twardowski’s Judgement and Meaning 103 notion of state of affairs or relationship, which is the object, and not the content of judgement. Twardowski was present when Łukasiewicz presented the paper, and it may have influenced Twardowski’s development towards the idea that the term “judgement” is ambiguous . It may refer to the act of judging and to the judgemental product as well, where the latter, being the proper bearer of truth and falsity, is the logical notion. Like Łukasiewicz, Twardowski is not willing to acknowledge a Platonic realm of meanings in specie, the way Husserl thinks is necessary to overcome psychologism in logic . In what way does the distinction between judgement as act and judgement as product help Twardowski to overcome his earlier psychologis‑ tic account of meaning? And, is he able to account for the absoluteness of truth by acknowledging a bearer of truth and falsity that is less dependent upon time than the act of judgement is?

4. Actions and Products

By uttering a declarative sentence one makes an assertion, at least, if there are no indications that the declarative sentence is used for other purposes . When sentences are written on a blackboard, the context shows that the oc‑ currence is not used to make an assertion . When I have made an assertion, I can be held accountable for it, and have to be able to give an answer when asked “How do you know that?” . The assertion is thus still valid after the moment of the act of assertion, and we thus speak of the correctness of the assertion made . We might say that the assertion made is justified, or that the assertions made by the witness were the basis for the conviction of the accused . A similar situation arises when a promise is made . The promise made is valid until one’s word is redeemed . In case the act consists of both mental and physical aspects, the product may have a concrete existence: the act of writing a letter results in a written letter that can be posted . An act of composing results in a composition, and if it is written down it results in a written composition . Why did Twardowski get interested in the distinc‑ tion between actions and products? What is the ontological status of purely mental products? Are they abstract objects created at a certain moment of time? If one does not acknowledge such abstract entities, would it still be possible to speak about judgements and assertions made? For Twardowski, the distinction between action and product is of central importance to prevent psychologism in logic . Whereas psychology studies the act of judgement, logic is concerned with the judgement product . The distinction between actions and products is therefore primarily of importance for the notion of judgement . In the lectures on psychology of thinking, which Twardowski delivered in the academic year 1908/1909, Twardowski argues 104 Maria van der Schaar that “the object of logic is thought, not thinking; not a mental function but rather its product” (Twardowski 1908/09, p . 135) . Twardowski says: Discussion of judgement is the domain of logic, which studies them according to their veracity or falsity, whereas psychology deals with the function of judging (idem) . It is thus the judgement product, and not the act of judging, to which truth and falsity pertain . Whereas psychology claims, according to Twardowski, that harbouring two contradictory beliefs is impossible, logic claims that two contradictory judgements cannot both be true, and cannot both be false . The psychological law may be refuted by the obtainment of new empirical evidence, but no such evidence would count against the logical law . Accord‑ ing to Twardowski, logic consists of two parts: a theoretical part, logic in its proper sense, and a practical part, a whole of rules and guidelines how to think . The latter is a technique that concerns our acts of judging and in‑ ferring . This practical part of logic is based on logic in the proper sense, a theoretical study of the truth of judgements made (p . 109) . The distinction between the act and product of judgement can already be found in Bolzano’s Wissenschaftslehre (1837) and in Julius Bergmann’s Reine Logik (1879), where the terms “Erzeugnis” and “Werk” are used to denote the product . According to Bergmann, logic is concerned not with the act of judging but with the judgement, the products (Erzeugnisse) of the act of judging (Bergmann 1879, p . 2) . In the same way, a thought (Gedanke) is what the act of thinking constructs (bildet) from its object, the object about .13 We can thus speak of a product as a construction or formation (Gebilde) . Language has the possibility to distinguish between judging (das Urteilen) and judgement (das Urteil) to make a distinction between the transitory act and its enduring product (p . 39) . Immediately before Twardowski makes the distinction for the first time in the academic year 1908/09, C . Stumpf and S . Witasek published on the distinction between actions and products in, respectively, 1907 and 1908 . Cf . (Witasek 1908, pp .71ff) . Both Stumpf and Witasek use the term “Gebilde” for the product of a mental process or action . Especially Stumpf’s separate work Funktionen und Erscheinungen (Stumpf 1907) must have stimulated Twardowski to elaborate on the distinction, and to use it to disambiguate the term “judgement ”. Stumpf’s distinction between function and product (Gebilde) should not be confused with the distinction between function and appearance (Erscheinung, Empfindung); the latter is comparable to

13 In der letzteren Bedeutung ist es so viel wie das Erzeugniß des Denkens, das Werk, welches das Denken aus seinem Objekte bildet, also gleichbedeutend mit Gedanke, sofern darunter nicht, wie oft geschieht, der Akt des Denkens verstanden wird. Nur in dem zweiten Sinne des Wortes ist das Gedachte Gegenstand der Logik (Bergmann 1879, pp . 11-12) . Judgement and Meaning 105

Brentano’s distinction between act and object . Functions may cover acts, states and experiences, whereas appearances are, for example, the sensory objects to which these functions are directed . Function and sensory object are dependent upon another, and they can only be distinguished by means of abstraction . The distinction between function and product is of a different kind . Products may have an objective character (Stumpf 1907, p . 30, note 1), and Stumpf compares the judgement product to Bolzano’s Satz an sich, although Bolzano himself explicitly distinguishes the Satz an sich from the judgement product . The judgemental product, also called a state of affairs, a Sachverhalt, by Stumpf, is expressed by a that‑clause, “that there are no Cyclopes,” or by an gerund construction “the nonbeing of Cyclopes ”. Ac‑ cording to Stumpf, we can think of a state of affairs without judging, but the state of affairs is only “real” as content of an actual judgement . Stumpf had introduced the notion of state of affairs already in 1888 for the content of judgement, as we have seen above . By identifying the judgement product with the state of affairs as judgemental content, he identifies the content with the product of judgement . Just as Stumpf, Twardowski says, in a footnote, that the distinction be‑ tween the function of presenting and the presentation product is identical to the distinction between act of presenting and content of presentation he had introduced in 1894 (Twardowski 1912c, p .114, note 30) . This identification may seem quite harmless if one looks at the content and the product of pre‑ senting . It is problematic, though, when applied to the notion of judgement . Whereas the judgemental force is absent from the content of judgement, it is present in the judgement made . The distinction might be less clear if one argues that both the content and the product are dependent for their existence on the act, as Stumpf and Twardowski do . As soon as one argues, though, that one of the two notions is an abstract entity that has a being independently of the act, it becomes clear that the two notions cannot be identified . For Bolzano, the Satz an sich, which has the role of being the content of a judgement, is distinguished from the judgement product that can arise only as the result of an act of judging (Bolzano 1837, I, §20) . For him, the judgement (das Urteil) is an effect (eine Wirkung, etwas Gewordenes) produced (hervorgebracht) by a judging (Urteilen) . The judgement is thus a product of the act of judging, and has to be distinguished from a Satz an sich, which is not a product of a particular judging or proposing (ein Setzen) . If Twardowski’s distinction between judgemental act and judgement as product would reduce to the distinction between act and content of judge‑ ment, his distinction would be far less original, as the latter distinction is already extensively made inside and outside the Brentano school before 106 Maria van der Schaar and around 1900 .14 Although there is sometimes a certain ambiguity in the notion of judgement product in the paper “Actions and Products,” in most cases it is clear that Twardowski intends to speak about the judgement made, the result of the act of judging, and not about the judgemental content . One may thus conclude that we should understand Twardowski’s notion of judge‑ ment product as being identical to the judgement made introduced at the beginning of this section . This interpretation is confirmed by Twardowski’s semantics and the introduction of the notion of represented judgement as we will see below . For Twardowski, purely mental products have no existence beyond the act; it is only the combination of a physical with a mental act that may re‑ sult in an enduring entity . Enduring products last longer than the activity from which they result . A building lasts longer than the activity of building; a cake endures at least some time after the activity of baking it . Judging in silence does not lead to an enduring product, but writing at the same time a declarative sentence that is intimately related to the judgement, results in a sentence written down . The written sentence is in a certain sense the ef‑ fect of the act of judging, but more is involved . A psychophysical product, the written sentence, is the result of a physical action, the writing, which is accompanied by a mental act which has some influence on the result‑ ing product (Twardowski, 1912b, p .109) . The psychophysical product, which is perceivable, may thus be understood as the external expression of a non‑perceivable product (p . 120) . A shout may be the expression of pain; a particular head movement may be the expression of an affirmative judge‑ ment . In a similar way, the written or spoken sentence can be understood as the expression of the judgement made, and the latter is the meaning of this particular sentence . A judgement as the product of the action of judging – that is to say, of passing judge‑ ments – expresses itself in sentences … Such sentences then signify judgements; in other words, their meanings are judgements (p . 129) . The judgement expressed by the utterance of a declarative is thus in standard cases a judgement passed, and not a judgemental content . If Twardowski had identified the judgement expressed by the declarative with the judgemental content, he would not have needed to introduce the notion of represented judgement, that is, a notion of judgement in which the judge‑ mental force is absent, a notion that will be explained below . The verb “to express” may be used to refer to a relation between a word or sentence and its meaning, but it may also be used as a synonym for “to intimate” . We appropriately say that the meaning is expressed by a sentence,

14 Hanna Buczyńska‑Garewicz (1980, p. 160) argues, though, that the differentation adds nothing essentially new to the earlier distinction . Judgement and Meaning 107 but we do not say in the same sense of “to express” that the utterance of a sentence expresses a judgement made . If the sentence is uttered with asser‑ tive force, it intimates that a judgement has been made, but the meaning of a sentence is something else; it is an identical element in the different uses of the sentence . Because Twardowski does not clearly distinguish between the content and the product of judgement, he is not able to make a distinction between the semantic relation of expressing and the pragmatic relation of intimating . He does not make a distinction between the semantic notion of meaning that can be fully expressed in language, and the pragmatic notion of assertive force that can only be shown . Twardowski’s notion of meaning, the judgement made in the case of a declarative, involves both the semantic and the pragmatic aspect . For Twardowski, what a sentence expresses differs depending on the use that is made of the sentence . A sentence is always a sentence token in use, for it is essentially a psychophysical product of an act of speaking or writing, accompanied by a mental act . This means that, for Twardowski, each time a sentence is used a different meaning is given to it . How can he account for the fact that a sentence has a meaning independent of the concrete use that is made of it? For Twardowski, the meaning of the utterance of a declarative is, in the primary sense, the judgement made which produced this utterance . The judgement made exists only at the moment the act of judging occurs, where‑ as the uttered declarative also has a meaning after the moment of judging . Twardowski thus needs to introduce a second notion of meaning . In the written sentence there potentially exist judgements made, in the sense that this sentence may be the cause of a judgement produced by some‑ one who reads the sentence: “the mental product exists potentially (though by no means truly and actually) in the psychophysical product” (p . 125) . These potential judgements form the meaning of the sentence in a second sense . The potential existence of the judgements that can be made is thus ontologically and conceptually dependent upon the actual existence of the written sentence . It may be said that the meaning of the sentence is contained in or embodied by the sentence insofar as the sentence is a partial cause of the emergence of these mental products . The meaning in this sense exists potentially in the sentence as psychophysical product . In the first meaning the judgement product is the cause of the sentence, whereas in the second meaning the sentence is the partial cause of the different judgement prod‑ ucts that may arise . In order to explain that we may speak of an identical meaning of a par‑ ticular sentence token in use, Twardowski introduces the meaning of a sen‑ tence in a third sense . We can distinguish a group of common attributes in the mental products elicited by a particular sentence token, provided “that 108 Maria van der Schaar these common attributes correspond to the intent with which that psycho‑ physical product was utilized as a sign ”. (p .127) . By an act of abstraction performed on the particular mental products elicited by this token we are able to speak about an identical meaning, although ontologically speaking there is not an identical abstract object . An act of abstraction consists, ac‑ cording to Twardowski, in an act of attention, a special element is brought into relief, together with an act of mentally eliminating other characteristics (Twardowski 1903, p . 85; see chapter III) . The product of this complicated act of abstraction, the abstractum, may function as meaning of the linguis‑ tic product, but there is ontologically speaking no entity separate from the common particular attributes or moments of the individual acts . Speaking of an identical element is merely a way of speaking . Twardowski refers in a footnote to Husserl’s notion of ideale Bedeu‑ tung (p . 128, note 55) . There is an important difference, though, between Twardowski’s notion of third meaning and Husserl’s notion of ideal mean‑ ing as developed in the first of the Logical Investigations (Husserl 1901, pp . 102‑110, A 97‑A 105, §30‑35) . The ideal meaning is a species indepen‑ dent of thought and language, and not a product of abstraction . For Hus‑ serl, the relation between the meaning and the expression that has meaning (bedeutender Audruck) is the same as the relation between the species red and the red object given in perception (Husserl 1901, p .111, A 106) . For Husserl, we are only able to apprehend the meaning as species through acts in which the species is instantiated, but this does not imply that the species is conceptually and ontologically dependent upon these acts, as it is for Twardowski . For Husserl, the ideal meaning comes first in the order of ex‑ planation to the acts that instantiate the meaning . The ideal meaning is the identical logical content of an act, not its phenomenological content, which differs from act to act . For Twardowski, the identical result of abstraction is not a strict and true identity, as it is for Husserl (Husserl, 1901, p . 117, A 112) . Furthermore, each sentence token has its own abstract meaning corresponding to the judgement that caused the sentence . How is it possible for Twardowski to make a distinction between asserted and unasserted contents? As Geach puts it: “A thought may have just the same content whether you assent to its truth or not; a proposition may oc‑ cur in discourse now asserted, now unasserted, and yet be recognizably the same proposition . … I shall call this point about assertion the Frege point,” (Geach 1965, pp .254 -255) . Twardowski acknowledges the problem, and he introduces the notion of represented judgement as a solution . In order to explain this notion, he introduces the distinction between an artificial and a non‑artificial product (Twardowski 1912b, p . 129) . The uttered declara‑ tive is a non‑artificial product insofar as it is the effect of an assertion or judgement . In standard cases, Twardowski argues, the sentence expresses Judgement and Meaning 109 an assertion or judgement made . On stage, the actions result in artificial products . The actor’s sayings and doings do not express real feelings and actions; instead, they express imagined feelings and actions . Declarative sentences on stage are the result of a judgement that is imagined or present‑ ed . We may generalise Twardowski’s point by saying that sentences uttered on stage or used in other deviant contexts are the result of presentations of assertions, questions and orders, not of real orders, questions or assertions . Declarative sentences uttered on stage express, according to Twardowski, not a judgement made, but the presentation of a judgement . A presentation of a judgement is neither an actual judgement not a judgement made . One thinks of a judgement, but does not make the judgement oneself, and the product of this act of thinking is the meaning of the sentence as artificial product . The presentations of judgements are also of importance to logic, for the logician often does not judge the examples of premises and con‑ clusions given (pp . 129-130) . If a logician gives an example, he does “not actually pass the judgements that comprise the meanings of these sentences … His sentences are not psychophysical products that express actual – that is to say, passed – judgements. They express merely ‘represented judge‑ ments’ ” (p . 130) .15 A represented judgement is not a judgement, according to Twardowski, for the term “represented” is used in its modifying sense . For Twardowski, the meaning of a sentence thus differs depending on the question whether it is the effect of a judgement or of a presentation, that is, depending on what use is made of the sentence . Although Twardowski acknowledges the distinction between asserted and unasserted sentences, he is thus not able to account for the Frege point . He does not allow for an identical element in the artificial and the non‑artificial case . In this sense Twardowski gives a traditional account of the distinction between asserted and unasserted sentences that can, for example, be found in the Port‑Royal logic 16. Twardowski, though, was probably inspired by Bolzano’s account of the way we think of a proposition (Satz an sich) with‑ out judging it . In God, the true propositions are judged, but the false ones are only objects of presentations (Bolzano 1937, §34) . In a similar way, to doubt a Satz an sich is to have a presentation of it . There is thus no account of different propositional attitudes directed to the same proposition, on Bolzano’s account, for in the judgement “Snow is white” the Satz an sich, Snow is white, is the content, whereas the same Satz an sich is the object in case one merely entertains the though that snow is white (cf . §143) 17.

15 The idea is already present in the Vienna Logic lectures: Was also die vorgestellten Urteile betriffft, so sind das überhaupt nicht eine Art von Urteilen (Twardowski 1894/1895, p . 177) . 16 For references, compare (Marušić 2014, p. 274). 17 For Bolzano, the Satz an sich either appears in our mind as content of an act of judgement, or is present to the mind as the object of a presentations about the Satz an sich . Ein bloss 110 Maria van der Schaar

We can now determine the different meanings of the term “judgement” in Twardowski’s writings . It may refer to (1) the act of judging and the (2) judgement product . But it may also stand for (3) our capacity to judge (Twardowski 1894a, p .125, note 51), which man develops when learning a language . Such a capacity to judge needs to be distinguished from (4) a belief, the fourth meaning of the term “judgement ”. Below I give a fifth meaning of the term . When we possess a belief, “all that means is that we harbour dispositions owing to which the same sorts of products can arise in us in the future as have in the past” (p . 116) . For Twardowski, we possess a belief only if we have made the corresponding judgement once “in the past .” The distinction between a capacity to judge and a belief may be elucidated by Aristotle’s distinction between dunamis and hexis (Latin: habitus; cf . De Anima 417a21‑417b1) . Although both notions can be understood as a poten‑ tiality, belief may also be understood as an actualisation of the capacity to judge . Belief as a habitus is a disposition, a more specific capacity to make particular judgements . Such a disposition is a potentiality, to be explained in terms of its actualisations, the corresponding acts of judgement . We may explain Twardowski’s notion of belief that S by means of two necessary conditions together forming a sufficient condition for belief . According to Twardowski, someone has a belief that S, where one may substitute any declarative sentence for S, if and only if that person: (1) has once passed the judgement that S, and (4) has the disposition to make a judgement of the same kind as the first judgement . Because of the first clause, Twardowski’s explanation of belief does not hide the ambiguity often present in modern accounts of belief pointed at by Robert Audi . According to Audi, we have to make a distinction between a dispositional belief and a “disposition,” a capacity, to make a new judge‑ ment (“a disposition to believe”; cf . (Audi 1994)) . Most of us have a dis‑ positional belief that the North Pole is a cold place, but we do not have a dispositional belief concerning the result of a complex calculation if we have never made the calculation . We only have a capacity to make a new judgement, at least, if we know how to calculate . The distinction can be elucidated by the distinction between the third and the fourth meaning of gedachter Satz, what Kant has called a “problematic judgement,” is, according to Bolzano, nothing but a presentation of a proposition (Bolzano 1837, p . 4) . The logician who asserts that “If A is, so is B” neither judges that A is, nor that B is . The logician judges that [the truth of] the proposition that B is, is a consequence of the truth of the proposition that A is (idem) . The logician makes a judgement about these propositions . He has a presentation of the propositions that A is, and that B is; these propositions do not function here as contents of non‑judgemental mental acts, but as the objects of such acts, just as is the case for Twardowski . Judgement and Meaning 111 judgement, (3) and (4) above . If we have developed a capacity to calculate, we have a capacity, a dunamis, to make a complicated judgement for the first time . If we then know that 19+33=52, and we can remember the result, we have obtained a belief in sense (4), a habitus . Finally, the term “judgement” may stand for (5) a represented judgement . Twardowski has understood that we are in need of a fifth meaning of “judgement”, where we use the term not for a judgement actually passed, but for a judgement that is understood although it may not be passed yet . In what way can we make sense of this notion of judgement without committing oneself to the problems invoked by Twardowski’s notion of represented judgement? Just as we can speak about a judgement not yet passed, we can speak about a dance no one has danced yet, or a house that has not been built yet, that may even be impossible to build . The house that has not been built may exist on paper, designed by an architect . So, it seems that the term “judge‑ ment” may indeed show another ambiguity . One may speak of a judgement or assertion not yet judged, the ideal judgement in Husserl’s terms . It is a proposition in the traditional sense, not a proposition in the modern sense, for it has the form of a declarative sentence, although it does not include judgemental force . If one apprehends the meaning of a sentence, it should be clear what the assertion conditions are, that is, under what condition one would be entitled to put forward the sentence with assertive force . I may not be able to make the assertion, yet, because I will not be able to justify my assertion . What I apprehend when I apprehend the assertion condition of the sentence, is the ideal judgement . The ideal judgement may be under‑ stood as the meaning of the declarative sentence, just as the ideal question may be given the role of meaning of the interrogative sentence . The ideal judgement may also function as the content of judgement, what is judged is not the proposition that snow is white, but the ideal judgement that snow is white is true . The ideal judgement could thus function as an alternative for Twardowski’s notion of represented judgement . If one wants to avoid the Platonic terminology of ideal judgement, one may simply speak of the proposition in the traditional sense . In a footnote, Twardowski seems to come close to this notion of ideal judgement . When speaking about the potential existence of certain acts of judgement and their products, he says: This potentiality may be conceived within the broadest possible scope, as when we speak, say, of the “existence” of truths no one knows yet, i e. ,. of the “existence” of true judgements that no one has ever passed . Obviously, what is involved here is the capability of passing these judgements, and that which exists is not the judgements but the capacity to pass them (Twardowski 1912b, p . 116, note 36) . This point would also count for false judgements no one has ever passed . With our capacity to judge we have a capacity to make false judgements, too . For Twardowski, a truth no one knows yet exists in us, in our capacity 112 Maria van der Schaar to judge, the third meaning of “judgement” mentioned above . For him, there is ontologically speaking no ideal judgement no one has ever passed yet independently of our capacity to judge . Twardowski’s thoughts in “Actions and Products” may rightly be called Aristotelian . Is the semantics Twardowski presents in “Actions and Products” psy‑ chologistic? The meaning of a sentence is, at least partly, explained in naturalistic terms . The judgement product is the primary meaning of the sentence only if it is a partial cause of that sentence; as far as the prima‑ ry meaning is concerned the relation between the psychophysical and the mental product, that is, the relation of sign and meaning, seems to be arbi‑ trary . In principle, a judgement that the weather is cold could produce the sentence “The weather is hot,” because the judger produces that sentence by accident . The relation between sign and meaning is here arbitrary . On the other hand, the sentence itself should be able to be the partial cause for the emergence of similar mental products in those who apprehend the sentence (Twardowski 1912b, p .121) . So, if the witness makes an assertion by uttering the sentence “I have seen the accused coming out of the build‑ ing on that evening,” but judges at the same time that the accused did not come out of the building that evening, the product of this act of judging is not the meaning of the uttered sentence, in the second, and third, sense of meaning introduced above (pp . 122, 124) . The sentence must be able to elicit similar judgement products in those hearing the sentence . The relation between the second meaning and the sentence is thus not arbitrary . There is a meaning in the second sense only if there is an agreement between the judgement causing the sentence, and the judgements caused by the sentence . Still, Twardowski’s account of the second meaning is given in naturalistic terms alone . Identifying the non‑enduring mental product with the meaning of the psychophysical product, Twardowski argues that the meaning exists potentially in the product insofar as the latter is the partial cause of similar mental products . The meaning of a declarative sentence as abstractum, the third meaning, is explained in terms of the common attributes of products of individual acts of judgement . This meaning is the result of a mental act of abstraction . The abstractum is thus a result of an act of abstraction, and in this sense there is no identical meaning without an individual’s act of abstraction . As the abstraction is based upon common attributes of the judgements made, Twardowski’s semantics need not be understood as psychologistic . The common attributes which may be distinguished in the different mental products are there independent of our act of abstraction . The abstraction that results from the act of abstraction is thus not the result of an arbitrary, subjective act; it is the result of an act of abstraction upon common attributes that can be found in the judgements made . Judgement and Meaning 113

A second reason why Twardowski’s semantics is not psychologistic concerns the relation between an act and its product . The act of abstraction is not the natural cause of the abstraction, the identical meaning, but is in‑ ternally related to the abstraction . Although Twardowski’s semantics has a naturalistic basis, it is not purely given in such terms . The main reason is that the relation between action and product is not a natural, but an internal relation: there cannot be an act of judging that the weather is cold without there arising a judgement product that the weather is cold, and there cannot be a product without there having been an act from which it results . If it would be a natural relation, the relation between action and product would be contingent, and one could have the action without the product . The act is not causally related to its product . The act of writing a letter, if complet‑ ed, necessarily results in the written letter, and the accompanying acts of judgement necessarily result in the judgements made expressed in the let‑ ter . The written letter itself may be the cause of further judgements in the one who reads the letter . If these judgements play a role in Twardowski’s semantics of the written letter, causal relations do form part of his explana‑ tion of meaning, but one should not forget that arbitrary judgements caused by the letter are excluded, for they can only be part of the constitution of the meaning in a second and third sense if these judgements agree with the original judgements of the writer . There are thus important reasons to en‑ dorse the thesis that Twardowski’s semantics is not purely psychologistic or naturalistic: an abstraction is based upon common attributes, and the relation between action and product is not a natural relation .

5. Twardowski’s Critique of Russell’s Multiple Relation Theory of Judgement

There is a tight relation between the theory of truth a philosopher presents and the account of judgement he gives: “The definition of truth presupposes a particular perspective on the essence of judgement” (Twardowski 1925, p . 194) . The critique Twardowski gives of Russell’s correspondence defini‑ tion of truth as presented in The Problems of Philosophy (1912) is therefore preceded by a critique of the multiple relation theory of judgement Russell defended at the time . Truth, according to Russell, pertains to judgement or belief, and consists in the correspondence of a belief with a fact . Russell argues that our activity of judging cannot consist in a two term relation of a mind to a fact – for such an account would give us no explanation of false judgement, as there would be nothing to judge . For Russell, judging is a relation connecting several terms . In case Othello judges that Desde‑ mona loves Cassio, the judging involves four terms: Othello, Cassio, love 114 Maria van der Schaar and Desdemona . The relation of judging connects the different terms, and gives the terms a sense or direction which places them in a certain order . The unity and order should be different from the one that is constituted by the relation of love if they actually love each other, for if it is a fact that Desdemona loves Cassio, this is not the result of Othello’s judgement . It may be true that women sometimes love another man as result of their husband’s judgement, but that seems to be an empirical fact, not a fact to result from one’s theory of judgement . Love does not have its relating function in the act of judging; it is merely a term to be connected by the act of judging . If the judgement is true, there corresponds to the judgement a complex in which loving does occur as a relation, actually relating Desdemona to Cas‑ sio in the same order as was done in the act of judging . Russell’s theory is of importance to Twardowski, because it is a variant of the traditional account of judgement as a combining or separating of ideas . The only two differences consist in the point that for Russell, in his mature account of the multiple relation theory, a judging connects objects, not ideas, and that there is for him no separate act of denial . There is today a revival of Rus‑ sell’s multiple relation theory of judgement . It is therefore of interest to see what Twardowski’s critique is . In the first place, Twardowski argues that judging is not a relation . The act of judging is not characterised by the fact that a relation is involved . As the act is sui generis, it is not to be explained as a relation together with a specific difference (Twardowski 1925, p .204) . Judging is something in the mind, owing to which a relation comes about, but it is itself not a relation (p . 198) . As Twardowski puts it, “the decision to take a trip is not a “relation” between the mind that decides and the taking of the trip” (idem) . One may speak about a relation between the mind and the judgement made, but this relation is constituted by the fact that one judges, and the latter fact has to be explained independently of the mentioned relation . Twardowski’s point is that judging is an activity, and that an activity is not to be understood as a relation . The activity of taking a walk, results in the walk taken, but one cannot explain the activity in terms of a relation between the walker and the walk taken, for we would then explain the action in terms of what comes second in the order of explanation, the product . It is true that the early Twardowski acknowledges an object of judgement, a simple object or a state of affairs, to which we are related in the act of judgement, but such a relation is not defining for the judgement, as there are other ways in which we may be related to such an object . Furthermore, for Twardowski, the judgement has a content, which directs it to this particular object rather then to another; the relation is thus constituted by the fact that the judgement has a certain content, and the latter is prior in the conceptual order to the judgement’s relation to the object . This content makes a par‑ Judgement and Meaning 115 ticular judgement into the judgement it is, but it cannot be used to define the act of judging as such . The latter is an undefinable act of affirming or denying, and the notion of content is not part of it . Second, Russell uses the terms “belief,” “judging,” “statement” and “judgement” interchangeably (p .198) . Twardowski argues that Russell does not see that we have to make a distinction between the act of judging and the judgement made . This is of importance for the question of truth, for it is the judgement made, not the act of judging, which is the proper bearer of truth and falsehood (idem; see chapter VI) . We may add that these terms also show an actuality/potentiality ambiguity, as we have to make a dis‑ tinction between the act of judging and the capacity to judge . Furthermore, Russell’s term “statement” may stand for the act of stating, the statement made, and the stated content, as we have seen above . Third, Russell’s explanation of judgement presupposes that the act of judging is an act of unifying . When Othello judges that Desdemona loves Cassio, the objects Othello, Desdemona, love and Cassio are unified by the act of judging . Furthermore, if the judgement is true, there exists another complex unity that of Desdemona loving Cassio . As Russell says: “[I]f Othello believes truly that Desdemona loves Cassio, then there is a complex unity, … , when a belief is false, there is no such complex unity” (Russell 1912, p . 74) . This point captures, Twardowski argues, what is meant by one of the Aristotelian explanations of truth: When Aristotle says that “he who thinks the separated to be separated and the com‑ bined to be combined has the truth,” he assumes that something combined or separat‑ ed exists (Twardowski 1925, p . 199) . For Aristotle and Russell, the act of unifying the constituents into a com‑ plex whole is essential to the act of judging, whereas this is not the case for Twardowski . Whether I think of something by means of a single concept or a combination of con‑ cepts is irrelevant to the essence of judgement . What is crucial here is simply the affirmation or denial of the existence of the represented object (p . 201) . The essence of judging consists in either affirming or denying the exis‑ tence of the object . Twardowski need not invoke here his more complicated account of judgement in which a distinction is made between existential and relational judgements . The A may be complex or relatively simple; the content of the judgement is existence or actuality; and the object of judge‑ ment “may be a person, a thing, a phenomenon, or – most importantly – a relation” (p . 202) . Because a combination of concepts is thus inessential to Twardowski’s elucidation of judging, a reformulation of the correspondence theory of truth is demanded . I come back to this in chapter VI . 116 Maria van der Schaar

Before this can be done, Twardowski needs to answer a possible critique that might be raised by Russell against his theory . If, as Russell thinks, judging is a relation, and the act may have only one simple object, it seems that we cannot judge falsely . For if the judgement were false, there could be no relation because one of the terms does not exist (pp . 203-204) . Russell’s objection not only presupposes that the act of judging is a relation; it also presupposes that a relation can exist only if the terms of the relation exist, and both theses are denied by Twardowski . It may also be argued, Twardowski continues, that on his Brentanist account, there is always a combination of the concept of an object with the concept of existence (pp . 204-205) . Twardowski answers in the first place that merely combining two concepts does not yield an act of judging . Com‑ bining the concept A with the concept of existence does not give an act of judging, for I might merely imagine that A exists . There is no judging unless there is assertion or rejection, affirmation or denial . Furthermore, existence is not a characteristic like extension, redness or weight: “Existence is [not] a characteristic of objects” (p . 205) . Existence is also not a relative char‑ acteristic of objects . “To exist” would mean then: “to be the object of a true affirmative judgement” (p .207) . These two phrases cannot by synonymous, though, for we can imagine an existing object that would never be the ob‑ ject of a true affirmative judgement . The only relative characteristic we can distinguish in an object of (a true affirmative) judgement is that it has judgedness (p . 207), and this needs to be distinguished from its existence, which is independent of our judging . “That “objectivity,” that “objective” validity, lies precisely in this independence of existence from our judging” (p .207) . I come back to Twardowski’s critique of Russell’s account of judgement, and its relation to his account of truth, in chapter VI, section 1 . 5.

KNOWING AND PREJUDICE. AN EDUCATIONAL MISSION

1. Some Conceptual Distinctions

The former chapter is a prelude to the one on knowing and prejudice, as both notions are explained by Twardowski in terms of judgement . The standard explanation of knowledge as justified true belief seems not to have judge‑ ment as part of its explanation, but as we have seen in the former chapter, the term “belief” needs to be disambiguated . The two notions that seem to play a role in the standard account of knowledge as a kind of belief are the notions of conviction and of disposition to judge, at least, if knowledge is understood as a disposition, a potentiality (a habitus, see chapter IV) . It is well known that there is an important problem with the standard explanation of knowledge, as we may create Gettier cases . We can imagine a case in which a belief is justified and true, whereas we would not call it knowledge . If someone looks at a clock, reads that it is three o’clock, and the time is also three o’clock, his judgement seems to be true and justified, and would normally count as knowledge . The Gettier case is created by adding to the example the perspective of a third person, who knows that the clock is defect . From this third person perspective, the person’s judgement can no longer be understood as knowledge . I will argue in this chapter that Twardowski’s account of knowledge is able to prevent such problems . Just as we have to make a distinction between the act of judging and the judgement made, we have to make a distinction between the act of knowing and knowledge as product . Examples of acts of knowing are acts of perceiving resulting in a perception made: an act of insight or understanding resulting in a certain insight, and an act of demonstrating resulting in a conclusion, a knowledge product . Sometimes an act of knowing is acknowledged, while the act is explained as an act of getting to know, an act through which we acquire knowledge (as product) . On such an account, knowledge is prior in the order of explanation to the cognitive act . Twardowski rightly saw that 118 Maria van der Schaar the product is the result of the act, and that the act is thus the notion that needs to be explained first . Second, we have to make a distinction between the act of knowing and knowledge as disposition or habitus; we may have knowledge, although we are sleeping . Knowledge as disposition is not an act; knowledge is a po‑ tentiality that can be actualised in acts of judging, which may be accom‑ panied by a state of conviction . The distinction is made by Brentano in his Psychologie vom empirischen Standpunkt . We may possess a storehouse of obtained cognitions (Erkenntnisse) without thinking of them . Such knowl‑ edge has to be understood as a disposition to make certain acts, not as itself being an act of knowing (ein Erkennen; cf . (Brentano 1874, p . 144)) . This may also be called the distinction between knowing and having knowledge . Third, knowledge is often contrasted with opinion . Whereas knowledge is always the result of a true judgement, opinions may be true or false . If someone utters a declarative sentence, one is entitled to ask “How do you know that?” . If no answer is given, we may draw the conclusion that only an intimation of a subjective opinion is given, and that the declara‑ tive sentence was not uttered with full assertive force . If the speaker has presented his subjective opinion as though it is valid for all, that is, if the declarative sentence is put forward with full assertive force, we may say that his opinion is a prejudice . Because the speaker has no ground for his assertion, he is not entitled to make the judgement . Prejudice is an opin‑ ion without judgement, as Voltaire put it: Le préjugé est une opinion sans jugement . We may call the prejudice a “judgement” but only in a deprived, modified sense of the term . Finally, we have to make a distinction between knowing an object and having knowledge about something, on the one hand, and knowing that … and having knowledge that …, on the other hand . The distinction is diffi‑ cult to formulate in English; most European languages make a distinction between what is called kennen and wissen in German, and in French con‑ naître and savoir . Seeing John is an example of knowing an object, and to know Paris is to have knowledge about Paris, to have become acquainted with Paris . Perceiving that John is wearing a red jacket is an example of knowing that … as an act, and knowledge that 7 + 5 = 12 is an example of having knowledge that so and so . The latter is sometimes called “proposi‑ tional knowledge”, but this is an inappropriate formulation, just as speaking of knowledge as a “propositional attitude” is confusing . These phrases seem to imply that we know a proposition, but that cannot be true . Knowledge that snow is white is not identical with the apprehension of the proposi‑ tional object that snow is white; such conceptual knowledge is not what is at stake . What we know is that a certain proposition is true . We may call Knowing and Prejudice 119 it therefore judgemental knowing and knowledge to be distinguished from knowing and knowledge of objects .

2. Brentano and Bolzano on Knowledge

The historical background for Twardowski’s explanation of knowledge is given by the accounts one can find in Brentano and Bolzano . Although the two accounts are opposed to each other in an important sense, they agree insofar as in neither Gettier cases can be created . The same is true for Twardowski’s explanation of knowledge, as I will argue . It is therefore of interest to see in what sense Twardowski’s account of knowledge differs from the standard explanation, and to what extent it is influenced by the explanation of knowledge in the writings of Brentano and Bolzano . Twardowski’s earliest position in the theory of knowledge is influenced mainly by Brentano . Brentano explains knowledge as the right judgement (“Das gute oder richtige Urteil ist Erkenntnis”; Brentano 1956, p . 2) . Al‑ though neither Brentano nor Twardowski make the act/product distinction at that time, the term “judgement” is best understood as referring to the judge‑ ment product; Erkenntnis is to be translated as “cognition” not as “act of cognizing .” Furthermore, the term “right judgement” in Brentano’s writings is used in a strict and in a lax sense: it may either mean a true judgement in the broad sense in which a blind judgement may be called true, although it is not justified, or it may mean a judgement that is logically as it should be (logisch berechtigt) . In the latter sense, the right judgement is opposed to any blind judgement, whether true or false (Brentano 1889a, pp . 20-21); has a certain inner rightness (gewisse innere Richtigkeit; cf . (Brentano 1889a, p .13)) . In order to count as knowledge, a judgement must not merely be true; it also has to be justifiably certain, that is, it has to be evident 1. Knowledge for Brentano is the evident judgement, that is, the judgement that is characterised by a certain inner rightness 2. The same point can be found in Twardowski’s logic manuscript: Knowledge is a true judgement that is judged with evidence (in opposition to a judge‑ ment that is true but blind, because it is judged without evidence) 3.

1 [U]m Erkenntnis zu sein, muß ein Urteil nicht bloβ wahr, sondern auch sicher sein. Mit ‘si‑ cher’ ist … gemeint, daß der Urteilende … mit Flug und recht seiner Sache gewiβ ist (Brentano 1925, p . 159) . On the distinction between the two meanings of “correctness” in Brentano, and the relation between the notions correctness and evidence . Cf . (Schaar 2002/2003) . 2 If Brentano would make a distinction between the act of judgement and the judgement product, he would says that the inner rightness pertains to the act of judgement, which results in the evident judgement . 3 Unter Erkenntnis verstehen wir ein wahres, mit Evidenz gefälltes Urteil (im Gegensatz zu den wahren, ohne Evidenz gefällten, blinden) (Twardowski 1894/1895, p . 209) . 120 Maria van der Schaar

The being evident of the judgement is infallible, according to Brentano . He thus aims to give an explanation of the Cartesian concept of knowledge: knowledge has to be infallibly certain . Twardowski’s account of knowledge as conforming to the standard explanation: knowledge is a judgement, which is evident, that is, justified, and thereby true . This does not mean that Gettier cases can be formulated on his account, for the notion of being evident cannot be identified with the notion of being justified as used in modern epistemology and the Gettier cases . For Brentano, if a judgement is evident, its truth is guaranteed . So‑called justifications in the Gettier cases are not able to make the judgement evident, Brentano would argue . The per‑ ception of a clock that is not working is not able to make one’s judgement evident or justified . The price Brentano has to pay for this seems to be too high, though, for we are never able to determine whether our judgement is infallibly evident . From 1911 on, Twardowski’s account of knowledge is influenced by Bol‑ zano, as we will see below . According to Bolzano (1837, I, §36), cognition (Erkenntnis) is a judgement that contains a true Satz an sich . Bolzano gives as examples of cognition: having cognition (Erkenntnis haben), the act of understanding (einsehen), and the act of perceiving (wahrnemen) . The term “cognition” can thus refer to the act of judging and to the product of the act, although Bolzano does not address the act / product distinction here . As the judgement product, for Bolzano, does not exist independently of the act, he considers this explanation of cognition too narrow, when he takes up the topic in the third part of the Wissenschaftslehre (Bolzano 1837, III, §307) . Anyone who is at the present moment not judging a certain truth, cannot be attributed cognition according to the explanation given in part I of the Wissenschaftslehre . We need an explanation of cognition as a “continuous state of the mind;” we need an explanation of knowledge as a disposition to judge, knowledge as habitus . According to Bolzano (Bolzano 1837, III, §307, p . 207), A’s cognition that a Satz an sich M is true is a “state” (Zustand) of the mind of an agent A, if and only if: (a) A has once passed the judgement J that M is true; (b) the Satz an sich M is true; (c) A is able to remember the judgement J; (d) A still holds M to be true . A holds true (being consistently committed to, fortdauernd zugetan sein; cf . (Bolzano 1837, III, §306, p .200) a certain Satz an sich M, consisting of a subject presentation S and a predicate presentation P, if and only if:

(a´) A has once passed the judgement J that M is true; Knowing and Prejudice 121

(b´) A is able to remember the judgement J; (c´) each time the presentations S and P, or the question whether S is P, comes to A’s mind, A passes the judgement J . Cognition is thus a special case of holding true, namely one where the Satz an sich that is held true, is actually true . In contrast to Brentano’s explanation, there is no Cartesian demand that cognition should be infallibly certain . Mark Siebel (1999) gives a slightly different analysis of holding true. According to Siebel, the first and the second clause are not part of Bolzano’s explana‑ tion . The context of the explanation of holding true, though, presupposes that we are speaking of a judgement once passed and now remembered . I therefore take the freedom to add these points as part of the explanation . The extra clauses are needed in order to distinguish the notion of holding true, that is, a belief as habitus, from the notion of a capacity, a dunamis, to form a new judgement (see chapter IV) . There are two important differences between Bolzano’s account of cog‑ nition and the standard account of knowledge as justified true belief . For Bolzano, a state of cognition can obtain only if the knower has at least once made the judgement . If we never have made the judgement, we can only speak of a capacity to form a new judgement, or of an implicit, unquestioned presupposition, and neither of these notions is discussed here . On Bolzano’s explanation, the act through which we acquire knowledge is accounted for by (a) . Second, Bolzano does not explain cognition in terms of justification . Bolzano’s notion of cognitive ground seems to be Bolzano’s alternative to the modern notion of justification . Why is the notion of cognitive ground not part of his account of cognition? The cognitive ground of a judgement is a collection of judgements made, which need not contain a true Satz an sich . It is also called a cognitive cause, as these judgements made can be understood as the cause of the concluding cognition . For Bolzano, cogni‑ tions may be mediated or they may be immediate judgements, and only the first have a cognitive ground . So, a cognitive ground is not essential to every cognition (§313) . Furthermore, the cognitive ground is a psycholog‑ ical notion, a cause, and it may or may not have the objective ground of the conclusion as its content . It is for these reasons that Bolzano does not take the notion of cognitive ground to be part of the explanation of cognition . The latter may be the result of pure luck, and it may be mediated by false judgements (§314) . In this sense, Bolzano’s account of cognition is far apart from Brentano’s idea of knowledge as the evident judgement .4 Because the

4 The fascinating notion of objective ground in Bolzano does not play a role in Twardowski’s writings . The objective ground of a Satz an sich is the reason why the latter is true . If we know the objective ground of a, we understand why a is true . Understanding (Begreifen / Einsehen) is thus a stronger notion than knowing . The act of understanding results in a clear insight 122 Maria van der Schaar notion of justification is absent from his explanation of cognition, no Gettier cases can be formulated for Bolzano’s account .

3. Knowledge, Science and the Cognitive Act

In the paper on actions and products from 1912, Twardowski distinguishes two meanings of the term “cognition:” the act of cognition and the cogni‑ tion product . The term “cognizing” is used for the cognitive act, that is, the act of knowing, and the term “cognition” is primarily used for the product (Twardowski 1912b, p . 113) . As cognition is a purely mental product, it is non‑enduring . In the paper on actions and products, Twardowski does not say anything about the philosophical meaning of the thesis that cognition is to be understood as secondary in the order of explanation to the act of cognizing . Neither does he say anything on what the cognitive act is . In the summer trimester of 1925 Twardowski gave a lecture course on the theory of knowledge . The manuscript has survived for the greater part, and was edited by Izydora Dąmbska in 1973, together with her notes from the first two lectures 5. Twardowski starts his lecture course with some conceptual clarifications . He draws our attention to the prefix “co‑” (“er‑” in the German “erkennen,” “po‑” in the Polish “poznać”) . The meaning of it is generally to start doing something, to enter in a particular state . “To co‑gnize = to enter into a state in which I know something” (Twardowski 1925, p .184) . This explanation leads one to the question what “to know” (znać = kennen) means: “ “To know” = “to have knowledge [wiedzieć = wissen]” ” (idem) . The central epistemic question for Twardowski is thus: What is to have knowledge? From the explanation one might draw the conclusion that for Twardowski the notion of having knowledge is prior in the order of expla‑ nation to that of cognitive act . I do not think, though, that these elucidations are explanations of definitions; they are intended to show how these notions are related . As we will see below, Twardowski explains having knowledge in terms of an act once passed . The way he elucidates these notions here is of importance for the distinction between the psychology of cognition and the theory of knowledge . According to Twardowski, the psychology of cognition is concerned with the mental function of passing a judgement . It investigates the conditions under which the passing of correct and incorrect judgements occurs . Only the theory of knowledge is concerned with the dis‑ tinction between true and false judgements, and this distinction pertains to

(cf . Bolzano 1837, III, § 316) . Bolzano also has a concept of Wissen or certain knowledge, knowledge with a high degree of confidence . Cf . Schaar (2007) . 5 The text is published in Polish in 1975, and published in English in 1999; cf . (Twardowski 1925) . Knowing and Prejudice 123 the judgement product rather than to the act of judging . Twardowski explains cognition as product of the correct judgement, whereas the act of cognizing is an act of judging resulting in the correct judgement . On Twardowski’s account, the theory of knowledge abstracts from the activity of judging, focussing on true judgement products . To come back to Twardowski’s central question, what does having knowl‑ edge mean? To have knowledge can be understood as the capacity to make a correct judgement . This definition Twardowski considers too broad . If someone asks me “How much money do you have in your pocket?”, I may not know the answer at this moment, although I have the capacity to make a correct judgement by counting the money . We would not call this knowl‑ edge, although there is the capacity to make a correct judgement . Essential to having knowledge is that I have once made the correct judgement . In this case, if someone asks me how much money is in my pocket, I will be able to give an answer on the basis of memory . This leads Twardowski to an explanation of knowledge that has a strong similarity with the explana‑ tion of knowledge given in the third part of Bolzano’s Wissenschaftslehre . According to Twardowski, a person P has the knowledge that J, if and only if (cf . (Schaar 2009)): (a) P has once passed a judgement resulting in the judgement J; (b) the judgement product J is true; (c) P is able to remember the judgement J; (d) P has the capacity / the disposition for making correct judgements similar to J . Although the notion of judgemental act or cognitive act is not completely absent in this explanation, as there cannot be a judgement product without a judgemental act, it is rather the judgement product J that plays a central role in each of the conditions . What is clearly absent in the explanation is a notion like justification . Brentano’s idea that knowledge is the evident judgement is absent from the explanation, and no alternative is given, apart from the idea that the judgement has to be true . Compared with the standard explanation of knowledge, condition (c) strikes one as new: without the pos‑ sibility to remember the first judgement, there is no knowledge . What is it precisely that we have to remember? Do we have to remember that we have counted the money? The fact that we have counted the money does not seem essential to Twardowski . He might argue that the act of judgement leaves a certain trace in our memory system, by which we are able to remember that the judgement is made . Twardowski is not explicit on this point, though . Because of the distinction between actions and products, the notion of cognitive act plays a role in Twardowski’s epistemology . This role is small, though, for in order to prevent a psychologism in epistemology, he argues 124 Maria van der Schaar that the cognitive act is not relevant for epistemology . I think, though, that logic and epistemology are in need of the notion of cognitive act, and I will give an example below . On the account of knowledge Twardowski has given in 1912 and in 1925, Gettier cases cannot be formulated, because the notion of justification is not part of his explanation of cognition . For Twardows‑ ki, a cognitive act is an act of judgement resulting in a correct judgement, where correctness is not understood in epistemic terms . As we will see in the next chapter, Twardowski explains truth in terms of correspondence . Would it be possible to strengthen Twardowski’s concept of cognition in such a way that a notion of justification is present without invoking Gettier problems? I think this would be possible by demanding of the cognitive act that it is cognitive in the sense that it is an act of perceiving, an act of un‑ derstanding, or an act of inference on the basis of other judgements made, that is, an act of demonstrating . In all three cases the judgemental act results in a judgement made that may count as knowledge, and not merely as a cor‑ rect judgement . For the judgement is made evident by an act of perceiving, an act of understanding, or an act of demonstrating . In the latter case, the judgements that function as the premises for the conclusion, may be called the justification for our knowledge . It is clear that in the other two cases, the perceiving and the understanding, the judgement made, the perception or the insight, is also the result of an act of cognizing in the strict sense, and may therefore count as knowledge, too . In all three cases the judgement is made evident by a cognitive act, and thereby counts as knowledge . Although the evidence of the judgement should not be understood in Brentano’s sense of infallible evidence, it is a stronger notion of justification than on most modern accounts of the notion . If we learn that our former act of perceiving that there is a lady on the stairs is an illusion, because we see now that it is a wax figure, we can no longer consider the act to be a cognitive act . We can no longer say that we have perceived that there is a lady on the stairs, only that it seemed to us so, then . So, we can no longer call it knowledge . There is thus an internal relation between the cognitive act and knowledge as product, just as there is between actions and products in general . What is called a justification in the Gettier cases, would not count as a justifica‑ tion on this account . If our conclusion is based on false premises, the act of inference is not a cognitive act . It is thus that epistemology may use the notion of the cognitive act 6. According to Twardowski, the task of a theory of knowledge is to in‑ vestigate the truth of the first premises of the special sciences . The first principles of the sciences cannot be proved . Science presupposes the truth of its first premises, and the task of philosophy is to investigate their truth .

6 How this account may be used to give an answer to the problem created by the Gettier cases, is elaborated in (Schaar 2011) . Knowing and Prejudice 125

One of these premises is the Principle of Causality: whatever happens has its cause, whose content is investigated by metaphysics . The theory of knowledge asks the question on what basis the principle is judged to be true . What are the necessary and sufficient conditions that must be satisfied by the ultimate presuppositions in order that we are entitled to judge them to be true? By distinguishing different groups of “primitive judgements” we are able to give those conditions for these different first principles (Twardowski 1925, pp . 188-189) . In the discussion of the distinction between the a priori and the a poste‑ riori sciences Twardowski comes to speak of first premises and grounding . In opposition to J S. . Mill, Twardowski claims that there is a fundamental distinction between a priori sciences and a posteriori sciences . Sometimes it is said that their difference consists in the fact that the former are based on reason, while the latter are based on experience . This formulation is ambiguous, for what is it precisely that is based on reason and experience? Twardowski agrees with Aristotle, Locke and Hume that all our concepts are based on experience, that is, that they have their origin in experience, so this idea cannot be used to give a foundation to the distinction between the two sciences . Does “being based on” mean that assertions and judgements have their origin in, respectively, reason and experience, that we discover the truths of these sciences in different ways? According to Twardowski, this is not what distinguishes the two kinds of sciences (Twardowski 1923, p . 173) . The distinction lies in the way we ground our assertions . The a priori sciences “ground their assertions without resorting to experience, while the [a posteriori sciences] ground theirs by appealing to experience” (p . 175) . By grounding these assertions we “show that … [they] are true, or at least sufficiently plausible to be accepted until more plausible ones are arrived at” (p . 174) . Relating the distinction to the question of the first principles, Twardowski claims that the distinction between the two kinds of sciences consists in the fact that a priori sciences ground their judgements on axioms, together with definitions and postulates . Such axioms are the result of insight, and they are in this sense not empirical . Whereas the first principles of the a posteriori sciences are perceptual judgements (Twardowski 1923, p .176), while the a priori sciences never appeal to such perceptual judgements . When Twardowski addresses the question of the possibility of knowl‑ edge, he comes to speak of a problem well known at the time . There has been a discussion between Neo‑Kantians, such as Leonard Nelson, and phenomenologists concerning the possibility of knowledge . According to Nelson, a criterion of truth should exclude the possibility of error . Because such a criterion cannot be given, a theory of knowledge is impossible . Nelson formulates the problem of the criterion as follows . Suppose we had a criterion by which we can distinguish a judgement that is knowledge from 126 Maria van der Schaar a judgement that is not . This criterion must be either known or not known . If it is known, it should be asked on the basis of what criterion we can say that we know it, and we have returned to our original problem . If it is not known, we have to make the criterion an object of knowledge . In order to determine whether our object of judgement, the criterion, is an object of knowledge, we need to apply the criterion . That is, the question whether the object of judgement is an object of knowledge must already be decid‑ ed (Nelson 1908, p .92) . Nelson’s critique concerns Meinong’s position in Über die Erfahrungsgundlagen unseres Wissens that was published in 1906, and Nelson’s book got a 60 page review from Höfler in his Erkennt‑ nisprobleme und Erkenntnistheorie from 1910 . Höfler criticises Nelson’s psychological interpretation of the a priori . A theory of knowledge, Höfler says, does not concern the psychological analysis of actual knowledge‑pro‑ cesses . It rather concerns the understanding of the meaning of such words as “Erkennen” and “Erkanntes,” “cognizing” and “cognition,” and their a priori relations, a point well taken by Twardowski . Höfler’s solution to the problem of the criterion in terms of evidence is no longer defended by Twardowski in his later years . It is clear, though, that he still considers the problem to be of importance: “The problem of the criterion of truth is one of the most important problems for a theory of knowledge” (Twardowski 1925, p . 189) . Twardowski’s answer to the problem consists in admitting that we have to acknowledge certain principles of reasoning and first premises, primitive judgements, as well . It is precisely here that the sciences are in need of philosophy . Twardowski is not a Neo-Kantian for whom questions of knowledge are prior to all other philosophical questions . Twardowski’s main philosophical question is the question of truth . Before we ask what knowledge is, we have to answer the question what truth is, according to Twardowski . And a definition of truth is not to be confused with a truth criterion (p . 239) . As Twardowski explains knowledge in terms of truth, it seems that the question of truth is the central question for a theory of knowledge . It is for this reason that Twardowski spends the greater part of his epistemology lectures on the question of truth, and I come back to the topic in the next chapter .

4. Prejudice and the Critical Mind

The Polish Philosophical Society, founded by Twardowski in Lwów in 1904, was a society of professional philosophers, and is characterised by its exact method: “the quest for the greatest possible precision and exact‑ ness in thinking and in the expression of what is thought,” as Twardowski puts it (Twardowski 1926, p . 28) . The society has “inscribed on its banner Knowing and Prejudice 127 the slogan of scientific criticism and critical science” (Twardowski 1929, p .276) . It is dedicated to the search of truth . For Twardowski, the search of truth is also a pedagogic ideal . In what is called “Twardowski’s testament” (Twardowski 1999, Introduction, p . 13), which contains the text of the lec‑ ture “On the Dignity of the University” from 1933, Twardowski writes that the search of truth has an ethical significance . It brings moderation to dis‑ putes, and it gives tools for reconciliation between people . The university is not one among many kind of schools, for it teaches scientific thinking as a mode of thinking that leads to scientific knowledge . A presupposition for scientific learning is an unconditional intellectual independence . Let me give Twardowski’s own words on these matters . Those “who fund and support [the University] would display a complete misunderstanding of the essence of the University if they tried to impede its research in any given direction, to guard against any of its findings, or to point it in the direction of desired results” (Twardowski 1933, p . 279) . These are not empty state‑ ments in the year 1933 . Twardowski was not an intellectual looking away from the political disturbances at the time: These ways of conceiving the University’s dignity and autonomy in no way deprive it of influence on life itself . They only deprive it of direct and causal influence on current affairs, but assure it instead a different, much more enduring and deeper in‑ fluence (p . 280) . Through the instruction of youth and the publication of scholarly works: [T]he University shines onto the whole of society, and broadens perspectives and convictions that none can impose on another as dogma, perspectives and convictions whose power resides solely in their scientific grounding (idem) . The students should learn to approach all problems “in a careful and critical manner and will strive for their objective analysis … instead of suc‑ cumbing to charismatic slogans and falling into the nets of various fishers of the soul .” (p .281) . This certainly is addressed to philosophical education in particular, and some of the problems in philosophy departments appar‑ ently have not changed . By making truth itself an object of investigation, philosophy has a central position among the sciences (p . 284) . Philosophy is to awaken in us a critical spirit, demanding “clarity and rigor wherever scientific arguments are invoked” (idem) . Scientific knowledge can thus only be obtained if we use a proper method . The theory of knowledge has to show, on the one hand, that scientific knowledge and truth is possible, and, on the other hand, it has to show in what way we can obtain such knowledge . It is precisely the environment of the university that allows for obtaining such knowledge, and to show the student how to guard oneself against uncritically accepting unquestioned dogmas . The meaning of science 128 Maria van der Schaar and the university is not only to bring society knowledge, it also cultivates the ideal of independent thought . Success in this area will not only result in people being better educated but also in them being better by being less prejudiced, and more mindful of the truth (Twardow‑ ski 1906b, p . 90) . Twardowski’s lecture on prejudices, delivered for the university com‑ munity in 1906, may be understood in this context . Part of our way towards scientific knowledge is a fight against prejudices . For these are not only present in daily life, but also in science . According to Twardowski: [The fertile ground for prejudice is] ignorance and the lack of a critical mind . Vacui‑ ty, mechanical repetition and accepting what one hears … The aim is to teach people to think independently, reasonably, critically, and accustom them to be aware whether their convictions are justified or not . The only way leads through promoting science and education (Twardowski 1906a, p . 80) . So, understanding what prejudice is, is an important part of the scientific pedagogic ideal Twardowski endorses . A prejudice, according to Twardowski, is a “preconceived, unjustified and erroneous conviction” (p . 75) . Twardowski does not distinguish in the lecture between the act of judgement and the state of conviction . We may understand a prejudice as a potentiality to make a particular judgement, which is accompanied by a strong degree of conviction . The guess of a doc‑ tor who judges that the patient will not recover, although he cannot give a justification for his judgement, need not be a prejudice . When the doctor has based his judgement on some empirical data, the English language gives one the opportunity to speak of an educated guess . The judgement may be unjustified and erroneous, but it is not “preconceived ”. Furthermore, the doctor will, hopefully, be open to evidence that speaks against his judge‑ ment, which is not the case with a prejudice . According to Twardowski, not every preconceived judgement is a prejudice, for the judgement need also to be erroneous to count as such . Columbus was convinced that he would reach dry land if he continued to sail West . Although his contemporaries be‑ lieved this to be a prejudice, we now know it was not so, Twardowski argues (p .75) . Only with the benefit of hindsight are we able to judge whether it is a prejudice or not in these kind of cases . We thus need a wider context to determine whether an opinion is a prejudice . If it turns out that people do not change their opinion when they are confronted with clear counter‑evi‑ dence, if they lack the critical attitude towards their opinion that is central to an intellectual education, and if they are not able to give a justification, we consider it to be a prejudice, but we might be mistaken in this, for the opinion may turn out to be true . 6.

TRUTH AND TIME. TWARDOWSKI’S IMPACT ON HIS STUDENTS

1. The Correspondence Definition of Truth

When Alfred Tarski presents his semantic concept of truth, he aims at a defi‑ nition of truth in conformity with “the classical Aristotelian conception of truth,” chosing the formulation of Metaphysics Γ, 1011b26‑27: To say of what is that it is not, or of what is not that it is, is false, while to say of what is that it is, or of what is not that is not, is true (Tarski 1944, p . 343) . Any definition of truth has to capture the intuition expressed in this for‑ mula, Tarski says . The formula cannot function as a definition of truth, as it is not precise enough, although it is more useful than most modern variants of it that use the vague term “correspondence .” According to Tarski, the laws of contradiction and of excluded middle can be deduced from the formula, and from his own definition of truth of an interpreted sentence: a sentence A is true if and only if A is satisfied by every infinite sequence of objects . This formulation expresses a concept of absolute truth in the sense that the truth of a sentence is not restricted to a certain domain (Tarski 1956, p . 199) . For Tarski, truth is not relative to circumstances . If A is a variable free sentence of a formalized interpreted language, either all sequences satisfy A, or no sequence satisfies A . This means that “every sentence is true or false under all circumstances … Thus Tarski defines an absolute concept of truth” (Woleński and Simons 1989, p. 423). Any truth definition should not only be formally correct in the sense that the metalanguage in which the definition is given is richer than the object language; it is also to be materially adequate: since “its consequences [should] include all those required by this convention [T]” (Tarski 1936, p . 195) . Apart from the distinction between object‑ and metalanguage, which comes from his teacher Leśniewski, and the idea of convention T, Tarski’s 130 Maria van der Schaar ideas mentioned above can be brought back to Twardowski . The fact that Tarski defends an absolute notion of truth, from which the (meta)logical principles can be derived (cf . (Tarski 1944, p . 354)), that the bearer of truth and falsity is an interpreted sentence, and that Aristotle’s weak version of the correspondence definition of truth is a key to the definition of truth can all be brought back to Twardowski’s theory of truth and truthbearers . Because he distinguishes between object‑ and metalanguage, Tarski makes a distinction between “semantical laws,” given in terms of “true” and “false,” and the logical laws of contradiction and excluded middle, a distinction not acknowledged by Twardowski . It is these semantical laws that can be deduced from the definition of truth, Tarski argues . We may start our analysis of Twardowski’s account of truth by pointing to a difference with Tarski . Whereas for Tarski truth pertains to sentences, Twardowski argues for the absoluteness of truth by claiming that the proper bearer of truth is not the sentence, but the judgement product . Tarski’s sen‑ tences are interpreted sentences, and truth is thus a property of sentences together with their meaning; formal languages, for which the truth‑defini‑ tion is given, are already interpreted . Artur Rojszczak has given an account of the different explanations Tarski has given of the sentence, one of them being that a sentence is the product of human activity . Tarski clearly uses here Twardowski’s distinction between actions and products 1. In general, Tarski’s thesis that sentences are the primary bearers of truth and falsity can be understood as a linguistic variant of Twardowski’s thesis that the judgement product is the proper bearer of truth and falsity . Although Tarski may be understood as having sympathies with the nominalism of his super‑ visor Leśniewski – neither of them would endorse Twardowski’s defence of general objects, the sign is for both an interpreted sign. Leśniewski’s in‑ terest in formal logic does not make him a formalist in Hilbert’s sense; the axioms are insights, not arbitrary starting points for a formal system . The theses and axioms of logic have intuitive validity (Leśniewski 1929, p. 78). And the young Tarski follows his former supervisor in this respect . Tarski’s thesis that the sentence is the bearer of truth and falsity is of importance for his thesis that truth is a semantic notion . Truth relates the interpreted sentence to the world, and is thereby a semantic notion . Notwithstanding the fact that Twardowski considers the judgement product to be the proper bearer of truth and falsity, while Tarski takes sentences to have this role, both understand truth and falsity to be independent of the occasion at which the bearer is used . Truth is not relative to context or circumstances . Simi‑ lar to Tarski, both Łukasiewicz and Leśniewski have presented a linguistic

1 “Normally expressions are regarded as the products of human activity (or classes of such products)” (Tarski 1933, p. 174); cf. (Rojszczak 2005, p. 206). Cf. (Woleński 2009, p. 50): “Leśniewski and Tarski both considered language to arise from acts of using expressions.” Truth and Time 131 variant of Twardowski’s thesis, the interpreted sentence is the proper bear‑ er of truth and falsity, and they have thereby mediated Twardowski’s idea that the judgement product is the bearer of truth and falsity (see section 2) . The greater part of Twardowski’s course on the theory of knowledge from 1925 is concerned with judgement and truth rather than with justification, and is in this respect different from a modern approach to the problem . As already noted in the former chapter, according to Twardowski, the definition of truth presupposes a particular perspective on the essence of judgment (Twardowski 1925, p . 194) . Often, the explanation of truth in terms of cor‑ respondence underlies a wrong idea of the bearer of truth and falsity . If it is said that something is true if it corresponds or agrees with its object, one often assumes that the bearer of truth is an idea or image, which may or may not give a correct representation of its object . The relation between a repre‑ sentation and its object “is a relation of faithfulness, not of truth,” on this account, and can therefore not be used as an explanation of truth (p . 194) . This interpretation of the correspondence definition of truth disappears as soon as one understands that the proper bearer of truth and falsity is not an idea, but a judgement . Some of Aristotle’s formulations of the definition of truth presuppose a special explanation of judgement . Besides the Aristotelian definition of truth given by Tarski, which speaks of correspondence only in a weak sense, there is also a definition of truth in book Θ of the Metaphysics, in which correspondence is to be understood in a stronger sense . In this definition the Aristotelian notion of judgement is presupposed: “he who thinks that what is divided is divided, or that what is united is united, is right; while he whose thought is contrary to the real condition of the objects is in error” (Metaphysics Θ, 1051 b 3‑5, translation by H . Tredennick) 2. In order for there to be truth, one has to divide or combine, and judging is understood as a combining or dividing . This is anathema to the Brentanist account of judgement, and Twardowski does not allow for an explanation of judgement in terms of a combination or division of ideas . In the judgement “Unicorns do not exist” only one idea is involved, that of unicorn; one denies the existence of unicorns, and existence is not an idea or a predicate . The fact that a judgement sometimes seems to be complex is not due to the fact that judging itself is a combining of ideas; the act of presentation underlying the judgement already presents a complex object . The judgement merely affirms or denies the existence of the object, whether simple or complex .

2 The distinction between the two definitions of truth in Aristotle is clearly put forward by Twardowski in his lectures on knowledge and truth . The definition in which the notion of weaker correspondence plays a role is of importance for Twardowski’s own definition of truth. Jan Woleński and have introduced the terminology of “weak” and “strong” correspondence with respect to the two definitions in Aristotle . 132 Maria van der Schaar

What we assert is the existence of God, ghosts, thundering, raining, the love of Desdemona for Cassio, or the equality of the product of two times two and four (Twardowski 1925, p . 204) . It is for this reason that Twardowski gives the following definition of judgemental truth: An affirmative judgement is true if its object exists, a negative judgement, if its object does not exist . An affirmative judgement is false if its object does not exist; a negative judgement, if its object does not exist . (p . 208) . Like Tarski, Twardowski endorses the weaker formulation of the definition of truth, just as Brentano had done before him . Twardowski takes this definition to be a variant of the correspondence definition, where the correspondence obtains between the affirming or negating character of the judgement and the existence or nonexistence of the object of the judgement . An affirmative judgement is true precisely if it acknowledges the existence of an existing object, while a negative judgement is true precisely if it denies the existence of a non‑existent, merely conceived object . And: [A]n affirmative judgement is false when it acknowledges the existence of a non‑ex‑ istent object, or [a negative judgement is false when it] denies the existence of one that exists (p . 209) . According to Twardowski, correspondence does not consist in an agree‑ ment between the act of affirming and the existence of the object . As Bren‑ tano put it in his lecture on truth: “to correspond” does not mean to be similar; it rather means fitting, being in harmony, conforming to the object, or to the existence or non‑existence of the object (Brentano 1889b, p . 25) . Twardowski’s explanation of truth presupposes that there are existing and non‑existing objects, and is thus in conformity with his earlier theory of objects . The objects may also be relationships . In case of a relational judge‑ ment, the judgement is true, if the relationship subsists . The judgement that John loves Mary is true if the relationship of loving between John and Mary subsists, that is, if it is a fact that John loves Mary . Twardowski tries to stay within the Brentanist account of judgement, but these relational judgements are not existential judgements; they rather affirm that so and so is the case, and thereby have a propositional structure . Twardowski should have made a distinction here between objects and objectives, and between the existential “is” and the veridical “is” . We shall see below that he does acknowledge different meanings of “is”, but the just mentioned distinction is not to be found there . All this should have consequences for the way the correspon‑ dence theory of truth is formulated: existence of a (complex) object is not the same as subsistence of a state of affairs, as the former, but not the latter, presuppose the existence of its parts . Of the lectures on knowledge, in which Twardowski has developed his formulation of the correspondence definition of truth, five lectures (lecture 6 till 10) concern Russell’s account of truth in Truth and Time 133 his defence of the multiple relation theory of judgement . Russell’s account is of importance because it is a variant of the correspondence theory com‑ parable to the stronger definition of truth presented by Aristotle . Although Twardowski agrees with the idea that any proper definition of truth has to be given in terms of correspondence, he does not agree with the account of judgement that underlies Russell’s definition . Russell’s multiple relation theory of judgement, in which the act of judging is essentially an act of combining, makes it necessary to give a formulation of the definition along the lines of the stronger variant in Metaphysics Θ . The first point of critique Twardowski mentions is that Russell “speaks interchangeably of belief, statement, judging, and judgement” (p .198) . If we clearly distinguish between the sentence and the judgement, and between the activity of judging and judgement as product, we understand that truth and falsity are primarily characteristics of the judgement product . Second, Russell calls a judgement a relation . “[J]udging [is] an act which is sui ge‑ neris, rather than being a relation” (p . 204; more on this and the former point, see IV .5) . Third, and most important for the definition of truth, on Russell’s account of judgement, combining is essential to judgement . As for Brentano, for Twardowski it is possible to judge without any combination of objects or ideas . There are simple judgements of existence, such as “God exists” and impersonal judgements, such as “It rains,” in which we do not combine anything, although we do judge . For Brentano, and Twardowski follows him, existence is not a logical predicate; it cannot be treated as a characteristic of objects (p . 206) . The verb “is” may express different logical aspects in the judgement . For example, in the sentence “Two times two is four,” the verb “is” expresses (1) the relation of equality that two times two has to four; the relationship, a complex of the relation and its (non-existing) terms, is the object of the corresponding judgement; (2) existence on the level of the content of the judgement; and (3) affirmation, that is, judgemental force (p . 202) . The point of the ambiguity of being does not seem to form an argument against Russell’s position in 1912, though, but it is of interest for understanding the problem mentioned above that Twardowski does not distinguish between existential and veridical being . What precisely is exis‑ tence on the level of the content of judgement, and how is the existence of the content related to the subsistence of a relationship or state of affairs? Twardowski argues that existence is here not a logical predicate; apparently, it is an aspect of judgement that shows itself in the content of judgement, but at other places it is implied that objects may exist or not exist, while states of affairs may subsist or not, and it thus seems that existence is after all a characteristic of objects, and subsistence a characteristic of states of affairs . Is Twardowski on the side of Brentano, or on the side of Meinong’s later account, or does he oscillate between the two positions, as I fear he does? 134 Maria van der Schaar

Twardowski’s central argument against Russell’s definition of truth is that Russell conceives of judging as a combining, that is, as a relation . For Russell, it would be impossible to judge falsely if judging were not explained as a combining of several objects . For Russell, the existence of a relation presupposes the existence of its terms; because thinking and judging is a relation, we can only think and judge of what is . Falsity can thus only be explained as wrongly combining existing objects . When Oth‑ ello judges that Desdemona loved Cassio, his judgement is false, because there is in reality no complex unity of Desdemona loving Cassio, although Othello has combined Desdemona, love and Cassio in his judgement . The judgement relates these different objects, but there is no complex unity to which the judgement is related . As Desdemona does not love Cassio, there is no such complex . If the judgement were true, there would correspond to the judgement a complex unity in the world of Desdemona loving Cassio, and we can speak here of “correspondence” in a stronger sense of the term . In case the judgement is true, there is a certain agreement between the com‑ plex created by the act of judgement and the complex in the world, at least, the objects are related in the same order . In contrast to Russell, Twardowski argues that if a judgement is false, the judgement has an object, although that object does not exist . So, his account of falsity does not need the idea of wrongly combining existing objects . Essential to judging is not an act of combining, but an act of affirming or denying: “there is no judgement here until [there is] assertion or rejection, affirmation or denial” (p . 205) . It is the act of affirmation and denial that captures the essence of judgement . Russell and Twardowski share the thesis that the judgement is the bear‑ er of truth and falsity, although Twardowski would not agree with Russell that the act of judgement is such a bearer . How can Russell account for the fact that we may speak of truths that are not actually judged by anyone . Twardowski has a story to tell here, as we have seen in chapter IV, but he does not criticise Russell on this point . It is sometimes argued that the correspondence definition of truth cannot serve as a criterion of truth: we cannot compare our judgement with the object to which the judgement is to correspond, and we are therefore never able to determine whether our judgement is true . Twardowski agrees, but he denies that the point can be held against the correspondence definition, for we have to make a distinction between a definition and a criterion . We have here a problem for all definitions of relative characteristics . Truth is a relative characteristic of judgements, that is, an affirmative judgement is true because of something that lies outside the judgement; it is true if it acknowledges the existence of an existing object (p .210) . If we define relative characteristics, we are often confronted with the problem that the definition does not give us a criterion . The definition of the relative char‑ Truth and Time 135 acteristic of being a murderer, a man who has killed another man, does not give a direct criterion to determine whether someone is a murderer, for we cannot determine whether someone is a murderer simply by investigating that person . Those who give an alternative to the correspondence theory, such as the coherence and the pragmatist account, confuse a definition of truth with a criterion of truth, according to Twardowski 3. Another argument that is taken seriously by Twardowski is the point that the correspondence theory involves certain metaphysical assumptions (p . 211) . Twardowski’s formulation of the definition presupposes that the object of judgement may exist as something independent of us, and that existence has a certain meaning . It may be called “the theory of external, object‑oriented, transcendent correspondence” (p . 211) . Apparently, Twar‑ dowski does not consider this to be a counterargument to the correspondence definition of truth, and he seems to be willing to accept these metaphysical theses . Metaphysics as a general theory of objects is not at all to be ex‑ cluded from a scientific approach to philosophy . The fact that the concept of existence has for us its origin in our judgements as acknowledgements of existence does not exclude the thesis that objects exist independently of our judging them to exist .

2. The Absoluteness of Truth and the Logical Principles

Due to the fact that we are familiar with the idea that different logical sys‑ tems may be developed in accordance with different philosophical princi‑ ples, we are now used to make a distinction between a principle outside the logical system and a law that can be proved within the logical system, or that may function as an axiom within the system . Today these principles are formulated in metalogical terms, that is, in terms of truth and whatever is assumed to be the truthbearer, the proposition, judgement or sentence . In the work of Brentano and Twardowski, we see that what is called the “law of contradiction” and the “law of excluded middle” are formulated in terms of “judgement” and “truth ”. The question whether these laws are metalogical principles or laws within the system is not easy to answer . On the one hand, it is clear that these laws are formulated in terms of “truth” and “judgement”, which would imply that they are metalogical principles; on the other hand, these principles are considered to be judgements which are immediately evident, and they can be used to prove that other judge‑ ments are correct . They function thus as axioms within the logical system . Apparently, the laws do double duty, depending on function and formulation .

3 A greater part of the lectures contain a discussion and critique of the coherence and the pragmatist theory of truth, and of the Neo‑Kantian account of truth, as well . 136 Maria van der Schaar

It might be suggested that the question whether these principles function as a metalogical principle or as a logical law is an anachronism . The same question arises with respect to Frege’s work . Logical questions are always questions within the ideography as logic is the most universal science there is . One cannot step outside it; in essence, there is only one system . I am not completely sure, though, that the question can be put aside without further investigation in this chapter, as Łukasiewicz does make the distinction in his mature writings. Already in 1910 Łukasiewicz claims that these principles are not self‑evident and cannot be proved . At the same time, he argues that these principles have a practical value . He is thus not speaking about laws within the logical system . What are precisely the principles that play a role in the discussion? In accordance with Twardowski, I will formulate them in terms of truth and judgement, which means that we speak of logical, not of ontological prin‑ ciples, as Łukasiewicz is to put it. The strongest principle, underlying classical logic, is the principle of bivalence, BV: Every judgement is either true or false . It is also possible to formulate the principle of bivalence in a weak version, in which case it becomes trivial . I will therefore not use it: BVW: Every judgement is either true or not true . The principle of contradiction can be formulated in terms of contradictory judgements . For Twardowski, two judgements are contradictory if and only if one affirms what the other denies . NC: Of two contradictory judgements, at most one is true . NC´: Of two contradictory judgements, if one is true, the other is false . In a strong version: NCS: Of two contradictory judgements, at least one is false . The logical principle of excluded middle, in a weak version is:4 EM: Of two contradictory judgements, at most one is false . EM´: Of two contradictory judgements, if the one is false, then the other is true . It is sometimes said that NC together with EM gives BV . This is not the case, though, for NC and EM can be correct, even though there may be

4 Such a weak version of the principle of excluded middle is also defended in (Brentano 1956, p . 175) . The text on the principle in Lehre vom richtigen Urteil is from 1916; these formu‑ lations can also be found in Brentano’s manuscripts EL 4 and 25 . Cf . (Schaar 2002/2003) . Truth and Time 137 judgements that are neither true nor false . It is also possible to defend BV, while rejecting EM, as Leśniewski has done (see section 3). One may also defend the principle of excluded middle in a strong version: EMS: Of two contradictory judgements, one is true . EMS´: If a judgement is not false, then it is true . EMS´ together with the thesis that if a judgement is not true, then it is false gives BV . I have called EMS and EMS´ stronger principles than EM and EM´, because, as we will see, Kotarbiński defends the weaker, while rejecting the stronger principles . It is not at all clear that this is a correct terminology, though, for, as we will see, Leśniewski endorses BV and EMS´ but denies EM´ and EMS. From Leśniewski’s point of view, the principles should not be put together the way it is done here . Twardowski argues for the absoluteness of truth, truth is not relative to time or whatever circumstance, and for classical truth as well . Twardowski defends the classical notion of truth insofar as he defends the principles of bivalence, contradiction and excluded middle . Sometimes, the term “classical notion of truth” refers to the correspondence theory of truth . I am in need, though, of a separate ter‑ minology . As we will see below, Twardowski’s students, if they think that truth is definable at all, defend a version of the correspondence definition of truth, but they do not endorse all the logical principles Twardowski assumes to follow from the correspondence theory of truth. The early Kotarbiński and Łukasiewicz deny the absoluteness of truth with respect to time, and thus do not defend a classical notion of truth. Leśniewski defends the ab‑ soluteness of truth and the principle of bivalence, while rejecting classical truth in its full force . In the paper “On so‑called relative truths,” written in 1900, Twardow‑ ski explains an absolute truth as a judgement which is true always and ev‑ erywhere, while a relative truth is a truth that is true only under a certain condition . A truth is either a relative or an absolute truth . In the paper, Twardowski defends the thesis that all truth is absolute, understanding truth as a property of the primary truthbearer, the judgement made . A judgement J is a relative truth, if and only if J is true under certain circumstances, while J is not true under other circumstances . A truth of judgement J an absolute truth, if and only if J is not a relative truth 5. Twardowski directly opposes in this respect Brentano . For Brentano, the same judgement may change from true to false, if the relevant object chang‑

5 Compare also the later text: “By truth we mean a true judgement; thus absolute truth means an absolutely true judgement, or a judgement which is unconditionally true, or a judgement whose truthfulness does not depend on any changing conditions” (Twardowski 1923/1924, p . 239) . 138 Maria van der Schaar es .6 In fact, Twardowski already made the anti‑Brentanian claim in the logic manuscript from 1895/96: “[T]rue sentences [zdania] cannot become false after some time” (Twardowski 1895/96, p .54) . For a logical realist like Bolzano it is not difficult to provide an answer to those defending the relativity of truth: the bearers of truth and falsity are timeless Sätze an sich, and their being true or false, is therefore timeless, and the properties of be‑ ing true and being false are timeless, too . Bolzano does not put it that way, though, for he says that propositions have these properties always and for ever (fortwährend, für immer) . The topic is dealt with at several places of the Wissenschaftslehre . With Wolfgang Künne we can make a distinction between atemporalism, the thesis that truth is timeless, and sempiternalism, the thesis that if something is true it is true since ever and for ever, that is, true at all times 7. In section 147, Bolzano admits that a sentence may seem to be now true, now false, depending on the time, place and objects to which the sentence relates . The proposition expressed by the sentence “This flower smells pleasant” (Diese Blume riecht angenehm) seems to change its truth‑value depending on whether the word “this flower” refers to a rose or a stapelia 8. The topic of the section is not the question whether truth is relative or absolute, though . Bolzano’s point is that we can consider certain parts of propositions as variable, using this idea to explain the concept of validity of a proposition . The sentence “This flower smells pleasant” ex‑ presses a propositional form due to the term “this,” which corresponds to a set of propositions (Sätze an sich) that may be obtained by considering one part of the proposition to be variable, here the part expressed by “this .” The example therefore does not refute Bolzano’s thesis that every Satz an sich is true, or false for ever. For Łukasiewicz, Bolzano’s account of the validity of sentences has been of central importance for the development of an account of probability, and the question whether certain propositions may be true only relative to a certain time . It is precisely Twardowski who pointed to Bolzano’s chapter on validity when Łukasiewicz was working on probability (Łukasiewicz 1913, p. 52, note). I come back to this at the end of the section . Although we find a similar, though not identical example in Twardow‑ ski’s treatment of the topic of the absoluteness of truth, “the smell of this flower is pleasant”, the idea of the validity of a proposition is not present

6 Ohne dass das Urteil selbst sich geändert hätte – wenn draussen die betreffende Realität erzeugt oder zerstört wird, gewinnt oder verliert ein solches Urteil oft seine Wahrheit (Bren‑ tano 1889, p . 26) . 7 See (Künne 2003b, p . 286) . I endorse Künne’s suggestion that Bolzano understands truth to be atemporal rather than omnitemporal, notwithstanding what he says in the passage given above . 8 A cactoid plant with fleshy leaves that smells of decayed meat, as Łukasiewicz (1913, p. 53) remarks, while dealing with Bolzano’s example . Truth and Time 139 in Twardowski’s writings . Section 125 of the Wissenschaftslehre seems to have been more important for Twardowski, as its title is “Every proposition is either true or false, and this for ever and everywhere” (für immer und allenthalben (Bolzano 1837, §125)) . Truth and falsity are properties of Sät‑ ze an sich, which do not change from time to time, or from place to place (idem) . This section is especially relevant for Twardowski, because Bolzano argues that when people speak of propositions that are both true and false, neither true nor false, or true only for a certain time or for a certain place, they do not speak of Satz an sich, but of the linguistic expression of a Satz an sich . In Twardowski’s paper, the distinction between a judgement and its linguistic expression also plays a role in the argument against the rela‑ tivity of truth . Twardowski must have read these sections, but perhaps not shortly before he discusses the topic . Especially section 25 of the Wissen‑ schaftslehre shows that Bolzano’s formulation of the problem is different from the way Twardowski puts it in his paper . According to Bolzano, the sentence “It snows” seems to express a transitory relationship (ein vorüber‑ gehendes Verhältnis) . In order to be true, such a sentence needs the addition of a time specification, and often of a specification of place as well: “It is snowing today, in this place” (Heute, in diesem Orte schneit es) . As we will see below, for Twardowski such an indexical term needs to be replaced by a definite description in order to account for the absoluteness of truth . According to Bolzano, there are simple intuitive presentations (Anschauun‑ gen), which stand for precisely one object – so‑called Einzelvorstellungen (§72) . These presentations are exact, although their linguistic expression is pre‑eminently by means of the indexical “this (A)” .9 Precisely because Bolzano acknowledges Sätze an sich as bearers of truth and falsity, he is not in need of Quinean eternal sentences “free of indicator words ”. 10 It is true that such Sätze an sich cannot always be completely expressed in lan‑ guage, as we may have no other means of expressing a proposition than by saying “this hurts”, because we talk about a specific pain for which we do not have a definite description, but as truthbearers these Sätze an sich are completely specified . Although the language we use is indeterminate, the Satz an sich expressed is completely determinate, and has a truth‑value for ever and everywhere . This is possible precisely because of the fact that the Satz an sich is independent of language .

9 The Vorstellung an sich expressed by “This” (Dies A) already relates to precisely one object; cf . (Bolzano 1837, p . 9) . 10 “[A]n eternal sentence [is] a sentence whose truth value stays fixed through time and from speaker to speaker . … An eternal sentence may be expected to be free of indicator words … To finish the job of eternalizing the sentence … we have to supplant the ‘now’ by a date …” (Quine 1960, pp . 193-194) . 140 Maria van der Schaar

Twardowski does not want to acknowledge such timeless bearers of truth and falsity, and argues that judgement products, which exist only at the mo‑ ment of the act of judging, are the proper bearers of truth and falsity . How can he defend the thesis that truth and falsity are not dependent upon time, while the truthbearer is dependent upon time? Twardowski argues that those who think that truth is relative do not make a distinction between a linguistic proposition (a declarative sentence, enuntiatio, Aussage), and the judgement expressed by the sentence . According to Twardowski, the external expres‑ sion may be associated with one judgement product on one occasion, and with another product in other circumstances (Twardowski 1900a, p . 149) . A linguistic proposition may be called relatively true insofar as it expresses under certain conditions a true judgement, and a false judgement under other conditions . Although the sentence can thus be called relatively true, this does not imply that the judgement, the proper bearer of truth, is relatively true . Twardowski argues that the distinction between relative and absolute truth only applies to sentences, not to judgements (p . 169) . When do two sentences express the same judgement? According to Twar‑ dowski, “one may only speak of one and the same judgement if – speaking in the language of the traditional logic – the same subject of the judgement is given, the same predicate, the same quality, number, etc .” (p .149) . Twardows‑ ki is giving here an identity criterion for judgements in a traditional termi‑ nology he himself does not endorse . Can we give an identity criterion in his own terms? On the one hand, each judgement is unique and irrepeatable . On the other hand, Twardowski is willing to speak of the identity of judgement through different but similar particular judgements . Judgements may form a set, a manifold, from which an identical judgement may be abstracted, as we have seen in chapter IV . The basis for belonging to such a set is formed by the common properties of the judgements . In order for the judgements to belong to the same set, those properties that are relevant for the determi‑ nation of the truth‑value of the judgements need to be present in all these judgements, and this may be used as an identity criterion for judgements . Most of the sentences we use are elliptical . We say such things as “I will,” “Yes”, or “It is raining .” Without any abbreviation we might say “It is rain‑ ing here and now ”. The addition is not enough, though, to determine the truth‑value of the sentence . Terms such as “here” and “now” are, according to Twardowski, “highly ambiguous”: “The term “now” changes its meaning on every occasion of its utterance” (Twardowski 1900a, p .152) . Today we say that the linguistic role or linguistic meaning of the term “now” does not change from occasion to occasion, and Twardowski has seen the point, for he says that the same general concept is expressed by each use of the term “now” (p .153) . We also may agree with Twardowski that not only the object referred to, but also the (non linguistic) meaning or content of the Truth and Time 141 word changes from context to context . On a Fregean account, the reference changes from context to context, and this implies that the sense also chang‑ es from context to context . For Frege, the Sinn is the way the Bedeutung is given . This means that two terms cannot have an identical Sinn insofar as their Bedeutung differs . According to Twardowski, the judgements ex‑ pressed by a sentence containing indexical terms are fully determinate; only the sentence is indeterminate . In colloquial speech, we perceive the circumstances, and are thereby able to determine the judgement made . The circumstances accompanying the speaker’s words “fill in what the words do not express” (p . 150) . In science and logic, though, we are in need of eternal sentences . Science and logic have to disambiguate indexical terms, and they have to substitute, for example, the phrase “on December 17, 1899; in accordance with the Gregorian calendar; at 0:13 in the morning; central European time” for the word “now .” According to Twardowski, this phrase has exactly the same meaning as the word “now” used in the appropriate circumstances (p . 150) . What he calls the meaning of these terms is thus the time referred to . If the circumstances change, the meaning changes, too, and one can no longer say that the same judgement is expressed by the sentence “It is raining here and now” . Although the judgement products, which are the bearers of truth and falsity, are dependent upon an act of judging, and thereby on a certain moment in time, no judgement product changes its truth‑value, once true . Does Twardowski also claim that the judgement has always been true, that is, that before the rain had fallen on the High Castle Hill in Lwów at a certain time, it was already true at an earlier time? Twardowski says that if a judgement is true, it is true always and everywhere (p . 153), that is, for all times and all places . So, if one judges on February 28, 1900 that it rains on High Castle Hill in Lwów on March 1, 1900, that judgement is true, pro‑ vided the rainfall occurs the next day, although the rain has not fallen yet . For Twardowski, the truth and falsity of each judgement J is omnitemporal: If J is true at time t, then J is also true at any time t’ before or after t; and, if J is false at time t, then J is also false at any time t’ before or after t . The omnitemporality of truth means that truth is not relative to circumstanc‑ es of time, and it is thus a specification of the absoluteness of truth . The latter is a stronger property, for truth may be omnitemporal, though relative to other circumstances . What are called relative truths, Twardowski says, are sentences that do not intimate their judgement accurately, as in “A cold bath is healthful,” and “It is good that frail children will be exterminated .” The first sentence seems to express a general judgement about all baths, in which case the judgement would simply be false . The judgement expressed is, in fact, not general; it is a particular judgement which states that a cold 142 Maria van der Schaar bath tends to be healthful . If this judgement is true, it “remains forever true” (p .156) . The example therefore does not prove that truth is relative to circumstances . If we understand that the judgement intimated on a par‑ ticular occasion by a certain agent is true, we understand that it has to be true always and everywhere . In the second sentence, the word “good” does not have the same meaning for the Spartans as for us, and this means that our judgement concerning frail children does not contradict the judgement of the Spartans, and it therefore does not mean that what was once true, is now false . Those relativists who say that a hypothesis is true within one framework of research, but may become false as the result of new discov‑ eries, are wrong . A scientific hypothesis should be called “probable” rather than “true,” and if it is in fact true, no new discoveries can conflict with the hypothesis (p . 160) . The relativist claims to derive his position from epistemological subjec‑ tivism . The subjectivist is to be taken seriously insofar as he makes a claim about the finitude of human knowledge (p .166) . The manner in which man represents things to himself, and judges about them, is dependent upon his constitution . There is no point in denying this, “yet, once some judgement made by man is true, it will never cease to be true for anyone” (idem) . I come back to this argument below . The subjectivist is a relativist insofar as he claims that the judgements we make are true only for us, and that these judgements may be false for beings that have a different constitution . The subjectivist and the relativist argue that the sentence “The scent of a rose is pleasant” is true for the person who likes the scent, while false for the person who does not like it . We have here, Twardowski argues, an elliptical sentence (p . 163) . A precise formulation would give: “The scent of a rose is pleasant to me”, and the phrase “to me” has a different meaning each time the sentence is uttered, for it designates different persons if different people are speaking . The sentences do not express the same judgement, and the example therefore does not show what the relativist claims . If the relativist makes the claim that not only some, but all judgements are true only for man, that is, if the relativist suggests “the possibility that other beings may no less correctly regard as false those judgements that man correctly regards as true” (p .163), his thesis is irreconcilable with the principle of contradic‑ tion . If someone correctly regards a judgement as false, the judgement is false, and if another correctly regards a judgement as true, the judgement is true . The relativist is thus committed to the claim that the same judgement may be both false and true . How does Twardowski formulate the logical principles of contradiction and excluded middle? Twardowski gives no explicit explanation of these principles in the paper on relative truth, but the context makes it clear that the two principles together imply that every judgement is either true, or Truth and Time 143 not true (p .161), whereas the principle of contradiction says that no judge‑ ment can be both true and false (p . 162) . From this it would follow, for the relativist, that a judgement and its contradictory are both true, which is a violation of NC . All proof relies on these principles, which are the funda‑ mental laws of thinking, according to Twardowski . Especially the principle of contradiction cannot be eliminated from human reasoning and thinking; it functions as a presupposition of thought . In the Wiener Logik, these highest principles of thought are axioms, a priori pieces of knowledge (Erkenntnisse; cf . (Twardowski 1894/1895, p . 228)), which are immediately evident . They are judged with full insight, and therefore do not need any grounding (Begründung) . In the Logik, Twardowski gives the following explanation of the principle of contradiction: Of two judgements, of which one affirms what the other denies, one must be false, that is, they cannot both be true (p . 230) . And the principle of excluded middle says: Of two judgements, of which one affirms what the other denies, one must be true, that is, they cannot both be false (p . 230) . It is to be noted that in both laws, the second explanation (“They cannot both be true/false”) does not have the same meaning as the first (“One must be false/true”) . The second explanation expresses, respectively, NC and EM, and is thus weaker than the first explanation, which express, respectively, NCS and EMS . The two principles given above are independent of each oth‑ er, according to Twardowski . The combination of the two principles is very powerful, as it says that of two contradictory judgements that one is true, while the other is false, a version of BV . This is due to the first, stronger part of the principles: “one must be false / true .” The claim that they cannot both be true / false is the weaker formulation that allows for truth‑value gaps . In the Wiener Logik, Twardowski used the principle of contradiction to argue against the skeptic . He explains the skeptic as someone who denies that there is a distinction between true and false (p . 234) . The skeptic does not argue that truth is relative; his position, though, is relevant with respect to the logical laws . Twardowski argues against the skeptic that his position cannot be defended without assuming the principle of contradiction; as soon as he makes a judgement, he claims that what he says is true, and he there‑ by presupposes that the distinction between truth and falsity makes sense . It is of interest that for Twardowski the question whether truth is ab‑ solute or relative is directly related to the logical laws, as the status of the logical laws is of central importance to his pupils Łukasiewicz, Leśniewski and Kotarbiński in their early writings. As we will see in the next section, Kotarbiński relates the question whether truth is relative to time directly 144 Maria van der Schaar to the question which formulations of the law of excluded middle can be regarded as universally valid . In his beautiful and clear monograph on the law of contradiction in Ar‑ istotle published in 1910, Łukasiewicz claims that the law of contradiction has, in Aristotle, an ontological, a logical and a psychological meaning . The ontological law says: No object can contain and not contain a certain property at the same time (Łukasiewicz 1910a, p. 10). Referring to all objects in general, it is a metaphysical principle, as Le‑ śniewski puts it (1913a, p. 49). The logical law says: two judgements, of which the one attributes a property to an object, which the other denies of the object, cannot both be true (p . 11) . Remember that for Łukasiewicz “judgement” means a declarative sen‑ tence together with its meaning . In his “Logic and Psychology” from 1907 Łukasiewicz argues that a confusion of logic and psychology is due to an ambiguity of the term “judgement .” It may refer to the mental act of judg‑ ing, which is a psychological notion, or it may stand for the judgement in a logical sense, the sentence, which is the proper bearer of truth and falsity (Rojszczak & Smith 2003, p . 253) . Twardowski himself was influenced by Łukasiewicz’ distinction when he wrote his paper on the distinction between actions and products, but Twardowski has given a different, non‑linguistic interpretation of the judgement in the logical sense, as we have seen in chapter IV . The psychological law says that two convictions corresponding to contradictory judgements, cannot exist in the same intellect at the same time (p . 13) . According to Łukasiewicz, the psychological law is an empirical law (p. 39), and can therefore not give a foundation to logic . The term “object” in the other two laws is to be understood in a most general sense . An object is everything that is something and not nothing: “things, persons, phenom‑ ena, events, relations, … thoughts, feelings, concepts, theories” (p .136), including fictional objects. Łukasiewicz’ term “object” is thus as general as Twardowski’s use of the term in 1894 11. There is no reason to exclude contradictory objects from the category of objects, as Meinong has pointed out (p . 135, note) . The ontological and logical law of contradiction should not exclude contradictory objects in an a priori manner . It is therefore es‑ sential to these laws that the definition of object does not contain the thesis that the object is something that cannot have and have not a certain property at the same time (p. 135). According to Łukasiewicz, a proof for the law of contradiction is in need of such a thesis, and this means that we cannot

11 Łukasiewicz worked with Meinong in Graz in the academic year 1908/1909. So, he probably got the broad notion of “object” directly from Meinong, and only indirectly from Twardowski . Truth and Time 145 prove the ontological or the logical law this way . This conclusion would be no problem for Twardowski’s early account of the law, as he takes the logical law of contradiction to be an axiom, whose truth is guaranteed by an act of insight . As a good Brentanist, Twardowski says that the law is immediately evident. This position is not available, though, to Łukasiewicz. For, he gives a severe critique of such a position: (1) he himself does not see that the law is evident; (2) being evident is not a criterion of truth, for if the notion of evidence is not internally related to that of truth, it can only refer to a mental state, a feeling that accompanies our judgements (p . 126) .12 As philosophers have often called judgements evident that turned out to be false, there is no internal relation between the notions of being evident and being true . Evidence is thus nothing but a feeling . Furthermore, it is not possible to prove the law by appealing to the psychological necessity of the law; the distinction between the logical and the psychological law makes it clear that the psychological necessity of the law only relates to the latter, but even in that sense there is no proof, for it is doubtful that there exists such a psychological necessity (p . 127) . For Łukasiewicz, the term “judgement” is central to the logical principle. What he calls a “judgement” is not a mental state of conviction or a com‑ bination of concepts; it is rather a string of words proposing (aussagen) that something is or is not (Łukasiewicz 1910a, p. 15). Judgement is for Łukasiewicz a linguistic entity, thus keeping the Twardowskian terminology of “judgement,” but changing the meaning of the term . The “judgement” is for Łukasiewicz a logico‑linguistic proposition, a declarative sentence; a logical, not a psychological fact . I will use the term “(linguistic) propo‑ sition” or “(logical) sentence” where Łukasiewicz speaks of “judgement”. The sentence essentially has a meaning, and it is the proper bearer of truth and falsity . When do two linguistic propositions express the same thought, that is, when do they have the same sense? Łukasiewicz acknowledges that problems around identity of sense are of great importance: The distinction between sense identical judgements and equivalent judgements with non‑identical sense belongs to the most difficult, but als to the most important prob‑ lems of logic (pp . 17-18, note) . If two sentences have the same meaning, then they are logically equiv‑ alent . It might be the case, though, that two logically equivalent sentences differ in meaning . Logical equivalence is a necessary, but not a sufficient condition for identity of meaning. According to Łukasiewicz, “Plato was

12 Die Verwendung des Begriffs der Evidenz als Wahrheitskriterium ist ein Überrest des ‘Psy‑ chologismus’, der die philosophische Logik auf den Holzweg brachte (Łukasiewicz 1910a, p . 127) . Thinking of Heidegger’s metaphor, not meant to be a Holzweg with a Lichtung, an appearance of truth, at the end . 146 Maria van der Schaar the teacher of Aristotle” and “Aristotle was the pupil of Plato” are logically equivalent, because each follows from the other . The two sentences do not have the same meaning, though, for the word “Aristotle” relates to another object than the word “Plato,” and the phrase “was the teacher of Aristotle” relates to another property than the phrase “was the pupil of Plato” (Łuka‑ siewicz 1910a, p .17) . Apparently, the linguistic structure of the sentence determines the structure of the meaning, certainly not a thesis endorsed by Frege. For Łukasiewicz, all logical sentences have either the form “Object O contains property P,” or “Object O does not contain property P” (p .16) . Sentences have the same meaning, if and only if “O” signifies the same object, “P” signifies the same property, and these sentences have the same form, that is, they are both positive or both negative . This means that the sentences “Aristotle was the founder of logic” and “The Stagirite was the founder of logic” have the same meaning. Leśniewski will criticise Łu‑ kasiewicz on this point . Central to Łukasiewicz’ logical principle is the definition of a true judgement: “True is an affirmative judgement, which attributes a property to an object that the object contains; true is a negative judgement that de‑ nies a property of an object the latter does not contain” (p . 20) . A definition Łukasiewicz considers to be supported by Aristotle’s weaker definition of truth given in Metaphysics Γ . Being is not only the logical ground, but also the real cause for the truth of the judgement, as Łukasiewicz puts it (p. 21). On the basis of the definition of a true judgement one can show the logical and the ontological law are equivalent, although they are not identical in meaning, for the first is about objects, whereas the latter is a law concerning meaningful sentences . Łukasiewicz’ doubts concerning the possibility to prove the law of con‑ tradiction is not meant as a diminishment of the importance of it as a prin‑ ciple . It is true that it has no logical value, as it cannot be proved within the system . It has, though, a considerable moral value . Without the principle, it would not make sense to say that a judgement contradicts other judgements made, and it would thus be impossible to point out that an error is made, that someone is lying, or to defend oneself in court against accusations, although one has an alibi (pp . 166‑169) . Łukasiewicz also expresses some doubts concerning the principle of ex‑ cluded third in this early period . Perhaps, abstract objects are not subject to the principle . Abstract objects, which include not only general objects, but also mathematical and logical objects, are, for Łukasiewicz, constructions of the mind 13. These created objects have relations among each other that

13 Łukasiewicz calls the general objects reconstructions of the mind, because they depend on experience, whereas mathematical and logical object are constructions, which are independent of experience . Truth and Time 147 cannot freely be chosen by the mind . They may even contain contradictions, as Russell’s paradox has shown, and they are in this sense independent of our will (pp . 140‑149) . Abstract objects do not exist in reality; they are nothing but products of the human mind (p .140) . We obtain them by comparing a sequence of concrete objects, focus on the common properties, and delete the properties in which they differ (p. 139). Łukasiewicz’ explanation of abstract, general objects is in this sense reminiscent of Twardowski’s ex‑ planation of general objects given in 1894 . Łukasiewicz follows Meinong in calling these abstract objects incom‑ plete . What makes the abstract object general is its incompleteness . Probably Łukasiewicz has discussed the topic of incomplete objects with Meinong when he was in Graz in 1909, although Meinong had not published his work on incomplete objects yet . When we speak of a concrete pillar like the Mickiewicz‑pillar in Lwów one can either correctly attribute or cor‑ rectly deny of the pillar any property one can think of . This means that the pillar is a complete object . Abstract general objects like the pillar as such form an important group of incomplete objects: there are properties, such as the property of being made of bronze, which can neither correctly be at‑ tributed nor correctly be denied of the incomplete object: “[T]he judgement “The pillar [as such] is made of bronze” is neither true nor false” (p . 139) . From this Łukasiewicz does not immediately draw the conclusion that the principle of excluded middle does not hold for abstract, that is, incomplete objects . Only if one claimed that each of the two judgements is false, one would be committed to the claim that incomplete objects are not subject to the principle of excluded middle (p . 139, note) . The formulation of the principle of excluded middle Łukasiewicz must have in mind can be found in the talk Łukasiewicz gave in Lwów in 1910: It is doubtful whether the principle of excluded middle holds for general objects, such as the triangle in general or man in general … this object is … not defined with respect to equilaterality and non‑equilaterality . For this reason, both the propositions “the triangle is equilateral” and “the triangle is non‑equilateral” appear to be false (Łukasiewicz 1910b, p. 69). The principle of excluded middle would thus be that two contradictory sen‑ tences cannot both be false, that is, the principle would not even hold in its weak sense, for this is EM mentioned above . The fact that the triangle in general is not determined with respect to the property of equilaterality ex‑ plains that the principle of excluded middle might not hold for it . The talk of 1910 in Lwów foreshadows the discussion of the principle of excluded middle in Kotarbiński’s article from 1913, which discussion will be the topic of the next section: 148 Maria van der Schaar

With regard to real objects, the principle of excluded middle seems to be closely con‑ nected with the postulate of universal determination of phenomena, not only present and past but also future ones . Were someone to deny that all future phenomena are today already predetermined in all respects, he would probably not be able to accept the principle in question (Łukasiewicz 1910b, p. 69). It is to be noted, though, that EM would be valid for future phenomena; it is EMS that does not hold for them. Łukasiewicz apparently does not make a distinction between the two formulations . Below we will see that Kotar‑ biński saw the distinction in 1913. Although Łukasiewicz thus expressed some doubts concerning the validity of the principle of excluded middle in 1910, he did not question the principle . In his early writings, he endorses the principle of bivalence for such sentences. In 1913, Łukasiewicz states that sentences which do not contain a free variable are either true or false, “even if before the event we can never know which of them are true and which are false” (Łukasiewicz 1913, p. 38). Suprisingly, in 1913 he defends the thesis that sentences concerning future contingents are either true or false . Sentences with free variables can be called probable. For Łukasiewicz, probability is neither purely subjective, nor purely objective . Probability is a creative construction of the human mind; it is a logical concept, as it is a property of indefinite sentences . These sentences have a similar relation to the world as negative sentences . Whereas true sentences have a counterpart in certain facts, false sentences have no objective correlates . Falsehood is characterized by a negative relation to facts (p . 37) 14. Although falsehood does not exist without the human mind, the concept of falsehood is free from subjective elements (idem) . Because each probable sentence has its counterpart in a group of true and false sentences, the notion of falsehood is part of the notion of probability . Of interest for later developments towards a three‑valued logic is that Łukasiewicz does acknowledge more than two “truth‑values.” Although Łu‑ kasiewicz uses Frege’s “truth‑value” terminology, calling it also the logical value of a sentence, there is an important difference between Łukasiewicz’ notion of truth‑value and that of Frege . Whereas for Frege only the Gedan‑ ke has a truth‑value, which means that there are only two truth‑values, for Łukasiewicz truth‑values pertain only to sentences with free individual variables . The truth‑value of a sentence with free variables is 1, if it yields only true sentences for all the values of its variables; it is 0, if it yields only false sentences for all the values of its variables; and it is a proper frac‑ tion, if it yields a true sentence for some of the values of its variables, and a false sentence for other values (p . 17) . The fraction is the ratio between the number of values for the variables for which the open sentence gives

14 A position one can later find in (Russell 1918, p . 208) . The concept of indefinite proposition or propositional function was taken from Russell’s Principles of Mathematics . Truth and Time 149 true sentences and the total number of values for the variables (idem) . In order to determine the truth‑value of an open sentence we therefore have to determine the range of the variable, and, according to Łukasiewicz, this range has to be finite and non‑empty . As was noted at the beginning of section 2, Twardowski drew the atten‑ tion of Łukasiewicz to Bolzano’s writings on the validity and the compar‑ ative validity or probability of a proposition 15. Twardowski was important for the spreading of Bolzano’s ideas also in cases where he himself did not specifically endorse them . Starting with a sentence with an indexical term, “This flower smells pleasant”, Bolzano draws our attention to the fact that we may consider a part of a Satz an sich as variable . In the sentence, the term “this flower” may refer to a rose or a stapelia, the notion expressed thus differs in the two cases, which means that the sentence may be considered as expressing two propositions . We are thus able to consider a certain notion in a proposition as variable (veränderlich) . If we then consider the different propositions that may thus be obtained, we are able to form the concept of validity of a proposition . If all the propositions resulting from the substitution of a variable part are true and objectual, the original proposition is called completely valid with respect to this part, and the degree of its validity is 1 . If only false propositions are obtained than the proposition is generally invalid with respect to this variable part, and the degree of validity is 0 . The degree of validity of a proposition with respect to a variable part is the ratio of true propositions to all propositions resulting from the substitution . Łukasiewicz compares his notion of truth‑value with that of validity in Bolz‑ ano and concludes that the concept of a variable and of an indefinite or open proposition are absent from Bolzano’s writings . As Wolfgang Künne (2003a, p . 183) has shown, though, there is a notion in Bolzano comparable to the notion of indefinite sentence: what Bolzano calls “forms of propositions,” such as “Some A are B” .16 Łukasiewicz’ notion of indefinite sentence plays a similar role in logic (Łukasiewicz 1913, p. 55). For Łukasiewicz, though, probability is a non‑relative property of indefinite sentences, whereas for Bolzano validity is a property of a complete proposition relative to certain parts of it. Łukasiewicz claims that for him not all sentences are either true or false, but this is only an apparent difference, for his notion of sentence includes indefinite sentences . We may conclude that he still adheres to the principle of bivalence for definite sentences in 1913 .

15 Łukasiewicz (1913, p. 52, note). He refers to the central sections §147 and §161 of (Bolzano 1837) . He also notes that his stay in Graz with Meinong in 1909 brought him to the topic of probability. Cf. (Łukasiewicz 1913, p. 49). 16 So kann mann sagen, die Logik betrachet nur Formen von Sätzen, nicht aber einzelne Sätze (Bolzano 1837, §12, p . 48) . Properly speaking only the linguistic expression van be called a form of a sentence, Bolzano adds . 150 Maria van der Schaar

3. Determinism and the Relativity of Truth to Time

Truth and Time If one acknowledges with Bolzano a realm of objective propositions as bearers of truth and falsity outside space and time, truth and falsity may be said to pertain to propositions independent of time and place . The fact that truth and falsity are relative characteristics (see section 1) does not make truth dependent upon time . It may be said that the fact that an object A has a certain property b makes the proposition that A has b true, but this relation of truemaking is an internal relation between a fact and the truth of a proposition; it is a logical, not a natural relation, and is therefore as independent of time as the logical proposition . It is natural for a logical re‑ alist to defend the absoluteness of truth, from which the logical principles of contradiction, excluded middle and bivalence follow . If one does not allow for objective bearers of truth and falsity independent of time, like Brentano, Twardowski and his pupils, how can one defend the thesis that truth is absolute, that truth is independent of time? For Brentano, the truth of a judgement is dependent upon time: the same judgement may be true at one time, while false at another time, but, at any moment t, the logical principles are valid for all judgements at t . Brentano is in that sense able to defend the law of contradiction and excluded middle, although he is willing to formulate the principle of excluded middle only in its negative version . Of two contradictory judgements, not both can be incorrect, that is, Brentano gives a variant of EM´ . Twardowski and his pupils are opposed to such a strong relativity of truth to time . Twardowski’s solution is to argue that all indexical terms are to be deleted from the logical judgement . This does not seem to form a final defence of the absoluteness of truth, for one may question the absoluteness of truth for sentences that do contain indexi‑ cals, and it may be doubted whether all indexicality can be eliminated from the bearer of truth 17. This need not be a problem for the logical realist as we have seen above, but it seems to be a problem for those defending that the sentence is the bearer of truth . On a particularist account that is given by Twardowski and his pupils, sentences and judgements have a being in time . If sentences and judgements are truthbearers in time, one may wonder whether the sentence or judgement “It rains on the Castle Hill of Lwów on August 14, 2014” is already true, or false, now, on August 13, 2014 . A similar question would not be asked if it is now August 15 of that year, for by that time it is definitely true, or false, that

17 Not all sentences can be translated to eternal sentences, as Peter Simons (2008, p . 8) has noted . His position is of interest as he defends the absoluteness of truth, while upholding that the truthbearer may contain essentially indexical elements . Truth and Time 151 it rained on the Castle Hill of Lwów on August 14, 2014, and the very same judgement uttered on August 13, 2014, is now, on August 14, true, or false . Although we speak about the same judgement, the point of evaluation has changed, and this means that the judgement has now a definite truth‑value . Michael Dummett has called the point of evaluation the time of assessment: The relativization of truth‑value … to a time t should not, however, be taken as re‑ lating to the time at which the sentence is being uttered … but, rather, to the time at which the truth‑value of the sentence is being assessed (Dummett 1981, p . 394) . We have to take into account (1) the time to which the sentence refers; (2) the time of utterance; and, (3) the time of assessment .18 A relativization of truth to the time of assessment, in distinction from the time of utterance, is also defended by John MacFarlane (2003), where it is argued that if we evaluate Jake’s assertion made yesterday that there would be a sea battle today, Jake is able to meet the challenge today by pointing to ships fighting, and we can thus say that his assertion is now correct, although it was not correct (“neither true nor false”, in MacFalane’s words) yesterday . Some‑ times, we speak of a similar or “identical” judgement “It rains on the Cas‑ tle Hill on August 14, 2014”, while abstracting from the time of utterance . Twardowski is willing to speak of an identical, abstract judgement indepen‑ dent of a specific time of utterance (see section 2 above) . Even a nominalist like Leśniewski is willing to speak of identical judgements or sentences in a figurative sense (see below) . On August 13, the judgement seems to be neither true nor false, but from August 14 on, the very same judgement is true, or false . On this account, the time of reference is constant, the time of utterance is irrelevant, while the time of evaluation is at issue . Is the truth of a judgement dependent on the time of evaluation? If so, the absoluteness of truth cannot be defended . And, if some judgements are neither true nor false at a certain moment of evaluation, the principle of bivalence, and a strong version of the principle of excluded middle cannot be defended either, al‑ though one still can say that of two contradictory judgements, if one is false, the other is true . One may think that truth is relative to time, because one believes that determinism is false, as Łukasiewicz and Kotarbiński have done, as we will see below; or, one may think that truth is relative to time, because one defends an epistemic notion of truth . If there is no method to determine whether a judgement is correct or incorrect, which seems to be the case with some judgements about the future, the judgement is neither correct nor incorrect . Sometimes, the arguments of the Polish logicians are given in epistemic terms, but as they defend a correspondence theory of truth in Twardowski’s sense, their main argument is not given in such terms .

18 Krystyna Misiuna (1998, p . 201) distinguishes a time datum from “a time of evaluation,” but seems to identify the time of evaluation with the time of utterance . 152 Maria van der Schaar

It is clear that Łukasiewicz’ doubts about the logical principles and the questions he had raised concerning the principle of excluded middle in the lecture given in Lwów in 1910 must have been very stimulating for Kotar‑ biński, as he is the first to doubt the principle of bivalence and the strong principle of excluded middle on the basis of a denial of determinism in a paper published in 1913. Leśniewski immediately reacted to this paper, arguing against Kotarbiński’s thesis, and Twardowski sides with Leśniews‑ ki, upholding the absoluteness of truth in a lecture course from 1913/1914 . I will focus here on the discussion around 1913, as much research has al‑ ready been done on later developments in Łukasiewicz’ writings. It may be helpful, though, to understand Łukasiewicz’ later position. As Kotarbiński’s paper is a first proposal, Łukasiewicz’ concepts are more developed; he gives, for example, a clear definition of determinism . I will therefore start with Łukasiewicz’ writings of a later date, in which a three‑valued logic is developed, before I go back to the discussion around 1913, in which Ko‑ tarbiński, Leśniewski and Twardowski participated.

Jan Łukasiewicz Łukasiewicz started to work on a three‑valued logic in the summer of 1917 (cf. (Łukasiewicz 1918, p. 86)), applying now the idea of a third value to definite sentences . In addition to true and false sentences, there are possi‑ ble sentences . Besides being and non‑being, there is objective possibility, symbolized by “½:” Possible phenomena have no causes, although they themselves can be the beginning of a causal sequence . An act of a creative individual can be free and at the same fact affect the course of the world (idem) . The idea of a three‑valued logic is thus right from the beginning con‑ nected with the denial of determinism and the possibility of free, creative actions, and thus connected to a metaphysical problem. Łukasiewicz’ view on logic is not identical with that of Russell or Frege, in which the laws of logic are the most universal laws of the world: The possibility of constructing different logical systems shows that logic is not re‑ stricted to reproduction of facts but is a free product of man, like a work of art . Log‑ ical coercion vanishes at its very source (idem) . As for Carnap, in logic there are no morals, but in contrast to Carnap, no conventionalism is part of Łukasiewicz’ conception of logic. For Łu‑ kasiewicz, endorsing classical logic implies an endorsement of determinism, as the principle of determinism follows from that of bivalence . In order for logic to be as metaphysically neutral as possible one should adopt a logical Truth and Time 153 system that does not have such metaphysical consequences. And, Łukasiewicz argues, a many‑valued logic does not imply the principle of determinism . What is determinism according to Łukasiewicz? He explains determin‑ ism in terms of truth and time, but not in terms of necessity . Determinism is true, precisely if it is the case that: If A is b at instant t, it is true at t´, earlier than t, that A is b at instant t (cf. Łukasie‑ wicz 1961, p . 113) . The time t´ is the point of assessment or evaluation (3, see p . 151, above), while t is the time of reference (1) . Łukasiewicz’ concept of determinism is a quite innocent semantic principle (cf . (Placek 2006, p . 172)), as it relates the truth of a sentence to something in the world to which the sentence may be said to correspond . It says merely that what will happen will happen; it does not say that what will happen will happen due to a determining chain of causal events . It does not imply that “everything in the world takes place of necessity and that every free and creative act is only an illusion” (Łukasiewicz 1961, p. 120). According to the principle of determinism, everything that will be true at some future time will already be true today, and this means, according to Łukasiewicz, that the future is as much determined as the past: The determinist looks at the events taking place in the world as if they were a film drama produced in some cinematographic studio in the universe … We are only pup‑ pets in the universal drama (idem, p . 113) . There are, according to Łukasiewicz, two arguments for determinism: one is based on the principle of excluded middle, the other on the law of causality . The principle of excluded middle says, according to Łukasiewicz, that: [T]wo contradictory sentences are not false together, that is, that one of them must be true (p . 114) . If the law merely says that two contradictory sentences are not false together, the possibility that determinism is false is not excluded, for if a sentence about a future contingent is neither true nor false, then neither it nor its negation is false . If the law also says that one of these sentences “must be true,” as Łukasiewicz adds in the explanation, putting forward the stronger version EMS, then the law implies that it is true now that John will be at home tomorrow noon or it is true now that John will not be at home tomor‑ row noon. Apparently, Łukasiewicz presupposes that “John will not be at home tomorrow noon” means the same as “It is not the case that John will be at home tomorrow noon” . Using the general principle “If it is true at in‑ stant t that p, then p” (idem, p . 115), we can derive that either John will be at home tomorrow noon, or that John will not be at home tomorrow noon . Then, Łukasiewicz argues, we can derive by propositional logic the thesis 154 Maria van der Schaar that if John will be at home tomorrow noon, then it is true at arbitrary instant t – and thus also with t´ earlier than t, that John will be at home tomorrow noon; the argument takes two pages, and will not be reproduced here (cf . idem, pp .116 -117) . If we generalise this argument we are thus able to derive the principle of determinism mentioned above from the strong principle of excluded middle . The more famous argument Łukasiewicz gives against determinism is based on the fact that one may doubt the principle of bivalence, and is in‑ spired by Aristotle’s famous passage in De interpretatione, chapter 9 (idem, p. 125ff). The principle is not evident to Łukasiewicz, and it is possible to give a logical system without the principle, he claims, in which a third value is acknowledged: a sentence may be true, false, or neither true nor false, that is, indeterminate . Sentences about future facts have no real cor‑ relate that would make them true or false . As their ontological correlate is a possibility, they are indeterminate (idem, p . 126) . The argument is well known . What seems not to have been noted in the secondary literature is the fact that Łukasiewicz extends the argument to sentences about the past. If something that has happened in the past has no effects today, what has hap‑ pened in the past has turned into a mere possibility, and this sentence about the past is neither true nor false, but is as indeterminate as a sentence about a future contingent (idem, p . 128) . If a sentence expressed today about an event in the future or the past does not “correspond” to a cause or effect of that event existing today, the sentence is neither true nor false. Łukasiewicz thus assumes that a sentence is true not insofar as it corresponds to an event existing in the past or future, but insofar as it corresponds with something existing now . The sentence can only be made true by what happens now . What exists, exists now . The future and the past can only exist insofar as something exists now that determines what happens in the future, or is determined by what has happened in the past. At the end of the paper, Łu‑ kasiewicz rejects what he seemed to endorse at the beginning, namely that all truth is eternal: what was once true remains true for ever . He no longer seems to believe “that if an object A is b at instant t, it is true at any instant later than t that A is b at instant t” (idem, p . 113) . His doubts concerning the principle of bivalence thus go beyond the question of determinism and future contingents .

Tadeusz Kotarbiński There is no sign that Kotarbiński developed an alternative logical system, in his paper “The Problem of the Existence of the Future” from 1913 . It can be shown, though, that he acknowledges a third ontological value besides being and non‑being . And perhaps one can say that he acknowledged a third logical Truth and Time 155 value as well, although only in the sense that a judgement may be neither true nor false (cf. (Woleński 1990b, p. 195)). Kotarbiński argues for the the‑ sis that not all judgements are true or false on the basis of the possibility of free actions and creativity. As in Łukasiewicz’ talk from 1910, the question is discussed in the context of the question whether all judgements about the future are now already true, or false . Like Brentano and Twardowski, Kotarbiński gives an existential account of judgement. A judgement is true if and only if its object exists . “To exist” does not mean the same as “to be present”, Kotarbiński adds. The object may be simple, but in most cases it is an “objective”, that is, a relation of inherence, or what Twardowski has called a “relationship”. Kotarbiński follows here Twardowski’s account of judgement, while using Meinong’s terminology (Kotarbiński 1913, p. 10). In the paper, Kotarbiński endorses the eternity of truth:

If a judgement is true at time t, then it is true at all times t´ later than t .19

This means that, given his explanation of judgemental truth, if something exists at t, it will exist at all times later than t . Kotarbiński denies the omnitemporality of truth, that is, what the Polish logicians are to call the sempiternity of truth: If a judgement is true at time t, then it is true eternally and it is true at all times t´ earlier than t . As Kotarbiński puts it, “every truth is eternal but not since always” (p. 10). A necessary condition for creativity, Kotarbiński argues, is that a judgement about the object that is to be created is neither true nor false at any moment before it is created . For, if the judgement about that object would be already true before it is created, that future object would already exist before it is created, and possess the properties to be created, which means that creation were impossible (p . 14) . This argument does not make use of the notion of causation . It does make use, though, of the idea that if something is true, the object(ive) about which the judgement is made exists . It may be said against Kotarbiński that the existence of an objective does not causally determine the truth of a judgement; truemaking is not a natural relation . If there exists now an objective O that is the cause of an event in the future, the existence of objective O is not the cause of the truth of the judgement about the fu‑ ture event (nor is the truth of that judgement a cause of the existence of O) . Kotarbiński seems to think that truths are “created” in this sense:

19 “Eternity” of truth is used by Kotarbiński and in this chapter as a technical term, in the meaning of truth for ever . In natural language, “eternal” may mean timeless (“God is eternal”), and omnitemporal (at all times), as well; these meanings of the term are here not involved . 156 Maria van der Schaar

Any judgement which will become true at a certain moment will be declared as true precisely at that moment and its truth will be then created (Kotarbiński 1913, p. 10). Truemaking, though, is an internal relation between the truth of the judgement and the existence of the objective . The fact that there is such a conceptual relation does not prove anything concerning the freedom of our actions, we may argue against Kotarbiński. One has to make a distinc‑ tion between the logical relation of truemaking and the ontological relation of causation. As we will see below Twardowski’s critique of Kotarbiński’s position is in this direction. Kotarbiński is thus in need of an extra thesis: “Every truth is a necessity, every falsehood is impossible” (Kotarbiński 1913, p. 19). As Kotarbiński explains, an affirmative judgement is true if there is a necessary fact or objective, and a negative judgement is true if there is an impossible objective . Necessity cannot mean here logical neces‑ sity . I assume that it means causal necessity, but this gives a problem for the existence of free actions that, by definition, do not have a cause . If they do not have a cause, they are not determined, and a truth describing the action once it has happened is therefore not a necessary truth . According to Kotarbiński, a necessary condition for creativity is that a judgement concerning an object not yet created may be neither true nor false . He thus relates the topic to the principle of excluded middle . Ko‑ tarbiński denies the validity of the stronger version, that is, of EMS, but endorses EM . If a judgement is false, its contradictory is true . This means that he also denies BV; not every judgement is either true or false . A judge‑ ment that is neither true nor false is “not‑definite,” so in this sense there is merely a truth‑value gap . Interestingly, there corresponds with a judgement that is not‑definite a special ontological category . A judgement and its con‑ tradictory are indefinite if there is a “two‑sided possibility,” if the objective about which the judgement is made is possible, that is, if it may or may not exist. It seems that Kotarbiński acknowledges here a third value, although perhaps not a third truth‑value. It is essential to Kotarbiński’s argument that we can speak about a judgement that is true at t, but one should not read this as though we have a predicate “being true at t” . Such a position would certainly make truth relative to time in a deep sense: J being true at t and J being false at t´ do not formally contradict each other. Kotarbiński’s question is rather whether a judgement may be true, at a certain moment of assessment, while being neither true nor false, at another moment of as‑ sessment . As he claims that this is possible, truth is for him relative to the time of assessment . The same judgement may be true at one time, while not being true at another time . At the beginning of this section, I noted that the thesis that not all judge‑ ments are either true or false is also defended by those who acknowledge an epistemic notion of truth. Kotarbiński does give such an epistemic argument, Truth and Time 157 but it seems not to be the main argument for him . The thesis that not all truth is true since eternity is, according to Kotarbiński, “a consequence of cogitations that a certain sphere of future events cannot possibly be known to us” (idem, p . 11) .

Leśniewski’s and Twardowski’s Reaction to Kotarbiński In the lecture series called “Ethics, Criminal Law and the Problem of Free Will”, delivered at the Lwów University in 1904‑1905, Twardowski argues that “determinism does not at all have the consequences inconsistent with ethics etc ., which indeterminists suggest” (Twardowski 1904‑1905, p .315) . The metaphysical question of determinism is not to be dealt with in sepa‑ ration from scientific research . Twardowski claims that determinism is sci‑ entifically more probable, and that we do not have to postulate new causes to explain free will, as we can almost always indicate our own motives as causes of our resolutions (idem, p . 317) . In this sense he follows Brentano’s endorsement of determinism .20 Although Twardowski himself did not question the validity of the principles of contradiction or excluded middle, there was apparently an atmosphere in which his pupils had the freedom to question these laws, only a few years later than L E. J. . Brouwer had expressed his doubts concerning the law of excluded middle in his dissertation from 1907 . Inspired by Łukasiewicz monograph from 1910, Leśniewski denies the validity of the principle of excluded middle in his paper “The Critique of the Logical Principle of Excluded Middle”, in the sense that “at least one of the two contradictory propositions has to be true” (Leśniewski 1913a, p . 74) . His reason is of a semantical nature: every sentence whose subject denotes nothing is false . This means that for him “Every centaur has a tail” and “A certain centaur does not have a tail,” which are contradictory sentenc‑ es, are both false, because the subject of each denotes nothing (Leśniewski 1913a, p. 59). For Leśniewski, the general term “centaur” denotes nothing if there are no centaurs; we would now rather follow Frege, and consider the term to have a Bedeutung, namely the concept of a centaur . On a Frege‑ an analysis, if there do not exist any centaurs, the general sentence is true, while the sentence that there is a centaur, with or without a tail, is false . Leśniewski’s general sentences, though, have an existence presupposition, as we know it from Aristotelian logic . Like Twardowski, and in opposition to Kotarbiński, Leśniewski defends in the paper “Is all Truth only True Eternally or is it also True without a Be‑ ginning” the thesis that truth is omnitemporal, that is, sempiternal, under‑ standing it to be a special case of the thesis that truth is absolute . All truth

20 Brentano considers the indeterminism hypothesis to be very improbable, and he does not believe that ethics is in need of it; cf . (Brentano 1952, p . 279ff) . 158 Maria van der Schaar is true eternally and without a beginning, and this does not make free cre‑ ativity superfluous. It is of interest that Leśniewski combines a particularism and nominalism with the idea that truth is absolute . If there is no identical truthbearer through time, how can one say that a judgement is true eternally and since always? In the paper on Kotarbiński, Leśniewski makes use of Ko‑ tarbiński’s “judgement” terminology, although for Leśniewski the sentence is the proper bearer of truth and falsity . How can a particularist possibly argue against Kotarbiński’s thesis that a judgement that is now true, once was neither true nor false? For, that thesis assumes that one can speak of an identical judgement through time . Even if we consider judgements with the same meaning to be “one and the same,” a judgement “ceases to last at the moment it is uttered for the last time” (p .96) . No true judgement is eternal in this sense . The idea that a judgement is an omnitemporal truth means, according to Leśniewski, that, if it is uttered by someone, it is always true (idem, pp . 96-97) . He is thus able to speak of identical judgements through time, although this is not identity in any literal sense . Sentences intimating the judgements we consider to be identical are not to contain any indexical terms (pp .98 -99) . As soon as indexical terms are substituted by expressions whose semantic function is constant, we will obtain truths that are “eternally true,” (p. 99). Instead of considering the sentence “Stanisław Leśniewski will die” we consider the sentence “Stanisław Leśniewski possesses the property of having ceased to be alive in the future of 2 p m. ,. March 2nd, 1913,” and the judgement intimated will be always true (idem). Leśniewski thus follows Twardowski’s argument that we have to consider judgements intimated by sentences without indexicals in order to understand that truth is absolute . We may now consider judgements to be identical precisely if they can be intimated by the same eternal sentence 21. What is Leśniewski’s argument against Kotarbiński’s thesis that not every truth is true without a beginning? As Kotarbiński endorses the principle of contradiction and the weak principle of excluded middle, but not bivalence or the strong principle of excluded middle, Leśniewski’s argument should ideally use only the former principles . In the presentation of the argument Leśniewski says that he applies the principle of contradiction, and men‑ tions no other principles (idem, p . 103) . He gives a reductio ad absurdum, assuming that some judgement “A is B” is a truth which had a beginning, and that there was thus a time t in which “A is B” was not true, although it is true now . He follows: “If at time t the judgement ‘A is B’ was not true, then the judgement ‘A is not B’ was true” (idem, pp . 102-103) . Apart from the question whether “A is not B” is the contradictory of “A is B,” which is

21 Cf . (Betti (2006), section 4) . Identity of sentences may also be understood in a figurative sense, in which case we can consider equiform sentences to be “identical”, presupposing here that we consider only eternal sentences, that is, sentences without indexical terms . Truth and Time 159 needed in the whole argument, Leśniewski assumes here the correctness of the strong principle of excluded third (EMS: of two contradictory judgements, one is true), a principle not endorsed by Kotarbiński. Leśniewski follows by making the statement that he had assumed that “A is B” was false at t, iden‑ tifying “not being true” with “being false”, endorsing thus bivalence (BV) clearly not a principle endorsed by Kotarbiński.22 Leśniewski then argues that, because “A is B” is true at present, “its contradiction ‘A is not B’ is always false and so it was at time t” (idem). Here Leśniewski presupposes that truth is not only eternal but also sempiternal: what is now true, has been true since eternity, and this is precisely what is at stake, which makes the argument as he presents it circular . The circle is not necessary, though . He could also have said that, as “A is not B” is true at t, it is true at any time later than t, as the eternity of truth is not the issue, and this contradicts the first assumption “A is B” is true now. It is also clear, though, that Kotarbiń‑ ski need not accept Leśniewski’s reasoning, for he does neither accept the principle of bivalence nor the strong principle of excluded middle . Finally, as Leśniewski puts it, “No truth can be created!” (idem, p .104) . From this it by no means follows that nothing can be created. Leśniewski’s argument here is mainly directed at Kotarbiński’s thesis that the truth of a judgement is explained in terms of the existence of an object . A judgement may be true, according to Leśniewski, although its object does not exist at present. Although Kotarbiński does not identify existing with presently existing, he does argue that a judgement about a future object is not true, because its existence is undetermined. Leśniewski’s point seems to be that the judgement about a future phenomenon is true at this moment, not be‑ cause of something that exists now, but because of something that exists in the future . In his lectures “On Ethical Skepticism” from 1913‑1914, Twardowski has given a reaction to Kotarbiński’s paper.23 He mentions Leśniewski’s arguments, immediately followed by his own doubts concerning Kotarbiń‑ ski’s core statement that the truth of a judgement which states the existence of a future event is related to the necessity of the future event . This may mean that (1) the necessity of the future event is a necessary and sufficient condition for the truth of the corresponding judgement; this means that the judgement is true only if the future event is necessary; or (2) the necessity

22 Woleński (1990b, p. 193) gives another explanation why Kotarbiński does not need accept Leśniewski’s argument. 23 Izydora Dąmbska writes that the text “On Ethical Skepticism” is based on lectures delivered in 1923‑24 . She also writes that the text of the lecture mentioned above is only to be found in a lecture series on the same topic from 1913‑1914 . Twardowski’s remarks can therefore be considered to be part of the Kotarbiński‑Leśniewski debate. Cf. (Twardowski 1913/1914, p . 254, note) . 160 Maria van der Schaar of the future event follows from the truth of the corresponding judgement (Twardowski 1913‑1914, p .253) . The second thesis would mean that if the judgement is true, the event must happen . Against (1), Twardowski argues that he sees no reason to assume that the future event has to be necessary in order for the judgement to be true; the fact that the event will take place can be understood as a sufficient condition for the truth of the judgement . One need not accept the stronger thesis expressed in (1) . Regarding the second interpretation, Twardowski admits that if the judgement about the future event is true, the event must happen . However, ‘must’ in question does not mean that a future event is causally determined. Similarly, in this case it also suffices to say that if a judgement on the existence of a future event is true, the event will happen. This ‘must’ concerns logical necessity, i.e. the fact of whether a judgement ‘An event will take place’ is true, or not, depends on whether the judgement ‘There is a future event’ is true or not. This mistake results from confusing two things: on the one hand, the possibility of judging in the present whether given judgements are true or false with, on the other hand, the actual truth‑ fulness or falsity of the judgements (idem, p . 254) . Twardowski seems to express here two forms of criticism: on the one hand, the relation between the truth of a judgement and the event in the fu‑ ture that makes it true is not a relation of causality; it is rather what I have called above an internal relation; that is, a conceptual or logical relation . In the last sentence he seems to express another point: although we are not able to judge whether a judgement about a contingent future event is true, this does not imply that the judgement is not in fact true . We have to make a distinction between the truth of a judgement and our epistemic access to it . The fact that we do not know whether a judgement is true or false does not imply that the judgement is neither true nor false . Twardowski took his students to be equal discussion partners, also when they criticised what was most dear to him: the absoluteness of truth . Twardowski would endorse the final sentence of Leśniewski’s paper on truth (Leśniewski 1913b, p. 114): “Absolute and unchanging, indestructible and impossible to create, true eternally and without a beginning, truth!”

Conclusion

Kazimierz Twardowski’s method inspired his students to a way of doing an‑ alytic philosophy that deviates from the better known variants, such as that of the Vienna Circle or British analytic philosophy . It can be characterised by a more positive attitude towards metaphysics as a general science of ob‑ jects that was already present in Twardowski’s work in 1894 . According to Twardowski, metaphysics, and philosophy in general, has to be done from Truth and Time 161 the bottom up, starting with a psychological description of experience and a logical analysis of linguistic facts . Before Meinong, and in opposition to Brentano and Bolzano, he argued that there are objects that do not exist . Although he opposed as a good Brentanist any form of Platonism, he ac‑ knowledged general objects, which are to be explained in particularist terms, in terms of the metaphysical parts or moments of objects . Twardowski’s aim is a scientific philosophy, that is, a philosophy that is exact, clear, and based upon scientific results . The empirical basis for philosophy is to be provided by psychology and linguistics . In this way one can discern a distinction between the content and the object of an act, and between the meaning of a name and the object named, which are starting points for every philosophical system . As for Brentano, Twardowski’s aim is to build a universal philosophical grammar in Leibniz’ sense, whose ele‑ ments are obtained by a combination of psychological, linguistic and logical analysis . Twardowski’s sensitivity to grammatical distinctions, such as the distinction between the internal and the external object of a sentence, and the distinction between modifying and determining adjectives, makes it possible for him to develop a unique way of doing analytic philosophy, in which these distinctions are used to disambiguate important philosophical concepts, such as judgement and presented object . Twardowski understood that philosophy is to be separated from psy‑ chology and linguistics . Psychological analysis, an analysis of mental acts and their contents, is not identical with metaphysical analysis, an analysis of objects in wholes and parts, or with logico‑linguistic analysis, an analy‑ sis of sentences aiming at the deep structure of thought . At the same time, these three forms of analysis are related to each other insofar as the struc‑ ture of the act’s content is identical with the structure of the object of the act . Furthermore, because the meaning is identified with the act’s content, a logico‑linguistic analysis is needed to see what the content precisely is . Although there is an analogy between language and thought, one has to keep them apart in order to understand that a philosophical grammar is universal . If one does not distinguish between the sentence and the judge‑ ment it intimates, one may think that truth is relative to time, an important theme for Twardowski . The fact that psychology as a science is of impor‑ tance for Twardowski made it possible for him to approach philosophical questions in a scientific manner, and to understand the role of mental acts, the act of judgement and knowing, in questions of truth and knowledge . It is not to be denied, though, that there are also psychologistic elements in Twardowski’s philosophy that may be regarded more critically, such as his psychologism in semantics . In his mature writings, he tried to prevent such a form of psychologism by introducing the distinction between actions and products . The judgement product, not the act of judgement, is the bearer of 162 Maria van der Schaar truth and falsity, and the meaning of a sentence can be explained in terms of the judgement product . The latter notion is essentially the result of an act of judging, and thereby brings in the notion of judging agent . Twardowski is thus able to argue for a less subjectivist semantics, in which the mental act has not completely disappeared . Precisely because products are the re‑ sult of actions, Twardowski’s account of the bearer of truth and falsity is unique and of interest, although there is a tension between his idea that the bearers of truth and falsity are products of mental acts, and his thesis that truth is timeless and absolute . Twardowski’s scientific attitude towards metaphysics, semantics and philosophy in general, and his fascination for the absoluteness of truth has had a decisive influence on his students, reaching beyond his direct students to Alfred Tarski . The topics of semantics, truth and the correspon‑ dence definition of truth led to the question what the bearer of truth and falsity is . The Lvov‑Warsaw School understands the bearers of truth and falsity to be sentences rather than propositions outside space and time . As the sentence is always interpreted, the nominalism of Twardowski’s school is to be distinguished from a purely formalistic approach . It has rather its origin in Brentano’s particularist ontology, and in Twardowski’s thesis that the bearer of truth and falsity is the judgement product, a created entity . Ac‑ cording to Twardowski, the correspondence theory of truth and the absolute notion of truth imply the validity of the logical principles of contradiction and excluded middle . His students took the liberty to doubt the validity of some variants of these principles without questioning the correspondence definition of truth . Eventually this meant that some of Twardowski’s students did not defend the absoluteness of truth, while others scorned his psycho‑ logical approach to the bearer of truth and falsehood and to semantics . It is especially the discussion around truth and time that shows that Twardowski must have stimulated his students to develop a philosophy, in which they dared to question the logical principles . In this open intellectual attitude Twardowski seems to have learnt more from Bolzano than from Brentano . BIBLIOGRAPHY

Albertazzi, L . & M . Libardi (1996) . The School of Franz Brentano. Dordrecht: Kluwer . Aristotle (1979) . Categories and De Interpretatione . Translated by J L. . Ackrill . Oxford: Oxford University Press . Audi, R. (1994). ‘Dispositional Beliefs and Dispositions to Believe. Nous 28, 419‑434 . Austin, J . L . ([1962] 1964) . Sense and Sensibilia. Oxford: Oxford University Press . Bergmann, J . (1879) . Reine Logik, being the first part of the Allgemeine Logik. Berlin: E .S . Mittler und Sohn . Betti, A . & M . van der Schaar (2004) . The Road from Vienna to Lvov . Twardowski’s Theory of Judgement between 1894 and 1897 . Grazer Philosophische Studien 67, 1‑20 . Betti, A. (2006a). Sempiternal Truth. The Bolzano‑Twardowski‑Leśniewski Axis. In: J. Jadacki & J. Paśniczek (eds.). The Lvov‑Warsaw School – The New Generation (Poznań Studies in the Philosophy of the Sciences and the Humanities, vol . 89), pp . 371‑399 . Amsterdam: Rodopi . Betti, A . (2006b) . The Strange Case of Savonarola and the Painted Fish . On the Bolzanization of Polish Thought. In: A. Chrudzimski & D. Łukasiewicz (eds.). Actions, Products and Things. Brentano and Polish Philosophy, pp . 55‑81 . Frankfurt: Ontos . Betti, A . (2013) We Owe it to Sigwart! A New Look at the Content/Object Distinction in Early Phenomenological Theories of Judgement from Brentano to Twardowski . In: M . Textor (ed .), Judgement and Truth in Early Analytic Philosophy and Phenomenology, pp . 74‑96 . Basingstoke: Palgrave MacMillan . Bolzano, B . (1837) . Wissenschaftslehre. Edited by J . Berg . In: Bernard Bolzano‑Gesamtaus‑ gabe . Edited by E . Winter & J . Berg . Stuttgart‑Bad Cannstatt: Frommann, 1985 & 1987 . Page references are to the four volume edition by W . Schulz . Leipzig: Felix Meiner, 1929‑1931 . Brandl, J . L . (1996) . Kazimierz Twardowski über Funktionen und Gebilde . Conceptus 39, 145‑156 . Brentano ([1862] 1960) . Von der mannigfachen Bedeutung des Seienden nach Aristoteles. Hildesheim: Georg Olms . Brentano, F . ([1870] 2014) . Logik . Manuscript EL 80 . Edited by R . Rollinger . Franz Brentano Archiv Graz . Available at: http://gams uni‑graz. at/archive/objects/context:bag/methods/. sdef:Context/get?mode=logik Brentano, F . (1874) . Psychologie vom empirischen Standpunkt. 2 vols . Hamburg: Felix Meiner, 1924, 1973/1971) . Brentano, F . (1883) . Miklosich über subjektlose Sätze . In: Brentano (1874), vol . 2, pp . 183‑196 . Brentano, F . ([1889a] 1969) . Vom Ursprung sittlicher Erkenntnis. Hamburg: Felix Meiner . Brentano, F . (1889b) . Über den Begriff der Wahrheit . In: O . Kraus (ed .), Wahrheit und Evidenz . Hamburg: Felix Meiner, 1930/1974 . Brentano, F . (1895) . Meine letzten Wünsche für Österreich. Stuttgart: J .G . Cotta . Brentano, F . ([1911] 1971) . Anhang zur “Klassifikation der psychischen Phänomene” . In: Psychologie vom empirischen Standpunkt, vol . 2, pp . 133‑277 . Hamburg: Felix Meiner . Brentano, F . ([1914] 1968) . Psychologie vom empirischen Standpunkt, vol . 3 . Edited by O . Kraus & F . Mayer‑Hillebrand . Hamburg: Felix Meiner Verlag . Brentano, F . ([1925] 1970) . Versuch über die Erkenntnis. Hamburg: Felix Meiner . Brentano, F . (1952) . Grundlegung und Aufbau der Ethik . Edited by F . Mayer‑Hillebrand . Bern: Francke . Brentano, F . (1956) . Die Lehre vom richtigen Urteil. Bern: Francke . Brentano, F . (1982) . Deskriptive Psychologie. Edited by R . Chisholm & W . Baumgartner . Hamburg: Felix Meiner . 164 Bibliography

Brożek, A. (2011). Kazimierz Twardowski. Die Wiener Jahre. Vienna: Springer . Buczyńska‑Garewicz, H. (1980). Twardowski’s Idea of Act and Meaning. Dialectics and Hu‑ manism 3, 153‑64 . Cavallin, J . (1997) . Content and Object. Husserl, Twardowski and Psychologism. Dordrecht: Kluwer . Chrudzimski, A . (2001) . Intentionalitätstheorie beim frühen Brentano. Dordrecht: Kluwer . Crane, T . (2013) . The Objects of Thought. Oxford: Oxford University Press . Dąmbska. I. (1978). François Brentano et la Pensée Philosophique en Pologne: Casimir Twar‑ dowski et son École . Grazer Philosophische Studien 5, 1978, 117‑129 . Dummett, M . (1981) . Frege. Philosophy of Language. 2nd ed . London: Duckworth . Fine, K . (1985) . Reasoning with Arbitrary Objects. Oxford: Basil Blackwell . Fréchette, G . (2010) . Gegenstandslose Vorstellungen. Bolzano und seine Kritiker. Sankt Au‑ gustin: Academia . Frege, G . (1879) . Begriffsschrift. Halle: L . Nebert . Frege, G . ([1884] 1987) . Die Grundlagen der Arithmetik. Stuttgart: Reclam . Frege, G . ([1918] 1967) . Der Gedanke . In: Kleine Schriften . Edited by I . Angelelli, pp . 342‑362 . Hildesheim: Georg Olms . Frege, G . ([1915] 1983) . Meine grundlegenden logischen Einsichten . In: H . Hermes, et al . (eds .), Gottlob Frege, Nachgelassene Schriften, pp . 271‑272 . Hamburg: Felix Meiner . Geach, P . ([1965] 1972) . Assertion . In: Logic Matters, pp . 254‑269 . Oxford: Basil Blackwell . Orginally published in Philosophical Review 74, 449‑465 . Grossmann, R . (1977) . Introduction: In: Twardowski (1894e), pp . vii‑xxxiv . Haas, W . (1989) . On “Categorical Grammar” . In: K . Szaniawski (ed .), The Vienna Circle and the Lvov‑Warsaw School, pp . 139‑164 . Dordrecht: Kluwer . Hamlyn, D .W . (1993) . Aristotle. De Anima Books II and III . Translated with introduction and notes by D . W . Hamlyn . Oxford: Clarendon Press . Herbart, J F. . ([1813] 1912) . Lehrbuch zur Einleitung in die Philosophie . 4th ed . Leipzig: Felix Meiner . Hickerson, R . (2005) . Getting the Quasi‑Picture: Twardowskian Representationalism and Hus‑ serl’s Argument Against It . Journal of the History of Philosophy 43, 461‑480 . Hickerson, R . (2007) . The History of Intentionality. Theories of Consciousness from Brentano to Husserl. London: Continuum . Höfler, A . (1890) . Logik . Prepared together with A . Meinong . Prague, Vienna, Leipzig: Temp‑ sky, Freytag . Hume, D . ([1739] 2009) . A Treatise of Human Understanding. Edited by D F. . Norton & M .J . Norton . Oxford: Oxford University Press . Husserl, E . ([1890] 1970) . Zur Logik der Zeichen (Semiotik) . In: L . Eley (ed ),. Philosophie der Arithmetik mit ergänzenden Texten, pp . 340‑373 . The Hague: Martinus Nijhoff . Husserl, E . (1894) . Intentionale Gegenstände . Edited by K . Schuhmann, Brentano Studien 3, 142‑76 . Although there are some later additions, probably from the year 1900, the relevant material can be dated as 1894 . Another edition is available in: Hua XXII & B . Rang (ed .), Aufsätze und Rezensionen, pp . 303‑348 . The Hague: Martinus Nijhoff . Husserl, E . (1896) . Besprechung von K . Twardowski, Zur Lehre vom Inhalt und Gegenstand der Vorstellungen . In: Hua XXII & B . Rang (eds .), Aufsätze und Rezensionen, pp . 349‑356 . The Hague: Martinus Nijhoff . Husserl, E . ([1901] 2009) . Logische Untersuchungen (Hamburg: Felix Meiner) Zweiter Band, nach Husserliana XIX/ 1‑2 . Edited by U . Panzer . The Hague: Martinus Nijhoff, 1984 . Husserl, E . ([1902] 1994) . Letter to A . Meinong, 5 April 1902 . In: E . Shuhmann & K . Shuh‑ mann (eds ),. , Briefwechsel, Bd . I, Die Brentanoschule, pp . 139‑145 . Dordrecht: Kluwer . Bibliography 165

Jacquette, D . (1991) . The Origins of Gegenstandstheorie: Immanent and Transcendent In‑ tentional Objects in Brentano, Twardowski, and Meinong . Brentano Studien 3, 177‑202 . Jadacki, J . (1999) . On Kazimierz Twardowski’s Descriptive Semiotics and Its Metaphysical Basis . In: J . Jadacki, Polish Analytical Philosophy, pp . 175‑199. Warsaw: Semper . Jadacki, J . (2005) . Kazimierz Twardowski. Filozofia i muzyka [Kazimir Twardowski. Philoso‑ phy and Music]. Warsaw: Semper . Kotarbiński, T. ([1913] 1968). The Problem of the Existence of the Future. The Polish Review 13, 7‑22 . Künne, W . (2003a) . Bernard Bolzano’s Wissenschaftslehre and Polish analytical philosophy between 1894 and 1935 . In: J . Hintikka et al. (eds ),. Philosophy and Logic. In Search of the Polish Tradition, pp . 179‑192 . Dordrecht: Kluwer . Künne, W . (2003b) . Conceptions of Truth. Oxford: Clarendon Press . Leibniz, G W. . ([1684] 1999) . Meditationes de cognitio, veritate et ideis . In: G W. . Leibniz, Sämtliche Schriften und Briefe, 6 Reihe, 4 Band, Teil A: Philosophische Schriften. Edited by H . Schepers et al, pp . 585‑592 . Berlin: Akademie Verlag . Leibniz, G .W . ([undated] 1965) . Characteristica universalis / On the Universal Science: Char‑ acteristic . In: P . Schrecker & A .M . Schrecker (eds .), Monadology and Other Philosophical Essays, pp . 11‑21 . Indianapolis: Bobbs‑Merrill . Leśniewski, S. (1913a). The Critique of the Logical Principle of the Excluded Middle. In: S .J . Surma et al. (eds .), Collected Works, vol . I, pp . 47‑85 . Dordrecht: Kluwer, 1992 . Leśniewski, S. (1913b). Is All Truth only True Eternally or Is It also True without a Beginning. In: S .J . Surma et al. (eds .), Collected Works, vol . I, pp . 86‑114 . Dordrecht: Kluwer, 1992 . Locke, J . ([1690] 1990) . An Essay Concerning Human Understanding . Edited by P .H . Nidditch . Oxford: Clarendon Press . Łukasiewicz, J. ([1910a] 1993). Über den Satz des Widerspruchs bei Aristotele. Translated by J . Barski . Hildesheim: Georg Olms . Łukasiewicz, J. ([1910b] 1987). On the Principle of the Excluded Middle. Translated and edited by J. Woleński & P. Simons. History and Philosophy of Logic 8, 67‑69 . Łukasiewicz, J. (1913). Die logischen Grundlagen der Wahrscheinlichkeitsrechnung. Kraków: Akademie der Wissenschaften . References are to “Logical Foundatons of Probability The‑ ory”, in: Łukasiewicz (1970), 16‑63. Łukasiewicz, J. (1918). Farewell Lecture. In: Łukasiewicz (1970), pp. 84‑86. Łukasiewicz, J. (1921). Two‑Valued Logic. In: Łukasiewicz (1970), pp. 89‑109. Łukasiewicz, J. (1961). On Determinism. In: Łukasiewicz (1970), 110‑128. Łukasiewicz, J. (1970). Selected Works . Edited by L . Borkowski . Amsterdam, London: North‑Holland Publishing Company . MacFarlane, J . (2003) . Future Contingents and Relative Truth . The Philosophical Quarterly 53, 321‑336 . Mach, E . ([1886] 1987) . Die Analyse der Empfindungen. Darmstadt: Wissenschafliche Buch‑ gesellschaft . Martin, W . M . (2006) . Theories of Judgement. Cambridge: Cambridge University Press . Marty, A . (1884) Über subjektlose Sätze und das Verhältnis der Grammatik zu Logik und Psychologie I, II, and III . Vierteljahrsschrift für wissenschaftliche Philosophie 8 . I have used the edition in Marty (1918) . Marty, A . (1892) . Review of “William James, The Principles of Psychology”. Translated by R .D . Rollinger . In: R .D . Rollinger, Philosophy of Language and Other Matters in the Work of Anton Marty, pp . 255‑299 . Amsterdam: Rodopi, 2010 . Marty, A . (1893) . Über das Verhältnis von Grammatik und Logik . In: Symbolae Pragensis, pp . 99‑126 . Prague, Vienna: F . Tempsky . 166 Bibliography

Marty, A . (1894) . Über subjektlose Sätze und das Verhältnis der Grammatik zu Logik und Psychologie IV, V, VI, and VII . Vierteljahrsschrift für wissenschaftliche Philosophie 8 . I have used the edition in Marty (1918) . Marty, A . (1908) . Untersuchungen zur Grundlegung der allgemeinen Grammatik und Sprach‑ philosophie. Halle: Niemeyer . Marty, A . (1918) . Gesammelte Schriften . Edited by J . Eisenmeier et al. Halle: Max Niemeyer . Marušić, J. S. (2014). Propositions and Judgements in Locke and Arnauld: A Monstrous and Unholy Union? Journal of the History of Philosophy 52, 255‑290 . Meinong, A . (1894) . Beiträge zur Theorie der psychischen Analyse. Hamburg, Leipzig: Leo‑ pold Voss . Meinong, A . (1889) . Phantasievorstellung und Phantasie . Zeitschrift für philosophische Kritik 95, 161‑244 . Published also in: Gesamtausgabe, vol . I . Edited by R . Haller, pp . 193‑271 . Graz: Akademische Druck‑ u . Verlagsanstalt, 1971 . Meinong, A . (1899 ) . Über Gegenstände höherer Ordnung und deren Verhältnis zur inneren Wahrnehmung . In: Gesamtausgabe, vol . II . Edited by R . Haller, pp . 377‑480 . Graz: Aka‑ demische Druck‑ u . Verlagsanstalt, 1971 . Meinong, A . (1902) . Über Annahmen. Zeitschrift für Psychologie und Physiologie, Ergän‑ zungsband 2 . Meinong, A . (1904) . Über Gegenstandstheorie . In: Gesamtausgabe, vol . II . Edited by R . Haller, pp . 481‑530 . Graz: Akademische Druck‑ u . Verlagsanstalt, 1971 . Miklosich, F . (1883) . Subjectlose Sätze. Vienna: Wilhelm Braumüller . Mill, J .S . ([1843] 1973, 1947) . A System of Logic. Collected Works of John Stuart Mill, vols 7 and 8 . Edited by J .M . Robson . Toronto, London: University of Toronto Press, Routledge . Misiuna, K. (1998). Truth and Time. In: K. Kijania‑Placek & J. Woleński (eds.), The Lvov‑War‑ saw School and Contemporary Philosophy, pp . 199‑208 . Dordrecht: Kluwer . Miśkiewicz, W. (2009). Leopold Blaustein’s Analytical Phenomenology. In: S. Lapointe et al. (eds .), The Golden Age of Polish Philosophy, pp . 181‑188 . Dordrecht: Springer . Moore, G .E . (1898) . The 1898 dissertation . The Metaphysical Basis of Ethics . In: T . Baldwin & C . Preti (eds ),. G.E. Moore. Early Philosophical Writings, pp . 115‑242 . Cambridge: Cambridge University Press . Moore, G .E . (1899) . The Nature of Judgement . Mind 8, 176‑93 . Mulligan, K . (1986) . Exactness, Description and Variation: How Austrian Analytic Philos‑ ophy Was Done . In: J C. . Nyiri (ed ),. Von Bolzano zu Wittgenstein, pp . 86‑97 . Vienna: Hölder‑Pichler‑Tempsky . Mulligan, K . (1987) . Promisings and Other Social Acts: Their Constituents and Structure . In: K . Mulligan (ed ),. Speech Act and Sachverhalt. Reinach and the Foundations of Realist Phenomenology, pp . 29‑90 . Dordrecht: Martinus Nijhoff . Mulligan, K . (1990) . Marty’s Philosophical Grammar . In: K . Mulligan (ed ),. Mind, Meaning and Metaphysics, pp . 11‑27 .Dordrecht: Kluwer . Nelson, L . ([1908] 1973) . Über das sogenannte Erkenntnisproblem. In: Gesammelte Schriften . Edited by P . Bernays, vol . II, pp . 59‑393 . Hamburg: Felix Meiner . Pakszys, E . (1998) . Women’s Contributions to the Achievements of the Lvov‑Warsaw School: a Survey. In: K. Kijania‑Placek & J. Woleński (eds.), The Lvov‑Warsaw School and Con‑ temporary Philosophy, pp . 55‑71 . Dordrecht: Kluwer . Parsons, T . (1970) . Some Problems concerning the Logic of Grammatical Modifiers . Synthese 21, 320‑34 . Partee, B . (2010) . Privative Adjectives: Subsective plus Coercion . In: R . Bäuerle et al. (eds .), Presuppositions and Discourse. Essays offered to Hans Kamp, pp . 273‑285 . Bradford: Emerald . Bibliography 167

Placek, T . (2006) . A Puzzle about Semantic Determinism . In: J . Jadacki & J . Pasniczek (eds .), The Lvov‑Warsaw School – The New Generation (Poznań Studies in the Philosophy of the Sciences and Humanities, vol . 89), pp . 171‑185 . Amsterdam, New York: Rodopi . Poli, R . (1993) . Twardowski’s Theory of Modification against the Background of Traditional Logic . Axiomathes 4, 41‑57 . Quine, W .V .O . ([1960] 1989) . Word and Object. Cambridge, MA: The M .I .T . Press . Reinach, A . ([1911] 1982) . On the Theory of the Negative Judgement . In: B . Smith (ed .), Parts and Moments, pp . 315‑400 . München,Wien: Philosophia Verlag . Rojszczak, A . (2005) . From the Act of Judging to the Sentence. Edited by J. Woleński. Dor‑ drecht: Springer . Rojszczak, A . & B . Smith (2003) . Truthmakers, Truthbearers and the Objectivity of Truth . In: J . Hintikka et al. (eds .), Philosophy and Logic. In Search of the Polish Tradition, pp . 229‑268 . Amsterdam: Kluwer . Rollinger, R D. . (1996) . Husserl’s Position in the . Utrecht: Ph . D . disser‑ tation . Rollinger, R D. . (2009) . Brentano’s Psychology and Logic and the Basis of Twardowski’s Theory of Presentations . The Baltic International Yearbook of Cognition, Logic and Com‑ munication 4, 1‑23 . Russell, B . ([1903] 1992) . The Principles of Mathematics. London: Routledge . Russell, B . ([1912]) . The Problems of Philosophy. Oxford: Oxford University Press . Russell, B . (1918) . The Philosophy of Logical Atomism . In: R C. . Marsh (ed ),. Logic and Knowledge, pp . 177‑281 . London: Allen & Unwin . Russell, B . (1919) . Introduction to Mathematical Philosophy. London: Allen & Unwin . Schaar, M . van der (1996) . From Analytic Psychology to Analytic Philosophy: The Reception of Twardowski’s Ideas in Cambridge . Axiomathes 7, 295‑324 . Schaar, M . van der (2002/2003) . Brentano on Logic, Truth and Evidence . Brentano Studien 10, 119‑50 . Schaar, M . van der (2006) . On the Ambiguities of the Term Judgement . An Evaluation of Twardowski’s Distinction between Action and Product. In: A. Chrudzimski & D. Łukasie‑ wicz (eds ),. Actions, Products and Things. Brentano and Polish Philosophy, pp . 35‑53 . Frankfurt: Ontos Verlag . Schaar, M . van der (2007) . Bolzano on Judgement and Error . The Logica Yearbook 2006 . Prague, 211‑221 . Schaar, M . van der (2009) . Twardowski on Knowledge and the Cognitive Act . Logic, Method‑ ology & Philosophy of Science, vol . 5, pp . 131‑141 . Warsaw: Semper . Schaar, M . van der (2011) . The Cognitive Act and the First‑Person Perspective: An Episte‑ mology for Constructive Type Theory . Synthese 180, 391‑417 . Schaar, M . van der (2013) . G.F. Stout and the Psychological Origins of Analytic Philosophy. Basingstoke: Palgrave Macmillan . Schaar, M . van der (2014a) . Wooden Horses and False Friends . On the Logic of Adjectives . In: K . Mulligan et al. (eds .), The History and Philosophy of Polish Logic. Essays in Honour of Jan Woleński, pp . 103‑116 . Basingstoke: Palgrave Macmillan . Schaar, M . van der (2014b) . Mere Belief and the Etiolations of Language . In: A . Reboul (ed .), Mind, Values, and Metaphysics; Philosophical Papers dedicated to Kevin Mulligan, pp . 163‑181 . New York, Dordrecht, London: Springer . Schaar, M . van der (2015) . The Things We Call True . Grazer Philosophische Studien 91, 43‑61 (in press) . Schnieder, B . (2007) . Mere Possibilities . A Bolzanian Approach to Non‑Actual Objects . Jour‑ nal of the History of Philosophy 45, 525‑550 . 168 Bibliography

Schuhmann, K . (1991a) . Husserls doppelter Vorstellungsbegriff: Die Texte von 1893 . Brentano Studien 3, 119‑36 . Schuhmann, K . (1991b) . “Einleitung” to Husserls Abhandlung “Intentionale Gegenstände” . Brentano Studien 3, 137‑42 . Schuhmann, K . (1993) . Husserl and Twardowski . In: F . Coniglione et al. (eds ),. Polish Sci‑ entific Philosophy: The Lvov‑Warsaw School (Poznań Studies in the Philosophy of the Sciences and the Humanities, vol . 28), pp . 41‑58 . Amsterdam: Rodopi . Simons, P . (2008) . Twardowski on Truth . The Baltic Yearbook of Cognition, Logic and Com‑ munication 4, 1‑14 . Simons, P. (2014). Jan Łukasiewicz. In: Stanford Encyclopedia of Philosophy, downloaded June, 14, 2014 . Smith, B . (1989) . Kazimir Twardowski . In: K . Szaniawski (ed ),. The Vienna Circle and the Philosophy of the Lvov‑Warsaw School, pp . 313‑373 . Dordrecht, Boston, Lancaster: Klu‑ wer . References are to the edition in Smith (1994) . Smith, B . (1994) . Austrian Philosophy. The Legacy of Franz Brentano. Chicago: Open Court . Steinthal, H . (1855) . Grammatik, Logik und Psychologie. Iihre Principien und ihr Verhältnis zu einander. Berlin: F . Dümmler . Stepanians, M . S . (1998) . Frege und Husserl über Urteilen und Denken. Paderborn: Ferdinand Schöningh . Stout, G .F . (1891) . Thought and Language . Mind 16, 181‑205 . Stout, G .F . (1893) . The Philosophy of Mr . Shadworth Hodgson . Proceedings of the Aristotelian Society 2, 107‑20 . Stout, G .F . (1896) . Analytic Psychology . 2 vols . London: Sonnenschein . Stout, G .F . (1899) . A Manual of Psychology. London, New York: University Correspondence College Press . Stout, G F. . (1921) . The Nature of Universals and Propositions . Proceedings of the British Academy 10, 157‑72 . Stumpf, C . (1873) . Über den psychologischen Ursprung der Raumvorstellung. Leipzig: Hirzel . Stumpf, C . (1888) . Syllabus for Logic . In: Rollinger (1996), pp . 241‑260 . Sundholm, G . (2004) . Antirealism and the Roles of Truth . In: I . Niniluoto et al. (eds .), Hand‑ book of Epistemology, pp . 437‑466 . Dordrecht: Kluwer . Szabó, Z . (2003) . Believing in Things . Philosophy and Phenomenological Research 66, 584‑611 . Tarski, A . ([1936] 1956) . The Concept of Truth in Formalized Languages . In: Logic, Semantics, Metamathematics . Translated by J .H . Woodger, pp . 152‑278 . Oxford: Clarendon Press Tarski, A . (1944) . The Semantic Conception of Truth . Philosophy and Phenomenological Research 4, 341‑376 . Thomasson, A . (2000) . After Brentano: A One‑Level Theory of Consciousness . European Journal of Philosophy 8, 190‑209 . Twardowski, K . (1892) . Idee und Perception. Eine erkenntnistheoretische Untersuchung aus Descartes. Vienna: C . Konegen . Twardowski, K . ([1894a] 1982) . Zur Lehre vom Inhalt und Gegenstand der Vorstellungen. Munich: Philosophia Verlag . Twardowski, K . ([1894b] 1977) . On the Content and Object of Presentations . Translated by R . Grossmann . The Hague: Nijhoff . Twardowski, K . (1894/1895) . Logik . Wiener Logikkolleg 1894/95 . Edited by A . Bet‑ ti & V . Raspa . Berlin: Walter de Gruyter . Twardowski, K . (1895) . Opening Lecture at the Lvov University . In: Twardowski (2015), pp . 35‑43 . Bibliography 169

Twardowski, K . (1895/1896) . Logika [Logic] . Lecture notes from the academic year 1895/96, Lwów University . Twardowski, K . (1897a) . Psychology vs . Physiology and Philosophy . In: Twardowski (1999), pp . 41‑64 . Twardowski (1897b) . Letter to Meinong, July, 11, 1897 . In: R . Kindinger (ed .), Philosophen‑ briefe – Aus der wissenschaftliche Korrespondenz von , pp . 143‑144 . Graz: Akademische Druck‑ u . Verlagsanstalt, 1965 . Twardowski (1898a) . On the Classification of Mental Phenomena . In: Twardowski (1999), pp . 65‑72 . Twardowski ([1898b] 1995). Imageries. Translation of “Wyobrażenia i pojęcia” [“Images and Notions”] . Axiomathes 1, 79‑104 . Twardowski, K . (1900a) . On So‑Called Relative Truth . In: Twardowski (1999), pp . 147‑169 . Twardowski, K . (1900b) . On Mistakes of Thinking . In: Twardowski (2015), pp . 91‑120 . Twardowski, K . (1901) . Indicatives, interrogatives, imperatives and optatives . In: J . Jadacki (ed .), Being and Duty. The Contribution of 20th Century Polish Thinkers to the Theory of Imperatives and Norms, pp . 119‑120 . Kraków: Copernicus Center Press, 2013 . Twardowski, K . (1902/1903) . The Theory of Judgements . In: Twardowski (2015), pp . 161‑180 . Twardowski, K . (1903) . The Essence of Concepts . Together with notes from 1923 . In: Twar‑ dowski (1999), pp . 73‑97 . Twardowski, K . (1904/1904) . Ethics, Criminal Law and the Problem of Free Will . In: Twar‑ dowski (2015), pp . 287‑321 . Twardowski, K . (1906a) . On Prejudices . In: Twardowski (2015), pp . 73‑80 . Twardowski, K . (1906b) . Independence of Thinking . In: Twardowski (2015), pp . 81‑90 . Twardowski, K . ([1912a] 1996) . Funktionen und Gebilde . Edited by J . Brandl . Conceptus 39, 157‑89 . This German version of Actions and Products, made by Twardowski, deviates from (1912b), which is a translation of the Polish paper . A shorter German version was published in 1911 . Twardowski, K . (1912b) . Actions and Products . Some Remarks from the Borderline of Psy‑ chology, Grammar and Logic . In: Twardowski (1999), pp . 103‑32 . Twardowski, K . ([1912c] 1976) . The Humanities and Psychology . In: Twardowski (1999), pp . 133‑40 . Twardowski, K . (1913/1914) . Judgements on Future and Epistemological Relativism . In: Twardowski (2015), pp. 250‑254. According to a footnote by Izydora Dąmbska, this text comes from lectures delivered in 1913‑1914 . It is added to the edition of the lectures “On Ethical Skepticism,” which given in Lwów in 1923‑1924, whose translation in published in: Twardowski (2015), pp . 237‑286 . Twardowski, K . (1921) . Symbolomania and Pragmatophobia . In: Twardowski (1999), pp . 261‑270 . Twardowski, K . (1923) . A priori, or Rational (Deductive) Sciences, and a posteriori, Empirical (Inductive) Sciences . In: Twardowski (1999), pp . 171‑179 . Twardowski, K . (1923/1924) . On Ethical Skepticism . Lectures given in Lwów in 1923‑24 . In: Twardowski (2015), 237‑286 . Twardowski, K . (1925) . Theory of Knowledge . A Lecture Course . In: Twardowski (1999), pp . 181‑239 . Twardowski, K . (1926) . Self‑Portrait . In: Twardowski (1999), pp . 17‑31 . Twardowski, K . (1927) . On the Logic of Adjectives . In: Twardowski (1999), pp . 141‑143 . Twardowski, K . (1929) . Address at the 25th anniversary session of the Polish Philosophical Society . In: Twardowski (1999), pp . 271‑276 . Twardowski, K . (1933) . On the Dignity of the University . In: Twardowski (1999), pp . 277‑286 . 170 Bibliography

Twardowski, K . (1999) . On Actions, Products and Other Topics in Philosophy (Poznań Studies in the Philosophy of the Sciences and the Humanities, vol . 67) . Edited by J.L. Brandl & J. Woleński. Amsterdam: Rodopi. Twardowski, K . (2015) . On Prejudices, Judgements, and Other Topics in Philosophy (Poznań Studies in the Philosophy of the Sciences and the Humanities, vol . 102) . Edited by A. Brożek & J. Jadacki. Amsterdam: Rodopi. Weidemann, H . (1994) . Aristoteles, Peri Hermeneias . Translation and commentary by H . Wei‑ demann . Berlin: Akademia Verlag . Witasek, S . (1908) . Grundlinien der Psychologie. Leipzig: Dürr . Woleński, J. (1989) Logic and Philosophy in the Lvov‑Warsaw School. Dordrecht: Kluwer . Woleński, J. & P. Simons (1989). De Veritate. Austro‑Polish Contributions to the Theory of Truth from Brentano to Tarski . In: K . Szaniawski (ed ),. The Vienna Circle and the Lvov‑Warsaw School, pp . 391‑442 . Dordrecht: Kluwer . Woleński, J. (1990). Marty and the Lvov‑Warsaw School. In: K. Mulligan (ed.), Mind, Meaning and Metaphysics, pp . 215‑223 . Dordrecht: Kluwer . Woleński, J. (1990b). Kotarbiński, Many‑Valued Logic, and Truth. In: J. Woleński (ed.), Ko‑ tarbiński: Logic, Semantics and Ontology, pp . 191‑197 . Dordrecht: Kluwer . Woleński, J. (2009). The Rise and Development of Logical Semantics in Poland. In: S. La‑ pointe et al., The Golden Age of Polish Philosophy, pp . 43‑59 . Dordrecht: Springer . Zimmermann, R . (1853) . Philosophische Propaedeutik für Obergymnasien (Vienna: W . Braun‑ müller) . The consulted edition, which also contains the second edition from 1860, is: Robert Zimmermanns Philosophische Propädeutik und die Vorlagen aus der wissenschafts‑ lehre Bernard Bolzanos . Edited by W . Winter . Vienna: Österreichischen Akademie der Wissenschaften, 1975 . NAME INDEX

Ajdukiewicz, Kazimierz 13, 14 Geach, Peter Thomas 55, 68, 108 Aquinas, Thomas 11, 34 Gettier, Edmund L . 117, 119, 120, 122, 124 Aristotle 11, 14, 19, 39, 41, 42, 47‑49, 59, 91, Grossmann, Reinhardt 57, 65 110, 112, 115, 125, 129‑131, 133, 144, 146, 154, 157 Haas, William 13 Audi, Robert 110 Heidegger, Martin 145 Austin, John Langshaw 9, 39, 102 Herbart, Johann Friedrich 10, 20 Hickerson, Ryan 56, 58 Baley, Stefan 12 Hilbert, David 13, 130 Beaney, Michael 70 Höfler, Alois 10, 11, 50, 53, 54, 65, 98, 126 Bergmann, Julius 104 Hosiasson‑Lindenbaumowa, Janina 13 Betti, Arianna 19, 46, 98‑100, 158 Humboldt, Wilhelm von 24‑26 Blaustein, Leopold 12 Hume, David 11, 20, 84, 89, 92, 125 Bolzano, Bernard 7, 10, 11, 14, 15, 20, 29, 31, Husserl, Edmund 7, 13, 31, 55‑57, 61‑68, 76, 34, 46‑48, 53, 54, 56, 58, 59, 62, 91, 99, 77, 80, 87, 90, 91, 95, 96, 102, 103, 108, 104, 105, 109, 110, 119‑123, 138, 139, 149, 111 150, 161, 162 Brentano, Franz 7, 9‑12, 14, 16‑22, 24‑28, 31, Jacquette, Dale 65 39, 43, 45‑49, 51, 53‑55, 57, 59, 60, 62, Jadacki, Jacek 10 65‑67, 69, 73, 85‑94, 96‑100, 105, 118‑121, Jadczak, Ryszard 22 123, 124, 132, 133, 135‑137, 150, 155, 157, 161, 162 Kant, Immanuel 21, 50, 110, 125, 126, 135 Brouwer, Luitzen Egbertus Jan 157 Kerry, Benno 30, 53, 54 Brożek, Anna 10, 12, 13 Kokoszyńska‑Lutmanowa, Maria 12 Brugmann, Karl 33 Kołodziejska‑Twardowska, Kazimiera 10 Buczyńska‑Garewicz, Hanna 106 Kotarbiński, Tadeusz 8, 12‑14, 137, 143, 147, 151, 152, 154‑159 Cajetan, Thomas 48 Künne, Wolfgang 14, 138, 149 Carnap, Rudolf 31, 152 Cavallin, Jens 56 Leibniz, Gottfried Wilhelm 9, 19, 31, 32, 161 Chrudzimski, Arkadiusz 19, 54 Leśniewski, Stanisław 8, 13, 14, 69, 129, 130, Chwistek, Leon 14 137, 143, 146, 151, 152, 157‑160 Crane, Tim 89 Lindenbaum, Adolf 13 Locke, John 57, 74‑76, 79, 125 Dąmbska, Izydora 12, 14, 122, 159 Łukasiewicz, Jan 8, 13, 14, 102, 103, 130, Descartes, René 11, 85 136‑138, 143‑149, 151‑155, 157 Dummett, Michael 151 Dzieduszycki, Wojciech 10 MacFarlane, John 151 Martin, Wayne M . 21 Eliot, Thomas Stearns 39 Marty, Anton 9, 11, 24‑28, 30, 31, 45, 48, 78 Marušić, Jennifer Smalligan 109 Fine, Kit 74‑76 Meinong, Alexius von 22, 50, 54, 55, 61, 69, Fréchette, Guillaume 52, 53 73, 81, 89‑91, 93, 94, 98, 100, 102, 126, Frege, Gottlob 11, 14, 20, 26, 27, 35, 61, 71, 133, 144, 147, 149, 155, 161 73, 76, 82, 86‑88, 90, 91, 108, 109, 136, Miklosich, Franz 25 141, 146, 148, 152, 157 Mill, John Stuart 28, 47, 48, 125 172 Name Index

Misiuna, Krystyna 151 Schnieder, Benjamin 41, 45, 46 Miśkiewicz, Wioletta 12, 22 Schuhmann, Karl 62, 67 Moore, George Edward 35, 55 Simons, Peter 14, 129, 131, 150 Mulligan, Kevin 9 Smith, Barry 22, 24, 144 Steinthal, Heymann 24 Nelson, Leonard 125, 126 Stepanians, Markus S . 46 Stout, George Frederick 55, 71, 77 Ockham, William 48 Stumpf, Carl 11, 33, 69, 97, 100, 102, 104, 105 Szabó, Zoltán 92, 93 Pakszys, Elżbieta 13 Szaniawski, Klemens 13 Partee, Barbara H . 43‑45 Szymborska, Wisława 5 Pelc, Jerzy 13 Placek, Tomasz 12, 153 Tarski, Alfred 13, 14, 69, 129‑132, 162 Poli, Roberto 41 Twardowski, Kazimierz (Kasimir) passim

Ribot, Théodule 22 Weidemann, Hermann 49 Rojszczak, Artur 14, 102, 130, 144 Witasek, Stephan 92, 104 Rollinger, Robin D . 19, 54, 67 Wittgenstein, Ludwig 87 Russell, Bertrand 7, 13, 14, 26, 28, 29, 34, 35, Woleński, Jan 12‑14, 35, 69, 129, 130, 131, 55, 60, 61, 90, 91, 102, 113‑116, 132‑134, 155, 159 147, 148, 152 Wundt, Wilhelm 11, 22, 70

Savonarola, Girolamo 46, 47 Zimmermann, Robert 10, 11, 53 Shaar, Maria van der passim