Actions, Products, and Things: Brentano and Polish Philosophy 9783110325706, 3938793066, 9783110325065

Citation preview

Arkadiusz Chrudzimski / Dariusz Łukasiewicz (Eds.) Actions, Products, and Things Brentano and Polish Philosophy

PHENOMENOLOGY & MIND Herausgegeben von / Edited by Arkadiusz Chrudzimski • Wolfgang Huemer Band 8 / Volume 8

Arkadiusz Chrudzimski Dariusz Łukasiewicz (Eds.)

Actions, Products, and Things Brentano and Polish Philosophy

ontos verlag Frankfurt I Paris I Ebikon I Lancaster I New Brunswick

Bibliographic information published by Die Deutsche Bibliothek Die Deutsche Bibliothek lists this publication in the Deutsche Nationalbibliographie; detailed bibliographic data is available in the Internet at http://dnb.ddb.de

North and South America by Transaction Books Rutgers University Piscataway, NJ 08854-8042 [email protected]

United Kingdom, Ire, Iceland, Turkey, Malta, Portugal by Gazelle Books Services Limited White Cross Mills Hightown LANCASTER, LA1 4XS [email protected]

Livraison pour la France et la Belgique: Librairie Philosophique J.Vrin 6, place de la Sorbonne ; F-75005 PARIS Tel. +33 (0)1 43 54 03 47 ; Fax +33 (0)1 43 54 48 18 www.vrin.fr



2006 ontos verlag P.O. Box 15 41, D-63133 Heusenstamm nr. Frankfurt www.ontosverlag.com ISBN 3-938793-06-6 2006 No part of this book may be reproduced, stored in retrieval systems or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise without written permission from the Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use of the purchaser of the work Printed on acid-free paper ISO-Norm 970-6 This hardcover binding meets the International Library standard Printed in Germany by buch bücher dd ag

Table of Contents Introduction ......................................................................................... 7 Twardowski, Brentano’s Dilemma, and the Content-Object Distinction ...................................................................... 9 DALE JACQUETTE On the Ambiguities of the Term Judgement. An Evaluation of Twardowski’s Distinction between Action and Product ......................... 35 MARIA VAN DER SCHAAR The Strange Case of Savonarola and the Painted Fish. On the Bolzanization of Polish Thought ............................................................. 55 ARIANNA BETTI Things and Truths: Brentano and Leśniewski, Ontology and Logic ....... 83 PETER SIMONS The Young Leśniewski on Existential Propositions .............................. 107 ARKADIUSZ CHRUDZIMSKI On the Phases of Reism ...........................................................................121 BARRY SMITH Brentanian Philosophy and Czeżowski’s Conception of Existence ....... 183 DARIUSZ ŁUKASIEWICZ Brentanism and the Rise of Formal Semantics ........................................217 JAN WOLEŃSKI Notes on Contributors ...................................................................... 233 Index of Names ................................................................................ 235

Introduction For a long time Franz Brentano has been widely perceived almost exclusively as the re-discoverer of intentionality and the founder of the continental phenomenology. It was only during the last 30 years that his immense importance for the development of analytic philosophy (and also the arbitrariness of the very division between analytic and continental philosophy) became clear. This volume is devoted to Brentano’s influence on the Polish Analytic Philosophy better known under the name of: “Lvov-Warsaw School”. The founder of this school – Kazimierz Twardowski was himself a student of Brentano. He took over Brentano’s intentionality thesis as well as many other elements of his philosophy (e.g. his non-propositional theory of judgement or the conviction that psychology is the only acceptable basis of any scientific philosophy), but at the same time, as early as in his doctoral dissertation On the Content and Object of Presentations, he severely criticised Brentano’s central idea of an ‘immanent object’. The first three papers in this volume centre on this important BrentanoTwardowski connection. Dale Jacquette addresses the aforementioned critique by Twardowski and elucidates his important distinction between content and object. Maria van der Schaar analyses Twardowski’s later development of the notion of content, which remained influenced by Husserl, and Arianna Betti argues that many aspects of Polish analytical philosophy could be better understood, if we focus rather on the traces of Bolzano’s thought in Twardowski’s philosophy. The next two essays concern the philosophy of Stanisław Leśniewski, who is (beside Alfred Tarski and Jan Łukasiewicz) probably the most important Polish philosopher. Peter Simons traces important parallels between Brentano and Leśniewski, focusing mainly on reistic or particularist ideas which are relevant for the late Leśniewski, but beginning with his early critique of Brentano’s non-propositional theory of judgement in “A Contribution to the Analysis of Existential Propositions”. While Simons writes that “[t]he paper on existential propositions is […], apart from some of its incidental features, a complete mess” (p. 87 in this volume), Arkadiusz Chrudzimski tries to concentrate on these incidental features and make some sense of them. Actions, Products, and Things. Brentano and Polish Philosophy. Arkadiusz Chrudzimski and Dariusz Łukasiewicz (eds.), Frankfurt: ontos, 2006, 7–8.

8

INTRODUCTION

Various facets of the reistic approach are also investigated by Barry Smith. He focuses mainly on Tadeusz Kotarbiński but also outlines some systematic relations between several kinds of reism. In the next paper Dariusz Łukasiewicz describes the evolution of Tadeusz Czeżowski’s views concerning the concept of existence, which are closely connected with the Brentanian non-propositional theory of judgement. If we were to choose the single most important and influential achievement of Polish philosophy, then the choice would most probably be the semantic definition of truth formulated by Alfred Tarski. In the last article of this volume Jan Woleński argues that Tarski’s discovery of semantics may have been influenced also by his remotely Brentanian background. First and foremost, we would like to thank all of the contributors who have made this collection possible. Our particular thanks go to Phillip Meadows for his valuable help with proofreading the English contributions. Most papers were written for this volume. Barry Smith’s article “On the Phases of Reism” was previously published (in: J. Woleński (ed.), Kotarbiński: Logic, Semantics and Ontology, Dordrecht / Boston / London: Kluwer, pp. 137–184). We would like to thank Springer Verlag for the kind permission to reprint this material. The work of Arkadiusz Chrudzimski was supported by the Austrian Fonds zur Förderung der wissenschaftlichen Forschung (FWF). This book is dedicated to the memory of Artur Rojszczak (1968–2001).

The editors Salzburg and Bydgoszcz February 2006

Twardowski, Brentano’s Dilemma, and the Content-Object Distinction DALE JACQUETTE

1. The Brentano School The students of Franz Brentano were independent-minded thinkers who found inspiration in Brentano’s teachings for their own tangential philosophical pursuits. Kazimierz Twardowski, the leading and first member of the Polish branch of Brentano’s school, is a prime example of the combination of partial loyalty to and dissent from certain of Brentano’s doctrines by philosophers who nevertheless considered themselves to be true Brentanians.1 Twardowski adopts Brentano’s central thesis of intentionality as the distinguishing feature of mental phenomena.2 At the same time, he is one of the outstanding ringleaders of the early breakaway group of descriptive psychologists within the Brentano circle who took exception to Brentano’s doctrine of immanent intentionality.3 The theory of immanent intentionality or intentional in-existence expresses Brentano’s insight that the intended objects of thought belong to and are contained within the thoughts by which they are intended. Intentional in-existence is not nonexistence, but rather existence in the psychological state by which an object is intended. Twardowski led his generation in advancing a distinction between the immanent content of thought and its mind-transcending intended objects that went beyond and in some ways contradicted Brentano’s original concept. If Brentano’s (1874) Psychologie vom empirischen Standpunkt set the agenda for turn-of-the-century scientific psychology and phenomenology, it was Twardowski’s (1894) treatise, Zur Lehre vom Inhalt und Gegenstand der Vorstellungen (On the Content and Object of Presentations), that most dramatically freed Brentano’s concept of intentionality from its implausible insupportable commitment to immanentism.4

Actions, Products, and Things. Brentano and Polish Philosophy. Arkadiusz Chrudzimski and Dariusz Łukasiewicz (eds.), Frankfurt: ontos, 2006, 9–33.

10

DALE JACQUETTE

2. Psychological Investigations in Philosophy Contrary to some popular histories of the early Austrian phenomenological movement, Twardowski was not absolutely the first to challenge Brentano’s theory of immanent intentionality.5 Indeed, in Zur Lehre vom Inhalt und Gegestand der Vorstellungen, he acknowledges Alois Höfler and Alexius Meinong in their (1890) Logik as having first proposed distinguishing between the content and object of presentations, and he draws on prior arguments to the same conclusion by Benno Kerry.6 Twardowski nevertheless rightly deserves credit as having been the first to make the content-object distinction the focus of a full-length study that upholds a qualified contra-immanentist version of Brentano’s intentionality thesis. Whatever implications have been attributed to Twardowski’s study for logic, metaphysics and phenomenology, Twardowski himself evidently thought of his project as a contribution to psychology, as the subtitle of the work, Eine psychologische Untersuchung (A Psychological Investigation), often omitted from its references in subsequent philosophical literature, makes abundantly clear. It is only as a psychological investigation, moreover, that the conclusions of Twardowski’s essay can be properly understood.7 Thus, when Twardowski turns to consider his famous four arguments for distinguishing between the immanent contents and thought-transcending objects of presentations, he does not merely appeal to the fact that it is in some way untenable to regard all intended objects as belonging immanently to the thoughts by which they are intended. Rather, he offers specific reasons from within a psychological, introspective or proto-phenomenological standpoint, providing inferences that can be reached internally concerning the direction of psychological acts with specific contents upon particular intended objects.8 The psychological perspective of Twardowski’s investigations into the act-content-object distinction is especially important within the context of his relation to Brentano. It is not merely the fact that Brentano was a pioneer in scientific thinking about the nature of psychological phenomena, nor that he and his students were associated with a new approach to longstanding philosophical problems of psychology. The importance of psychology for Twardowski as a member of Brentano’s school has more to do

THE CONTENT-OBJECT DISTINCTION

11

with a deep commitment to the limitations of empiricism as the only justifiable methodology for discovering truths of fact. The emphasis on psychology in Brentano is combined with a particular attitude toward the possibilities of justifying commonsense beliefs that go beyond the limits of a strict philosophical empiricism. Whether in David Hume and George Berkeley’s day or in Brentano’s time or even today, it is the contents of thoughts with which the empiricist is most intimately and properly concerned as a source of truth concerning the nature of the world. What, if anything, exists beyond what occurs within experience, understood as a stream of immediate sense impressions, is a problem to whatever extent a philosophy aspires to a purely empirical metaphysics and epistemology. Empiricism is a methodology fundamentally based on the contents and limitations of experience as a psychological occurrence, and on what can and cannot be learned from the resources of perception.9

3. Scientific Philosophy and the Content-Object Distinction Brentano’s revolution in the late nineteenth and early twentieth century, in a philosophical milieu dominated by post-Kantian idealism and transcendentalism in the German-speaking world, fostered a return to Aristotle and Hume, in which the methods of natural science would be recognized also as the only proper methods of philosophy.10 Twardowski is in sympathy with Brentano’s scientific orientation.11 It is largely for this reason, in the interests of sustaining a genuinely scientific psychology, that Twardowski opposes Brentano’s immanence theory of intentional in-existence. He draws a sharp distinction that Brentano was not prepared to recognize between the psychological act, its immanent content, and its thought-transcending intended object.12 Like others of Brentano’s students, Twardowski came to believe that there was a tension in Brentano’s scientific empiricism. Insofar as Brentano sought to make philosophy scientific, he was proposing to follow the patterns of inquiry that had been successful in the natural sciences. As a latter-day adherent, moreover, of Aristotle’s (‘naïve’) realist metaphysics, Brentano ought to have acknowledged the commonsense existence of entities outside the mind as the intended objects of thoughts. Hence, he ought

12

DALE JACQUETTE

to have avoided at all costs the proposition that intentionality relates psychological acts to immanently in-existent intended objects that belong to and exist only in the thoughts by which they are intended. As we know from the clear testimony of the Psychologie, Brentano did not recognize the conflict. Insofar as he was also deeply committed to a methodology of strict philosophical empiricism, he no doubt felt that, like Hume before him, the only responsible attitude to take toward the ontic status of intended objects is to consider them merely as they are presented to experience, as something found in thought as an internal part of immediate consciousness. To venture speculatively beyond the limits of phenomenal psychology would have been a betrayal of Brentano’s Humean methodological empiricism in the interests of an Aristotelian metaphysical realism that is also, in a distinct, non-Humean way, empiricist.13 Such a deep conflict needed to be admitted and resolved. Scientific psychology and philosophy was drawn in two opposed directions. Its allegiance to natural science and common sense inclined it toward accepting the existence of intended objects as transcending the contents of thought. Twardowski eventually makes the argument in Zur Lehre vom Inhalt und Gegenstand der Vorstellungen, as other mutineers from the immanent intentionality thesis also came to see. On the other hand, the demands of a strict methodological empiricism implied that philosophy could not compromise with the purely experiential limits of what can be known to exist. It is a historical puzzle that has long provoked Brentano scholars as to why Brentano did not see the difficulty posed by his dual devotion to both the metaphysics of Aristotle as a foundation for the methods of natural science and the strict empiricism that owes its origins more directly to eighteenth century proponents in the writings of John Locke, Hume, Berkeley, and J.S. Mill. The fact remains that it was left to his students, Twardowski chief among them, to recognize that what Brentano had referred to as the immanently intentional in-existence of the intended object was really more appropriately designated the thought’s content [Inhalt], and that its object [Gegenstand] was generally something else again. The importance of Brentano’s insight that all psychological occurrences were intentional, that presentations, judgments and emotions were the key to understanding the natural history of consciousness and the science of the activities of mind, made it important for Twardowski to disengage the intentionality thesis as

THE CONTENT-OBJECT DISTINCTION

13

such from its unfortunate connection with the immanent intentionality thesis in Brentano’s original exposition of empirical psychology. As necessary as Twardowski’s distinction between content and object may have been to the progress of scientific psychology and phenomenology, it remains part of Brentano’s indelible legacy to have honestly set forth the implications of accepting the principles of a strictly empiricist philosophical methodology. The problem remains for the philosophy of science generally, and for the philosophy of psychology in particular when it aspires to be both faithfully empiricist and in harmony with the ontic expectations of the natural sciences as they are actually practiced. As reflected in Brentano’s choices in Psychologie vom empirischen Standpunkt, these opposing implications of Aristotelian empirical science, common sense, and a radical Humean and Berkeleyan philosophical empiricism could only play themselves out in later critical interaction on the part of thinkers like Höfler, Meinong, Kerry, and Twardowski. Each of these philosophers had both a strong empathy for Brentano’s empiricism and sufficient distance from the direct authorship of his theory to be able to see its limitations. It was the capacity to understand what was essential to Brentano’s scientific psychology and where its extreme philosophical empiricism had led the theory astray that induced the followers of Brentano in the decades after the publication of his masterwork to draw a sharp distinction where Brentano had conflated psychological content and intended object, in the greater interests of a new less strictly empiricist psychology that could take its rightful place among the other natural sciences.

4. Twardowski’s Perception of his Task and Purpose Whether Twardowski succeeds in demonstrating a distinction between thought content and intended object within a sufficiently robust philosophical empiricism is perhaps another question. It all rather depends on what we mean by and what we expect from a ‘sufficiently robust philosophical empiricism’, and on what, for that matter, we mean by and expect from the concept of empiricism. There are many possibilities for specific formulations of empiricism in the marketplace of ideas, as there have been since ancient times. Brentano

14

DALE JACQUETTE

throws in his lot with an extreme form that, like Hume, discourages him from thinking of reference and knowledge as extending beyond the contents of thought – in Hume’s terminology, the mind’s ideas and impressions – to the ‘external’ or ‘outside’ world.14 For Hume, this is the point where philosophical justification gives out and there is only commonsense psychological compulsion to explain the belief we have that something beyond thought may positively correspond with the contents of immediate sense experience.15 Since Brentano’s mission is to undertake the empiricist investigation of psychology itself, he does not quite have the same range of options as Hume. Yet he is faced with the same division between what can and cannot be empirically known. We are left to conjecture whether Twardowski in superimposing a distinction between content and object, however intuitive and attractive in other ways it may be, has lost sight of the deep commitment to a methodological empiricism that in many ways first motivated Brentano’s efforts to articulate the principles of a scientific psychology. We may then want to question whether Twardowski weakens the empiricist strand in Brentano’s original system to the point where it no longer supports the standards of epistemic justification, the cautious agnosticism concerning the existence of things that arguably are not directly experienced but only inferred from the contents of experience. The larger issue is not whether Twardowski’s distinction between content and object is true to Brentano’s vision of a scientific psychology, but whether philosophy can ever be adequately validated in its acceptance of the existence of objects outside the mind, and, in particular, outside what Brentano and Twardowski agree in referring to as the contents of presentations. The only way to decide the question is to look in detail at the arguments Twardowski gives for distinguishing between the immanent content and thought-transcending intended objects of thoughts. We may see no reasonable alternative but to admit that the objects of most states of mind exist outside of thought, as when we think of the Eiffel Tower or the square root of 2. No matter how forceful we may find these considerations, if we are classical empiricists in the spirit of Hume, as Brentano certainly seems to be, then we will regard such considerations as no more than the psychological compulsion which Hume acknowledges is with us whenever we inquire into the existence of an external world that seems to be revealed

THE CONTENT-OBJECT DISTINCTION

15

in sensation, but which he also and with equal conviction asserts drives us to accept beliefs for which we lack any adequate philosophical justification. If science, including psychological science, is content to proceed without its epistemic credentials in good order, that in one sense is entirely its own business. We cannot expect philosophical empiricism of the most rigorous sort to which Hume and Brentano subscribe simply to fall in line with whatever practice prevails if there is not also a sound circumspect rationale for a science’s metaphysical commitments. Thus, we must ask whether Twardowski provides good enough reasons for breaking down the barrier between mental content and intended objects that purportedly exist outside and independently of the mind. We must furthermore ask whether the arguments Twardowski gives are in keeping with roughly the kind of radical Humean philosophical empiricism that Brentano had hoped to weld together with the actual practice and commonsense metaphysics of the natural sciences, or whether Twardowski violates the requirements of the philosophical empiricist framework that Brentano seems to have wanted to preserve. If, finally, it turns out that Twardowski has bent or snapped the bounds of philosophical empiricism in the sense Brentano may have meant to endorse for the new science of psychology, then we may at last be in a position to ask whether Twardowski was right to do so. We shall then be asking in effect whether the specific kind of philosophical empiricism that Brentano seems to have inherited from Hume is supportable as a basis for natural science, or whether Twardowski inadvertently demonstrates that the empiricist tradition from Sextus to Hume, Berkeley, Thomas Reid, and others, finally breaks on the shoals of Brentano’s empirical-scientific psychology. Turning to Twardowski’s four arguments, we first summarize each and then consider the categories to which they belong as compatible or incompatible with a radical philosophical empiricism, before assessing their strengths. It is worth remarking that Twardowski begins his discussion of the distinction between content and object with a straightforward appeal to commonsense considerations concerning thoughts about the external existence of things, or, to use Berkeley’s expression, ‘without the mind’.16 The appeal to pretheoretical ‘naïve’ realist commitments to the existence of intended objects outside of thought is a natural one to offer to an audience

16

DALE JACQUETTE

that need not be assumed to have any prior commitment to Brentano’s Humean brand of philosophical empiricism. It is nevertheless revealing when Twardowski opens his treatment of the topic with these anti-Humean sentiments: That the content and object of a presentation are different from each other will hardly be denied when the object exists. If one says, ‘The sun exists,’ one obviously does not mean the content of one’s presentation of the sun, but rather something which is totally different from this content.17

Twardowski contrasts this case with the intentionally very different situation in which the intended object of thought is ostensibly something that does not exist. What he has in mind here is evidently the prelude to a Gegenstandstheorie or object theory of the sort that Meinong and Ernst Mally among others were later to develop.18 The very fact that Twardowski is willing to countenance mental acts directed toward nonexistent intended objects indicates that he is in sympathy with Brentano’s proto-phenomenological intensionalist and intentionalist standpoint in logic, semantics, and philosophy of mind. He lays the groundwork for such considerations in previous chapters of the text. He is notably not limiting the scope of his arguments to the kind of extensionalist presuppositions associated with Gottlob Frege’s philosophy of language, or with Bertrand Russell’s Fregean extensionalism, manifested after 1905 with the publication of his stridently anti-Meinongian essay ‘On Denoting’ (Russell 1905).19 Twardowski continues: The case is not so simple for presentations whose objects do not exist. It is tempting to believe that in this case there is no real difference between content and object, but only a logical one; that in this case content and object are really one; and that this one entity appears sometimes as content, sometimes as object, because of the two points of view from which one can look at it.20

Previously, on the preceding page, Twardowski had anticipated Meinong’s categories of incomplete and impossible nonexistent intended objects (unvollständige, unmögliche Gegenstände), both types of which are treated phenomenologically as on a par with existent entities insofar as they can also properly stand as the intended objects of thought:

THE CONTENT-OBJECT DISTINCTION

17

Hence nothing stands in the way of asserting that to every presentation there corresponds an object, whether the object exists or not… But there are many presentations whose objects do not exist, either because the objects combine contradictory determinations [Bestimmungen] and hence cannot exist, or because they simply do in fact not exist. Yet in all such cases the object is presented, so that one may speak of presentations whose objects do not exist, but not of presentations which are objectless, of presentations to which no object corresponds. 21

What is significant in the first two paragraphs of Twardowski’s Chapter 6, on ‘The Difference Between Content and Object’, is that the commonsense distinction between content and object and the non-immanence of existent entities like the sun plays no role in any of the four arguments by which Twardowski proposes to distinguish between the content and intended object of presentations. He introduces his topic to the reader by appealing to ordinary beliefs about the existence of things outside the mind, observing that in presentations involving existent objects the distinction between content and object is clearcut, whereas for presentations that ostensibly intend nonexistents more penetrating arguments are required. He states: On the contrary, a brief consideration shows that the differences between content and object of a presentation which can be ascertained when the object exists also are present when the object does not exist. We shall list the most important of these differences and try to show for each one how it occurs for existing as well as nonexistent objects.22

Twardowski thus aims at a fully general distinction between the content and object of every presentation. His target is not merely of presentations that intend existent objects, but of presentations generally, regardless of whether or not they happen to intend existent entities. In this, Twardowski may have tried to go too far, since there seem to be presentations that intend their own contents, as when in a self-referential presentation we think about the contents of the presentation itself, in which the contents and objects of thought are identical, and hence phenomenologically indistinguishable. Self-referential and self-applicational thoughts in a number of categories can be invoked to raise doubts about the full generality of Twardowski’s effort to distinguish between the content and object of any and all presentations.23

18

DALE JACQUETTE

5. Four Arguments for the Content-Object Distinction Twardowski’s analysis of psychological presentations into act, content, and object, falls squarely in the ambit of Brentano’s scientific descriptive psychology, in what Brentano later referred to as Psychognosie (Psychognosy) or phenomenology.24 Twardowski’s reasons for wanting to extend the distinction between content and object from existent to nonexistent intended objects are easy to understand, especially in retrospect with the needs of object theory in mind. If there is no distinction between content and object in the case of ostensibly nonexistent objects in mind, then there is no basis for advancing a theory of intended objects that is indifferent to their ontic status and that respects the independence of the determinations or so-being (Sosein) of an intended object from its being (Sein), as Meinong was later to maintain.25 These anticipations of Meinong’s object theory are most definitely out of keeping with Brentano’s metaphysics and theory of meaning, since he never countenanced nonexistents as the intended objects of presentations, judgments, or emotions. As to whether Twardowski’s distinction between content and object is consistent with Brentano’s radically Humean philosophical empiricism now remains to be seen. Twardowski’s arguments in summary are these. He concludes that presentational content and intended object must always be distinct because: (1) Otherwise it would not be possible for the same presentation content to present the object of a true as of a false judgment concerning the object’s existence. Since it is possible for the object of a presentation not to exist, but the content of the presentation exists whether or not it presents an existent object, content and intended object cannot be the same; and since this is true of any judgment concerning the existence of any contingently existent or nonexistent object, the distinction holds generally for the content and object of any presentation. (2) The properties of presentational content and intended object are different. For example, an impossible object cannot exist, but the content of the presentation of an impossible object evidently exists. The same is also true with respect to more ordinary or constitutive properties. Thus, in the case of a presentation of a golden mountain, the property of being golden and a mountain holds true of the intended object even if no golden mountain exists. However, the property does not hold true of the presentation’s

THE CONTENT-OBJECT DISTINCTION

19

content, which, again unlike the intended object, is something psychological, not made of gold and not a mountain [attributed to Benno Kerry]. (3) Thoughts can be directed toward the same intended object by means of different psychological contents, so that the object cannot possibly be identical to several distinct contents. Since there is no basis for concluding that the object is identical to one content rather than the other, it follows that the object must be distinct from any of the contents by which it is presented to thought. (4) The contents of general thoughts are unitary things but can sometimes present a plurality of objects. This is true, for example, of such presentation contents as those related to the judgment that ‘All cats are mammals’. The content is one unitary entity, but it seems to intend all cats, a multiplicity of things, which consequently cannot be identical to the presentation’s content [also attributed to Kerry, and with reservations as to its legitimacy].26 Twardowski does not accept argument (4), but considers it rather for the sake of completeness. He denies without argument that a plurality of objects falls under a unitary presentation content. He may have in mind an alternative account of the meaning of general terms and general contents as referring to or intending a totality or aggregate of things possessing the requisite combination of properties, rather than the individuals belonging to such collectivities.27 The idea is undeveloped in Twardowski, but it is easy to see that if such a view were to be preferred it would obviate Kerry’s argument for the distinction between content and object. Twardowski concludes that the previous arguments (1)-(3) are sufficient to uphold the distinction between content and object, ‘even without this argument, the reasons listed above seem to show sufficiently that one has to distinguish between content and object of a presentation’.28 As a general comment on the orientation of Twardowski’s four arguments, we may consider their relation to Brentano’s immanent intentionality intentional in-existence thesis. We ask whether each of the arguments would have any persuasive effect on someone like Brentano. Could Twardowski’s reasoning convince Brentano that even though intended objects are in some sense contained immanently within the thoughts by which they are intended there nevertheless remains a distinction between the contents and intended objects of thought? The question in other words is whether or

20

DALE JACQUETTE

not Twardowski’s four (or first three) arguments considered only in and of themselves are sufficient to prove that intended objects are mind-independent. There might then be a sharp distinction between thought contents and intended objects even if both contents and objects possess at most what Brentano refers to as immanently intentional in-existence. The run-down of Twardowski’s arguments indicates that they can be readily interpreted as internal considerations offered from within the standpoint of Brentano’s immanent intentionality thesis, in keeping with Twardowski’s own description of his inquiry as a psychological investigation. Argument (1) is supposed to distinguish between the content and intended object of a presentation regardless of whether or not the object exists. Let us suppose that an intended object does not exist, but that the thought content by means of which the object is intended exists as a feature of a real thought. Thus, content and object cannot be identical, even without supposing that intended objects are necessarily mind-independent. The trouble with this first argument of Twardowski’s is that it is open to the complaint that the intended object exists in precisely the same sense as the content, and that according to Brentano’s descriptive psychology it then has what Brentano expressly refers to as intentional in-existence. This is the same sense in which the content of thought exists, so that the argument does not go far if any distance at all toward proving that the content and object of a presentation must be distinct. There is nevertheless an essential difference, when, for example, we say that the intended object of a thought ostensibly about a golden mountain does not exist, that there is no golden mountain, but that the content of the thought exists. We mean that in real space and time there is no golden mountain, even if space and time should themselves turn out to be only something mental, but that the content of the thought exists at least in time. If this is the meaning and import of Twardowski’s first argument, then as an internal argument aimed at proponents of Brentano’s intentional in-existence thesis, it can at best demonstrate that some thought contents are distinct from some intended objects. The argument may nevertheless serve a vital function in opening up the theory of intentionality to the possibility of a more generalized distinction between content and object as the thin edge of a maximally dispersive wedge. Argument (2) presses the distinction somewhat further by recognizing distinctions among the properties of contents and intended objects. This is

THE CONTENT-OBJECT DISTINCTION

21

clear already in the first argument if we consider the property of existing in real time as a property of the thought content but not of the intended object in the case of a thought ostensibly about the golden mountain. Twardowski wants to extend the reasoning to include other kinds of properties as well. Limiting attention for the moment to the golden mountain as an intended object, the possession of distinguishing properties apparently goes in the opposite direction, where it seems intuitively more correct to say that the golden mountain is in truth both golden and a mountain even though the golden mountain does not exist. The content of a thought about the golden mountain as something existent, a psychological occurrence in real time, in contrast, is itself neither golden nor a mountain. If the distinction among the properties of the contents and objects of thoughts can be fully generalized, then Twardowski has directly in hand the basis for a distinction between the content and object of every presentation. We have already suggested that this might not be the case where content and object coincide, as in thoughts that intend their own contents, including some of the thoughts we must inevitably entertain in considering Twardowski’s four arguments. We can safely conclude that if Brentano would not recognize a distinction between thought contents and intended objects on the grounds of Twardowski’s second argument, then it is for a purely internal reason to which he and his followers who accept the doctrine of immanently intentional inexistence ought to have been more sensitive. It is the argument, namely, that a close phenomenological scrutiny by means of inner perception reveals that contents as a rule are different from intended objects because they do not typically share all of their properties in common. The argument works to the extent that it does as the kind of reasoning that ought to carry weight without simply assuming that intended objects exist outside of thought as mind-independent or mind-transcending, and therefore once again as a friendly internal refinement rather than external refutation of Brentano’s thesis of intentionality as the distinctive mark of the psychological. Argument (3) is in some ways the most interesting of Twardowski’s proofs for the distinction between the contents and objects of thoughts. It also clearly involves an internal phenomenological kind of distinction that does not bluntly presuppose that intended objects transcend thought while thought contents are immanently contained within them. We can judge

22

DALE JACQUETTE

merely by reflecting on the inner structures of various thoughts that the same object can be intended by different thoughts with different contents. Twardowski considers the example of thinking about the same city alternatively as the birthplace of Mozart and the site of the Roman Juvavum (Salzburg, Austria). These are evidently altogether different thought contents, and it could come as a kind of revelation to discover that thoughts with these contents as a matter of fact intend the same object. The situation would then be much the same as Frege says in his essay ‘Über Sinn und Bedeutung’ with respect to thoughts about, or the coincident denotation of the proper names, ‘the evening star’ and ‘the morning star’ (Venus).29 The generality for the distinction between content and object that Twardowski seeks is afforded in this argument by the fact that in principle it is always possible to have different thoughts with distinct contents directed upon the very same intended object. Thus, it does not matter if it should happen to be true that in the psychological life even of a great number of given thinkers there is an exact one-one correlation of specific thought contents and specific intended objects, since it is clear from a single case that it is always at least logically possible for there to be a many-one relation between contents and objects. Since, however, the contents in question are distinct, they must also be distinct from their correlated intended objects, so that contents and objects are once again distinguished. If we take note of the general exception for thoughts that are about their own contents, it appears that in normal cases Twardowski has an internal phenomenological reason in his third argument for distinguishing between the contents and objects of thoughts. The argument does not simply assume that intended objects are mind-independent or mind-transcending, or that they have some ontic status that places them beyond the immanent contents of presentations. It is instead the kind of justification for the distinction that ought to hold sway with Brentanians who accept some form of the doctrine of the immanent in-existence of intended objects to convince them that even so there is a distinction between the content of thought and its intended object. Turning, finally, to argument (4), we find Twardowski offering a reason for the distinction that is equally internal or phenomenological. Introspectively, it is easy to understand why someone like Kerry would want to hold that a single thought content can be correlated with multiple intended objects, without supposing that intended objects exist outside the mind.

THE CONTENT-OBJECT DISTINCTION

23

Twardowski seems to limit the argument to generalizations of the type found in quantificational logic, as when we think of ‘All cats’, thereby putatively intending a plurality of distinct individual animals. The same consideration ought to apply in any case to such evidently unitary thought contents as ‘the Karamazov brothers’ that seem intuitively to be related to a plurality of intended objects, to Dimitri and his siblings Ivan and Alexei (Alyosha); or, say to ‘the Allies’, intending thereby Great Britain, Russia, Canada, Australia, China, and the United States. Twardowski appears to include Kerry’s argument only grudgingly for the sake of completeness, while not accepting the inference that a single general content can be correlated in thought with many different intended objects. He does not offer a reason for doubting Kerry’s conclusion, however, which in some ways is more in line with Brentano’s later reism, although it is possible that he imagines in such cases that the intended object is really also a unitary set, totality or collectivity of things rather than the individuals belonging to its membership considered individually. Even if the argument were accepted, it does not afford a fully general distinction between the contents and objects of all presentations, but at most only of those general thoughts that happen to involve unitary contents directed toward a multiplicity of intended objects.30

6. Content-Object and the Mind-Independence of Intended Objects The moral of this review of Twardowski’s four arguments for the contentobject distinction is, first, that none of the justifications he presents are external challenges that merely assume what the arguments are meant to prove, that the contents and objects of every psychological presentation are distinct. They are friendly amendments to Brentano’s intentionality thesis, or should be considered as such, as far as Brentanians committed to the immanently intentional in-existence of intended objects are concerned. Second, however, and perhaps equally important, is the fact that for this very reason, Twardowski’s arguments, contrary to what is often said about their implications for an object theory or ‘transcendental’ phenomenology (not in the later Husserlian sense), at their best do not actually establish the mind-independence of intended objects, but only their distinction from

24

DALE JACQUETTE

thought contents. As far as any of Twardowski’s four arguments go, it could still be the case that both contents and intended objects, admittedly as distinct from one another in all but exceptional self-referential or selfapplicational cases, are alike immanently in-existent, existing only within or contained as distinct parts of particular psychological presentations. This is a remarkable limitation of which Twardowski himself seems completely unaware. For, in the first chapter of his work, ‘Act, Content, and Object of the Presentation’, he declares: It is one of the best known positions of psychology, hardly contested by anyone, that every mental phenomenon intends an immanent object. The existence of such a relation is a characteristic feature of mental phenomena which are by means of it distinguished from the physical phenomena…But if a confusion between a mental act and its content is thus prevented, an ambiguity – pointed out by Höfler – still remains to be overcome. After having discussed the characteristic relation of mental phenomena to a content, he continues: “(1). What we called ‘content of the presentation and the judgment’ lies just as much completely within the subject as the act of presentation and of judgment itself. (2) The words ‘thing’ and ‘object’ are used in two senses: on the one for that independently existing entity…at which our presentation and judgment aim, ‘picture’ of that real entity which exists ‘in’ us. This quasi-picture (more accurate: sign) is identical with the content mentioned under (1). In distinction to the thing or object, which is assumed to be independent of thinking, one also calls the content of a presentation and judgment (similarly: of a feeling and willing) the ‘immanent or intentional object’ of these mental phenomena [Logik, paragraph 6].”31

The tribute to Brentano in the first sentences of the paragraph is unmistakable. Twardowski does not mention Brentano by name in the body of the text, but only in the accompanying footnote, perhaps out of respect for a teacher whose central contribution to descriptive philosophical psychology he is soon to criticize. The repeatedly emphasized contrast between immanent content of thought on the one hand, and the thought’s intended object on the other, makes it clear in Höfler as in Twardowski that the goal of a demonstrated distinction between content and object is to set off intended objects as other than immanent, indicating that they are to be understood as non-immanent or thought- or mind-transcending. This is clear also when Twardowski continues, after the completion of his quotation from Höfler above: ‘One has to distinguish, accordingly, between the object at which

THE CONTENT-OBJECT DISTINCTION

25

our idea “aims, as it were,” and the immanent object or the content of the presentation’.32 To this inference it hardly needs to be added that the only sense in which an intended object unlike a thought content can fail to be immanently existent within a presentation is for the object to transcend the presentation, to exist beyond, without, or outside the mind. It is the mind-independent existence of intended objects, even or especially when the objects do not exist, that Twardowski hopes at least the combined effect of his four arguments will establish. If he can prove that the objects of thoughts are not a part of the thoughts immanently contained within them, then he will have restored a measure of commonsense realism to Brentano’s otherwise valuable proposition that the mental or psychological is distinguished from all physical phenomena by virtue of the intentionality of presentations. If Twardowski’s arguments are successful in the way that he appears to require, then the ‘naïve’ metaphysics of Aristotle can also be restored to Brentano’s descriptive psychology, along with the commonsense realism whereby we ordinarily assume that the things we think about do not merely exist within our minds as an immanent constituent of our thoughts, but outside of thought in what we are pleased to speak of as the external world. Importantly, then, if Twardowski’s arguments achieve their intended effect, they equally restore the expectations of natural science whose methods Brentano wants to assimilate as the only proper methods of philosophy. The problem now is that Twardowski’s four arguments do not in fact guarantee this further conclusion. What he shows is merely that the immanent content of a thought is distinct from its intended object, allowing as always for exceptions involving thoughts about their own contents. What he does not show, but seems to believe he has shown, is that the contents and objects of thoughts are distinct by virtue of their contents being mindimmanent and their intended objects being mind-transcending. The difference is crucial. Given the friendly phenomenological orientation of all four of Twardowski’s arguments in Chapter 6 in relation to Brentano’s immanent intentionality thesis, there is no prospect for Twardowski to have possibly accomplished anything more than this. The only way in which Twardowski could have properly established the mind-independence or thoughttranscending of intended objects is to have single-handedly solved the

26

DALE JACQUETTE

great riddle of phenomenalism with which most of the history of metaphysics and epistemology in the Western tradition has struggled. We can begin with Descartes in the Meditations on First Philosophy, trying valiantly to prove that the external world corresponds positively with his sensory representations, but whose best solution is only the flimsiest effort to demonstrate the existence of a veracious God who would not permit Descartes to be deceived when he entertains clear and distinct perceptions. Then we might turn to Hume, who resolves the problem in A Treatise of Human Nature only by admitting defeat, acknowledging that in the end he can offer no adequate philosophical justification for the existence of an external world in which he nonetheless finds himself psychologically compelled to believe. Berkeley before him accepted without qualification the world of physical things as existing only in the mind, by which he made the foundation of his radical idealism an implication of his radical empiricism in A Treatise Concerning the Principles of Human Knowledge and Three Dialogues Between Hylas and Philonous. It is also the problem with which Brentano in Psychologie vom empirischen Standpunkt makes an uneasy Humean sort of compromise at the expense of Aristotelian natural scientific realism by remaining ontically agnostic about the existence of intended objects outside the mind, committing himself only to their immanent intentional in-existence within thought, as existing in the mind. Had Twardowski managed to prove the existence of any intended objects without the mind, had he solved the great riddle of phenomenalism, he would have justly earned a more famous place in the history of thought than Descartes, Berkeley, Hume, Brentano, or virtually any metaphysician or epistemologist before or after these giants of philosophy. He does no such thing, however, nor is there any chapter or verse anywhere in his valuable treatise in which he purports to accomplish anything quite so momentous. What, then, does he try to do, and how does he think he can manage to reconcile the limited objectives of the four arguments for the content-object distinction with his realist avowal of the mind-independent or thought-transcending ontic status of intended objects? How are we to understand exactly what Twardowski achieves in Zur Lehre vom Inhalt und Gegenstand der Vorstellungen, and what he may have considered to

THE CONTENT-OBJECT DISTINCTION

27

have been his purpose and the extent of his success in going beyond Brentano by distinguishing between the content and object of presentations? There is room for conjecture about certain aspects of Twardowski’s project, and of how he may have conceived of his project. Whatever disputed issues may continue to surround Twardowski’s analysis of the components of psychological presentations into the triad of act, content, and object, there seems little dispute about the fact that Twardowski understood his investigations as a contribution to a broadly Brentanian descriptive psychology, and that he also wanted to be able to uphold the robust Aristotelian scientific realism that locates ordinary intended objects of thought in the external world outside the mind. He wanted, in other words, to be some kind of empiricist, but not as strict or ontically austere an empiricist as Brentano or Hume, if he could only do so at the relaitvely high cost of commonsense realism and the Aristotelian naïve metaphysics that goes along with the natural sciences – for, above all, he wants to make philosophical psychology into a natural science within a general Brentanian intentionalist framework. Accordingly, the most reasonable way to interpret Twardowski’s main goal in the text appears to be that of breaking the hold of Brentano’s immanentism with respect to the objects of thought in order to make it possible even if not obligatory to adopt a commonsense metaphysics of mindtranscending objects. The ordinary objects intended by perceptual presentations, the things we see and hear and touch and taste and smell, along with the intended objects of many other kinds of psychological occurrences, including judgment, memory, and what Meinong would later distinguish as a separate category of assumption, are for Twardowski as they are for common sense and the metaphysics of natural science, real existent things in an external world that transcends the mind beyond the contents of thought. As Brentano leaves things in the Psychologie, there is no possibility of treating intended objects as anything other than mind-immanent or intentionally in-existent, occurring only within thought. It is only when Twardowski (and Höfler, Meinong, and Kerry among others before him) distinguish between the content of thought and its intended object on descriptive psychological or phenomenological grounds that an opportunity is created for regarding the immanent component of thought as its internal content, and for positing intended objects of thought in at least some if not

28

DALE JACQUETTE

most standard cases as independent of thought, transcending, which in this non-Kantian context is just to say existing, outside the mind. Twardowski does not prove that the intended objects of presentations are not also immanent like their contents. Once he has proved that intended objects are distinct from contents, however, he makes it possible for common sense to declare their independence from and transcendence of mental events. The result Twardowski attains represents an uneasy compromise in rejecting the radical Humean empiricism with which Brentano himself seems most comfortable. Twardowski in effect qualifies Brentano’s empiricism, diluting it in order to provide for the existence of an external world of intended objects in agreement with common sense and scientific realism. Where empiricism in its pure form stands in conflict with the belief in the external existence of the objects of perceptual and other kinds of presentations, Twardowski does not hesitate to abandon the methodological limitations of experience for what he takes to be the greater good of the most compelling realist externalist metaphysics of empirical science. This, of course, is nothing other than what Hume maintained all along, and as Brentano may have also implicitly understood, that while we lack philosophical justification for belief in an external reality, we are irresistibly driven to believe that a world beyond and outside the contents of thought exists. The judgment of history has been that Twardowski most probably made the right decision in adjudicating between radical empiricism and intuitive realist metaphysics. As reflected at least in the movement away from Brentano’s original immanent intentional in-existence thesis for intended objects that has characterized object theory, phenomenology, and intentionalist philosophy of mind and semantics by the vast majority of Brentano’s students even during his lifetime, the relation of thought to intended objects outside of mental states has proven itself to be a more sustainable formulation of Brentano’s intentionality thesis. The tradeoff has been to sacrifice a pure empiricism that eschews ontological commitments that cannot be rigorously upheld within the limits of experiential evidence. Twardowski’s approach has been widely preferred in almost all the later fruitful intentionalisms that have since been developed around an underlying recognition of the content-object distinction. These have virtually unanimously rejected Brentano’s immanent intentionality thesis. Whether

THE CONTENT-OBJECT DISTINCTION

29

or not Twardowski in a deeper sense was right to turn away from Brentano’s radical Humean empiricism in order to accommodate commonsense beliefs about the external extra-phenomenological existence of intended objects of thought may have yet to be decided, and certainly Twardowski says nothing to definitively settle this longstanding controversy.33

Notes 1

See Roman Ingarden, ‘The Scientific Activity of Kazimierz Twardowski’, Studia Philosophica 1939–1946, 3, 1948, pp. 11–17. 2 Franz Brentano, Psychology from an Empirical Standpoint, translated by Linda L. McAlister (London: Routledge & Kegan Paul Ltd., 1973), pp. 88–89. 3 I discuss these issues in Dale Jacquette, ‘The Origins of Gegenstandstheorie: Immanent and Transcendent Intentional Objects in Brentano, Twardowski, and Meinong’, Brentano Studien, 3, 1990–1991, 277–302. 4 The importance of Twardowski in the Brentano school and in the development of Polish philosophy is detailed by Roberto Poli, ‘Kazimierz Twardowski 1866–1938’ in The School of Franz Brentano, edited by Liliana Albertazzi, Massimo Libardi, and Roberto Poli (Dordrecht, Boston, London: Kluwer Academic Publishers, 1996), pp. 207–231. 5 See, for example, Reinhardt Grossmann, Meinong (London: Routledge & Kegan Paul, 1974), pp. 48–56. Grossmann writes, p. 48: ‘In 1894, there appeared a rather slim book by K. Twardowski, a student of Brentano’s, which influenced the course of philosophy. In this book, Twardowski argued that the intention of a mental act – its ‘object’ – is in no sense of the term ‘immanent’ to the act. He argued, in other words, that the intention of a mental act is never a part of that act. Twardowski, therefore, distinguished between the content of an act of presentation – what I shall sometimes call an ‘idea’ – and the object of this act. Without this distinction, I am convinced, there would be neither phenomenology nor a theory of entities [sic.: objects]. Meinong, as we shall presently see, adopts Twardowski’s distinction. He even presents the same arguments for it as Twardowski.’ Grossmann is wrong on several counts. Twardowski adopts the act-content-object distinction from Höfler and Meinong, not the other way around. Meinong does not accept Twardowski’s third and fourth arguments, as we read in Meinong, ‘Über Gegenstände höherer Ordnung und deren Verhältnis zur inneren Wahrnehmung’, Zeitschrift für Psychologie und Physiologie der Sinnesorgane, 21, 1899, pp. 181–272. More importantly, if we have interpreted his Chapter 6 correctly, then Twardowski does not argue that intended objects are nonimmanent, but only draws the content-object distinction in order to make it possible to agree with

30

DALE JACQUETTE

common sense that intended objects are generally not constituents of the thoughts by which they are intended. 6 Kasimir [Kazimierz] Twardowski, On the Content and Object of Presentations: A Psychological Investigation. Translation of Zur Lehre vom Inhalt und Gegenstand der Vorstellungen [Vienna: Hölder Verlag, 1894] by Grossmann (The Hague: Martinus Nijhoff, 1977), pp. 1–2; 15, 17–19, 26–31. All references to this translation unless otherwise indicated. Twardowski refers to Kerry, ‘Über Anschauung und ihre psychische Verarbeitung’, Vierteljahrschrift für wissenschaftliche Philosophie, 10, 1886, pp. 419–467. See Alois Höfler, in collaboration with Alexius Meinong, Logik (Vienna: Tempsky Verlag, 1890), pp. 6–7. 7 Barry Smith, ‘Kasimir Twardowski: On Content and Object’, in Austrian Philosophy: The Legacy of Franz Brentano (Chicago and LaSalle: Open Court Publishing Company, 1994), section 6, ‘From Psychology to Logic’, pp. 185–191. The precursor is Smith, ‘Kasimir Tardowski: An Essay on the Borderlines of Ontology, Psychology and Logic’, in Klemens Szaniawski, editor, The Vienna Circle and the Lvov-Warsaw School (Dordrecht: Kluwer Academic Publishing, 1989), pp. 313–375. 8 Twardowski, On the Content and Object of Presentations, Chapter 6, pp. 27–31. 9 Brentano, ‘Forward’ to the first edition of Psychology from an Empirical Standpoint, p. xxvii: ‘My psychological standpoint is empirical; experience is my teacher. Yet I share with other thinkers the conviction that this is entirely compatible with a certain ideal point of view.’ Brentano’s famous fourth thesis from his Habilitationsschrift defense at the Bayerische-Julius-Maximilians-Universität at Würzburg states: ‘Vera philosophiae methodus nulla alia nisi scientiae naturalis est’ (The true method of philosophy is none other than that of the natural sciences) (Aschaffenburg: J.W. Schipner, 1866). 10 See Jacquette, ‘Fin de Siècle Austrian Thought and the Rise of Scientific Philosophy’, History of European Ideas, 27, 2001, pp. 307–315. Jacquette, ‘Brentano’s Scientific Revolution in Philosophy’, Spindel Conference 2001, Origins: The Common Sources of Analytic and Phenomenological Traditions, Southern Journal of Philosophy, Spindel Conference Supplement, 40, 2002, pp. 193–221. 11 The point is too obvious to require documentation. Twardowski’s opening paragraph in the book is ample indication of his indebtedness to Brentano’s scientific approach to psychology, and is further supported by all his subsequent work in philosophy. See Liliana Albertazzi, ‘Brentano, Twardowski and Polish Scientific Philosophy’, in Polish Scientific Philosophy: The Lvov-Warsaw School, edited by Francesco Coniglione, Roberto Poli, and Jan Woleński, Amsterdam: Rodopi, 1993 (Poznań Studies in the Philosophy of Sciences and Humanities 28), pp. 11–40. 12 See Grossmann, ‘Introduction’ to his translation of Twardowski, On the Content and Object of Presentations, pp. xxviii–xxxiv. 13 I offer this interpretation of Brentano’s commitment to strict philosophical empiricism as a way of understanding his early immanentism in Jacquette, ‘Brentano’s Con-

THE CONTENT-OBJECT DISTINCTION

31

cept of Intentionality’ in Jacquette, editor, The Cambridge Companion to Brentano (Cambridge: Cambridge University Press, 2004), pp. 98–130; especially pp. 121–124. 14 David Hume, A Treatise of Human Nature [1739/40], edited by L.A. Selby-Bigge, second edition revised by P.H. Nidditch (Oxford: The Clarendon Press, 1978), Book I, Part I, Section I, ‘Of the Origin of our Ideas’, pp. 1–7. 15 Hume, A Treatise of Human Nature, Book I, Part II, Section VI, ‘Of the idea of existence, and of external existence’, p. 67: ‘[N]o object can be presented resembling some object with respect to its existence, and different from others in the same particular; since every object, that is presented, must necessarily be existent. A like reasoning will account for the idea of external existence. We may observe, that ’tis universally allow’d by philosophers, and is besides pretty obvious of itself, that nothing is ever really present with the mind but its perceptions or impressions and ideas, and that external objects become known to us only by those perceptions they occasion’. Part IV, Section II, ‘Of scepticism with regard to the senses’, p. 187: ‘We may well ask,’ he writes, ‘What causes induce us to believe in the existence of body? but ’tis in vain to ask, Whether there be body or not? That is a point, which we must take for granted in all our reasonings’. Part II, Section V, pp. 63–64: ‘’Twill probably be said, that my reasoning makes nothing to the matter in hand, and that I explain only the manner in which objects affect the senses, without endeavouring to account for their real nature and operations… I answer this objection, by pleading guilty, and by confessing that my intention never was to penetrate into the nature of bodies, or explain the secret causes of their operations. For besides that this belongs not to my present purpose, I am afraid, that such an enterprize is beyond the reach of human understanding, and that we can never pretend to know body otherwise than by those external properties, which discover themselves to the senses’. 16 George Berkeley, Three Dialogues Between Hylas and Philonous and A Treatise Concerning the Principles of Human Knowledge, The Works of George Berkeley Bishop of Cloyne, edited by A.A. Luce and T.E. Jessop (London: Thomas Nelson and Sons Ltd., 1949), Vol. II. 17 Twardowski, On the Content and Object of Presentations, p. 27. 18 Alexius Meinong, Alexius Meinong Gesamtausgabe, 8 volumes, edited by Rudolf Haller and Rudolf Kindinger in collaboration with Roderick M. Chisholm (Graz: Akademische Druck- u. Verlagsanstalt, 1969–1978). For a useful summary of the main principles of Meinong’s mature Gegenstandstheorie, see Meinong, ‘Über Gegenstandstheorie’, in Meinong, editor, Untersuchungen zur Gegenstandstheorie und Psychologie (Leipzig: Verlag von Johann Ambrosius Barth, 1904), pp. 481–530; Alexius Meinong Gesamtausgabe II, pp. 483–450. The best English introduction to Meinong’s thought in the secondary literature remains J.N. Findlay, Meinong’s Theory of Objects and Values, Edited with an Introduction by Dale Jacquette, from the second edition, Oxford University Press, 1963 (Aldershot: Ashgate Publishing, 1995) (Gregg Revivals).

32 19

DALE JACQUETTE

Bertrand Russell, ‘On Denoting’, Mind, 14, 1905, pp. 479–493. Twardowski, On the Content and Object of Presentations, p. 27. 21 Ibid., p. 26. 22 Ibid., p. 27. 23 See Jacquette, ‘Twardowski on Content and Object’, Conceptus: Zeitschrift für Philosophie, 21,1987, pp. 193–199. Jacquette, Meinongian Logic: The Semantics of Existence and Nonexistence (Berlin and New York: Walter de Gruyter, 1996), pp. 193–199. Edmund Husserl describes situations in which the contents and intended objects of thoughts can coincide in Logical Investigations, second edition [Logische Untersuchungen, 1913], translated by J. N. Findlay (New York: Humanities Press, 1970), volume I, pp. 287–291. 24 Brentano, Descriptive Psychology, translated and edited by Benito Müller from Brentano, Deskriptive Psychologie, edited by Roderick M. Chisholm and Wilhelm Baumgartner (Hamburg: Felix Meiner Verlag, 1982) (London and New York: Routledge, 1995), Appendix 5, ‘Psychognostic Sketch: Different Adaptation’, under item 6, Brentano writes, p. 167: ‘Psychognosy as an experiential science [Erfahrungswissenschaft]. There are sciences which, at least according to the sententia communis, are built up completely a priori. Psychognosy, in any case, is incapable of being so. It, too, must start with what is immediately evident. But [what, in its case, is immediately evident] are immediately evident facts which are not of apodeictic but of purely assertoric character. It is the sort of fact upon which every experiential science is based in its own way.’ 25 Meinong, ‘Über Gegenstandstheorie’, Alexius Meinong Gesamtausgabe, II, pp. 490– 493. The independence of Sosein from Sein thesis is formulated by Meinong’s student Ernst Mally, ‘Untersuchungen zur Gegenstandstheorie des Messens’, in Meinong, editor, Untersuchungen zur Gegenstandstheorie und Psychologie (Leipzig, Verlag von Johann Ambrosius Barth, 1904), p. 127. See also Findlay, Meinong’s Theory of Objects and Values, p. 44. On Meinong’s thesis of the independence of Sosein from Sein, see especially Karel Lambert, Meinong and the Principle of Independence: Its Place in Meinong’s Theory of Objects and its Significance in Contemporary Philosophical Logic (Cambridge: Cambridge University Press, 1983), pp. 87–96. Nicholas Griffin, ‘The Independence of Sosein from Sein’, Grazer Philosophische Studien, 9, 1979, pp. 23–34. 26 Twardowski, On the Content and Object of Presentations, pp. 27–31. 27 Ibid., Chapter 15, ‘The Objects of General Presentation’, pp. 97–105. See Smith, ‘Kasimir Twardowski: On Content and Object’, section 3, ‘The Theory of the General Object’, pp. 162–171. 28 Twardowski, On the Content and Object of Presentations, p. 31. 29 Gottlob Frege, ‘On Sense and Reference’, translation of ‘Über Sinn und Bedeutung’ by Max Black in Translations from the Philosophical Writings of Gottlob Frege (Oxford: Basil Blackwell, 1970), especially pp. 56–59. 20

THE CONTENT-OBJECT DISTINCTION 30

33

Twardowski is also likely to have been influenced by his teacher Robert Zimmermann, who was not directly part of the Brentano school. See Rudolf Haller, ‘Einleitung’ to the recent German republication of Twardowski, Zur Lehre vom Inhalt und Gegenstand der Vorstellungen: Eine psychologische Untersuchung (Munich and Vienna: Philosophia Verlag, 1982), pp. vii–xiii. 31 Twardowski, On the Content and Object of Presentations, pp. 1–2. 32 Ibid., p. 2. 33 This essay is dedicated to the memory of Karl Schuhmann.

On the Ambiguities of the Term Judgement. An Evaluation of Twardowski’s Distinction between Action and Product MARIA VAN DER SCHAAR

Introduction A theory of judgement has to make a distinction between the act of judging, the product of that act, and the content of the judgement (§1). Kazimierz Twardowski, in a nice, subtle paper dating from 1912, makes a distinction between actions and products, of which the distinction between mental acts and products is a special case. In this paper Twardowski identifies the judgement-product with the content of the judgement (§2). Twardowski’s distinction between action and product is evaluated, and it is argued that the judgement-product is not to be identified with the content of the judgement (§3). Besides the distinction between judgement as act and judgement-product, Twardowski makes a distinction between judgement as act and judgement as disposition (belief). The latter distinction has an application in the theory of knowledge, as it can be found in Twardowski’s lectures on the Theory of Knowledge from 1925 (§4).

1. Act, Content, and Product of Judgement and Assertion By the utterance of a declarative sentence one standardly makes an assertion. The speech act of assertion is to be distinguished from its product, the assertion made. We may follow Dummett (and Wittgenstein) in the explanation of an act of judgement as interiorisation of an act of assertion (cf. Dummett 1973, 363). More traditional accounts explain the speech act of assertion as intimation (Kundgebung) of the mental act of judgement (Frege 1918, 62). Just as the act of assertion results in an assertion made, the asserted assertion, the act of judgement results in the judgement made or passed. Whereas the act has a spatio-temporal existence, the product which arises in the act is an abstract entity. Mahler’s (complex) act of Actions, Products, and Things. Brentano and Polish Philosophy. Arkadiusz Chrudzimski and Dariusz Łukasiewicz (eds.), Frankfurt: ontos, 2006, 35–53.

36

MARIA VAN DER SCHAAR

composing das Lied von der Erde in 1909 resulted in the famous song cycle, which is an abstract entity created by Mahler in the act of composing. Similarly, all speech acts result in a product. The result of an interrogative act is the question asked. A promise made is the result of an act of promising. The question asked or the promise made can be determined by means of memory or written signs, traces, at any moment after the act was done. The verdict passed by a judge in court can be determined by the traces left in court, such as the notes made by a secretary. The relation between an act and its product is not a natural relation; it is not the case that the act is the cause of its product. The relation between act and product is internal; in the act of judging a judgement-product is necessarily constituted. When the action is done on pre-existing material, for example, when one writes a letter by the use of ink and paper, the resulting object, the written letter, is also internally related to the action of writing the letter. Relations of causality exist only between hand-movements and a certain distribution of ink on paper. Actions need not be done deliberately. Walking on the beach one leaves one’s foot-prints; they result from imprinting one’s feet on the sand. A foot-print is explained by its being the result of a foot being imprinted; it is not explained in terms of (sand-) configurations. The judgement-product is to be distinguished from the content of the judgement. Whereas the judgement made is there only as result of an act of judgement, this is not the case for the content of the judgement. Although the role of being the content of a particular act of judgement is dependent upon that act, the entity that fulfils this role is independent of this particular act of judgement, and is called the ‘proposition’ since Russell used that term as translation for Frege’s term ‘Gedanke’. Often, before making a judgement, we ask whether the relevant proposition is true. Before one passes the judgement ‘Snow is white’, one may ask ‘Is snow white?’ In order to explain that the judgement is an answer to the relevant question, we need a common element, the proposition that snow is white, which may function as content of both acts. Frege’s logical realism has it that the proposition is independent of any mental act. One need not be a logical realist, though, to acknowledge the distinction between judgement made and judgemental content. The content of the judgement, the proposition, and the judgement made can be distinguished by their linguistic counterparts. A judgement made is

ACTION AND PRODUCT

37

most aptly made known by a declarative sentence, and thus has indicative form. The proposition, on the other hand, is indicated by a that-clause. The judgement made, Snow is white, is thus to be distinguished from the judgemental content that snow is white. The distinction between judgemental act, judgement-product and content of the judgement is already made by Bolzano in his Wissenschaftslehre. There is a slight difference between the modern, Fregean account of the content of the judgement and that of Bolzano. For Frege, the judgemental content is denoted by a that-clause. For Bolzano, the Satz an sich, which functions as judgemental content, has the form A hat b, that is, it has indicative form. Notwithstanding this difference in form, the crucial point is also made by Bolzano: we may apprehend a Satz an sich without acknowledging its truth or falsity. The mere apprehension of a Satz an sich, according to Bolzano, is an act of presentation whose content is a Vorstellung an sich, whereas the Satz an sich is the object of that presentation (Bolzano 1837, § 124, § 127; cf. § 19 and § 34). For Bolzano, the proposition cannot be the content of an act of presentation. Bolzano explicitly asserts that a Satz an sich is not the result of an act of ‘setzen’. Similarly, a Satz an sich is not the judgement (product) that results in an act of judging (idem, § 20). Besides the act of judgement, the judgement made and the content of the judgement, a fourth notion can be discerned (cf. Sundholm 1999). One may speak of a judgement not yet passed, as one can speak of a question not yet asked. When the sentences ‘Snow is white.’ or ‘Is snow white?’ are used as examples, there is no judgement or question made. Someone who understands these sentences apprehends not merely the common proposition that snow is white, he also understands that the former example is standardly used for making an assertion, whereas the latter example is standardly used for asking a question. Someone who understands a declarative sentence apprehends what may be called the judgement-candidate or the assertion-candidate: it is the assertion stripped of its assertive force, a potential assertion. The declarative and the interrogative sentence just given express the same proposition, but they differ as far as the meaning of mood is concerned. The judgement or assertion-candidate

MARIA VAN DER SCHAAR

38

snow is white is equivalent to that snow is white, is true. The … is true-part of the assertion-candidate is indicated by the indicative mood of the declarative ‘Snow is white’. The interrogative mood of the sentence ‘Is snow white?’ indicates the part is it true…(?), which together with the proposition that snow is white makes up the question-candidate is it true that snow is white(?), or, shorter, is snow white(?). The question-mark is put in brackets because no question is actually asked. It is now easy to make up the other candidate notions, such as the ordercandidate let snow be white, or, more transparently: let it be the case that snow is white. Bolzano’s notion of a Satz an sich, with its indicative form, is perhaps closer to the notion of judgement-candidate than to the modern notion of proposition explained above. The term ‘judgement’ not only shows an ambiguity between act, content, product and candidate. Reinach, a pupil of Husserl, accuses Brentano of the fact that he does not make a distinction between judgement as act and judgement as belief. According to Reinach, judgment as belief is a conviction, a state of consciousness, and is distinguished from the act of judging in being enduring and involving degrees. The act of judgement or

ACTION AND PRODUCT

39

assertion is necessarily “accompanied by an underlying conviction which is such that both the assertion and the belief relate to something strictly identical.” (Reinach 1911, 320). Lying, for Reinach, is merely a quasi-assertion. Finally, the term ‘judgement’ may stand for the faculty of judgement. One need not presuppose that such a faculty has a substantial existence. It can be explained in terms of acts of judgements: the faculty of judgement is the general capacity to make such acts.

2. Twardowski on Actions and Products There is a grammatical difference between the two sentences (1) Hans is reading a letter. (2) Hans is writing a letter. Although ‘a letter’ is in both sentences the direct object (Akkusativ), there is a difference between the two objects (Grammarians use the term ‘object’ to denote the grammatical function of a phrase). In (1), ‘a letter’ is the affected or external object of the sentence, that is, the indicated object or person forms 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 effected or resultant or internal object, that is, the object indicated is brought about by the activity of writing.1 The grammatical distinction between external and resultant object is thus explained in semantic terms. Because the grammatical distinction is given a semantic explanation, it can be used to elucidate the distinction between action (Funktion) and product (Gebilde), a terminology that Twardowski took from Carl Stumpf.2 Twardowski is thus able to extend the distinction between external and internal object to non-linguistic entities. An interesting variant of the resultant object of a sentence to which Twardowski draws our attention, one may find in the sentence (3) Hans is singing a song for them.

40

MARIA VAN DER SCHAAR

The direct object ‘a song’ is called the cognate object.3 In such cases the noun and the verb are semantically and morphologically related (cognate). Here and in similar cases, like ‘to dream a dream’ or ‘to spring a sprong’, the verb is standardly intransitive, but is made transitive for stilistic reasons. The figure of speech concerned is called figura etymologica.4 This figure is explained as the junction of two words into one expression, in which the second word repeats the stem of the first. Usually, a verb repeats the stem of the direct object. The rhetorical function of the figure is that of emphasis. German (and Dutch), for example, uses the figure in the saying ‘selbst sein Grab graben’ (‘zijn eigen graf graven’) for the English phrase ‘to undo oneself’. The linguistic distinction between a verb indicating an action and the direct object indicating the product of that action 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. In ‘My dreaming was unexpected’, it is said that the activity of dreaming as such was unexpected. In ‘My dream was unexpected’ it is rather the content of the dream that is said to be unexpected. The two terms, ‘my dreaming’ and ‘my dream’, illuminate different aspects of the activity of dreaming. Although there is not an ontological distinction, there is a semantic, or, as Twardowski says, a ‘logical distinction’ corresponding to the linguistic distinction. In ‘They kept on making noise’, the topic is the activity of noise-making, whereas in ‘The noise made at your party was too loud’ it is the static aspect that is at issue. There is thus a logical distinction between the ‘dynamic’ aspect and the ‘phenomenal’ or ‘static’ aspect of the same activity. Philosophically, the distinction between action and product becomes relevant, if one uses the distinction to disambiguate terms such as ‘presentation’, ‘cognition’, ‘assertion’ and ‘judgement’ (Twardowski 1912e, 111ff).5 Each of these terms may denote the corresponding activity, 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

ACTION AND PRODUCT

41

with judgements and concepts as products. Confounding the two types of notions leads to psychologism in logic. Whereas psychology is interested in the causal relations between acts of judgement, logical consequence is a relation that holds between judgement-products (Twardowski 1912d, 165). According to Twardowski, the term ‘judgement’ can also mean someone’s capacity to judge in general, as we use it in ‘good judgement’ to indicate a given trait of a person’s character. Twardowski uses ‘judgement’ here in the traditional sense of faculty of judgement, giving it a psychological application. Besides, terms like ‘concept’, ‘conviction’ and ‘judgement’ have a more specific dispositional meaning. To possess a concept of something means to have the disposition to make a concept of that thing (idem, 167). Similarly, an enduring conviction, which may also be called a ‘judgement’, is a disposition to make similar acts and products in the future as one did in the past (idem, 169). Such a conviction is a potentiality in relation to its actualisations. Twardowski uses the distinction between action and product to explicate the notion of meaning.6 Before giving an exposition of his theory of meaning, it is necessary to explicate Twardowski’s distinction between enduring and non-enduring products, and his distinction between psychical or mental acts and products, physical actions and products, and psychophysical actions and products. Enduring products last longer than the activity from which they result. A building lasts longer than the activity of building, and a cake lasts longer than the activity of baking. Enduring products have in common that the activity indicated by the verb performs on preexisting material. According to Twardowski, psychical products, which result from nothing but a mental act, are non-enduring (Twardowski 1912d, 174). A thought that results from the activity of thinking is non-enduring; the judgement-product is non-enduring too. Physical and psychophysical products may be enduring. A physical product arises from a physical activity: a (non-enduring) sprong arises as a result of springing; an (enduring) burrow arises as the result of a rabbit’s burrowing. A psychophysical product is the result of a physical action that is accompanied by a mental act which has some influence on the resulting product (Twardowski 1912d, 174; 1912e, 109). A drawing or sentences spoken or written down are examples of psychophysical products.

42

MARIA VAN DER SCHAAR

The psychophysical product that is perceivable may be the external expression (Ausdruck) of a non-perceivable mental product (Twardowski 1912e, 120; 1912d, 175). A shout can be an expression of pain; a particular head movement an expression of an affirmative judgement. In general, the psychophysical product is the expression of a mental product, if (a) the mental act and product are partial causes of the psychophysical product, and (b) the mental act and product are not perceptible, whereas the psychophysical product can be perceived (Twardowski 1912d, 175). A linguistic product such as a (token of a) declarative sentence expresses the judgement made that was the (partial) cause of the sentence. The judgement made is the meaning of the declarative(-token) upon a further condition: the declarative potentially elicits, and thus partially causes, an act of judgement when someone reads the sentence, and the judgements elicited are similar to the judgement expressed by the sentence (Twardowski 1912d, 176–179). The non-enduring mental product, the meaning, may thus be said to ‘exist potentially’ in the psychophysical product in so far as the latter may become the partial cause of that mental product. We may abstract from the differences between the individual judgement-products that are elicited by a certain declarative, and discern a group of common attributes in these individual mental products, arriving thus at an abstractum, which is one and the same meaning of the linguistic product (idem, 182). Twardowski’s identification of this notion of meaning with the notion of ideal meaning in Husserl’s Logical Investigations is evaluated in the next section. Twardowski calls the mental product, the meaning, also the content (Inhalt) of the psychophysical product (idem, 179). A certain judgementproduct is thus not only the meaning but also the content of the declarative sentence to which it stands in a relation of causality. This is in accordance with Twardowski’s identification of the distinction between the act of presentation and the product of presentation he makes in Actions and Products with the distinction between act of presentation and content of presentation, introduced in Zur Lehre vom Inhalt und Gegenstand der Vorstellungen. For Twardowski, the product of an act of presentation is nothing but the presented content (Twardowski 1912d, 167; 1912e, 114, note 30). Like the mental product, the content of a mental act exists only as long as the act exists (Twardowski 1894, 30).

ACTION AND PRODUCT

43

Twardowski acknowledges that there are deviant uses of the declarative sentence in which it is not elicited by a mental act of judging. In case the declarative is used as an example, or if it is used by an actor on stage, the sentence is not the expression of a judgement-product. Logicians often work with examples of reasoning, whose premises may be false. In such examples, the premises are not expressions of judgement-products. According to Twardowski, such declaratives are expressions of artificial products that substitute for actual judgements. What is expressed is not a judgement passed, but a representation of a judgement (Twardowski 1912d, 184; 1912e, 129, 130). A representation of a judgement is a representation that has a judgement as object. This object is not a judgement that actually exists; as represented judgement it is not a real judgement. According to Twardowski, this explanation of the notion of represented judgement gives the best way to understand Bolzano’s notion of Satz an sich (Twardowski 1912d, 185).

3. An Evaluation of Twardowski’s Distinction between Action and Product, and its Application to Logic and Semantics As so many of Brentano’s better pupils, Twardowski was not a close follower of Brentano’s ideas. But Twardowski stayed faithful to the main characteristics of Brentano’s theory of judgement, which is non-propositional and anti-platonistic. As far as the theory of judgement is concerned, all acts of judgement are affirmations or denials of existence.7 Further, all judgements depend on acts of presentation, for the judgement gets its object through an act of presentation, and every act has an object, which may not exist. The judgements ‘God exists’ and ‘God does not exist’ both have God as object. According to Twardowski, these judgements differ as far as their content is concerned: the affirmation has the existence of God as content, whereas the denial has the non-existence of God as content.8 By bringing the distinction between content and object into focus, and applying this distinction systematically not only to the theory of presentation but also to the theory of judgement, Twardowski starts to deviate from Brentano, who does not give an important philosophical role to the distinction.

44

MARIA VAN DER SCHAAR

Twardowski’s reading of Husserl’s Logical Investigations not long after its publication in 1901, reinforced Bolzano’s influence on Twardowski, as well as his struggle with psychologism. In contrast to Bolzano and Husserl, Twardowski tries to defend himself against the criticism of psychologism without acknowledging an independent realm of logical entities. Bolzano and Husserl explain the objectivity of logic, semantics and science by the acknowledgement of propositions, which are explained independently of the notion of mental act. Propositions function as: (1) the contents of our judgements, (2) the meanings of declarative sentences, (3) the (primary) bearers of truth and falsity. For Brentano, the primary bearer of truth and falsity is the judgement, and the objectivity of logic is founded on the correctness, and eventually the evidence, of the judgement. One of the weaker points of Brentano’s explanation of the objectivity of logic is contained in the ambiguity of his use of the term ‘judgement’.9 Whereas Reinach has become famous for showing that Brentano’s concept of judgement is ambiguous between the act of judgement and judgement as belief or conviction, Twardowski has to be credited for pointing at another ambiguity in Brentano’s concept of judgement. In Brentano’s writings ‘judgement’ may mean either act of judgement or judgement-product. According to Twardowski, the primary bearer of truth and falsity is not the act of judgement, but the judgement-product. According to Twardowski, each of the three roles mentioned above are fulfilled by the judgement-product, as will be explained below. Twardowski identifies the product of a mental act with the content of that act. The judgement-product is thus nothing but the (1) content of the judgement, which, according to Twardowski, has the form the (non-) existence of A, that is, it does not have the indicative form. Judgement-products, as we have explained them in the first section, have the indicative form: A exists, or A does not exist. The judgement-product cannot both have and have not the indicative form. Therefore, the judgement-product cannot be the content of the judgement.

ACTION AND PRODUCT

45

Even if one asserts that the judgemental content is dependent upon the act of judging, as Twardowski does, one has to acknowledge the difference between the content and the product of a judgement for another reason. The same type of entity that functions as content of a judgement, may also function as the content of a question, or as the partial content of a hypothetical judgement. That God exists, or, as Twardowski would say, the existence of God can be the content of a judgement, but it (or a similar entity) can also be the content of a question. The content of a judgement cannot be the same type of entity as the judgement-product, for the latter is essentially the result of an act of judgement. Further, the notion of product of an action is different from that of content of an act, as the former is used with respect to mental, physical and psychophysical actions, whereas one can speak only of the content of mental acts. According to Twardowski, the judgement-product functions as (2) meaning of the declarative sentence. As there is no judgement-product in case the declarative is used to give an example, or if it is used as the antecedent of a conditional sentence, Twardowski needs to accord his theory of meaning. In such cases, he says, it is not the judgement-product that is the meaning of the declarative: it is merely a represented judgement that is the meaning of such uses of the declarative. That is, a declarative has two meanings depending on what we use the declarative for. But a theory of meaning should be able to give one meaning to each unambiguous term or sentence. It seems that another confusion has played a role in Twardowski’s explanation of the notion of meaning. Twardowski says that (a certain occurrence of) a sentence expresses the act of judgement and the corresponding judgement made, like a scream expresses pain (Twardowski 1912d, 176). This use of ‘to express’ in the sense of to make known or to intimate, needs to be distinguished from the use of ‘to express’ which takes the (linguistic) meaning as its object. Whereas a sentence expresses its meaning, a certain use of a sentence intimates the act of judgement and the judgement made. Notwithstanding the ambiguities in Twardowski’s theory of meaning, the notion of represented judgement is an interesting notion, for it has some similarities with the notion of judgement-candidate introduced in the first section. It is the judgement deprived of its judgemental force. It is not the actual judgement, but a potential one. In The Essence of Concepts,

46

MARIA VAN DER SCHAAR

Twardowski uses the notion of represented judgement to explain understanding and communication in general: “the understanding of the statement quite obviously consists of representing to oneself the judgment that comprises the meaning of the statement; the passing of a judgment is something entirely irrelevant to the understanding of a statement” (Twardowski 1924, 81). There are also some important differences between Twardowski’s notion of represented judgement and the notion of judgement-candidate. Whereas the judgement-candidate is an abstract notion explained in terms of the (potential) act of judgement, it is independent of any actual act of judgement. Twardowski’s notion of represented judgement is essentially the object of an actual act of representing. As object of an act of representing it is a concrete, but non-actual act of judgement. Twardowski’s notion of represented judgement is also different from Bolzano’s Satz an sich, a notion explained independently of the notions of act of judging or act of representing. In order to determine to what extent Twardowski’s semantics is psychologistic, Twardowski’s comparison of his notion of meaning as one and the same abstractum through a multiplicity of differences with Husserl’s notion of ideal meaning is relevant. Husserl deals in the first of his Logical Investigations with the notion of meaning, and relates it to the problem of the one and the many. Husserl identifies the meaning of a declarative sentence with the judgement in a logical sense. The judgement in a logical sense is the logical content, which is one and the same content in similar individual acts of judgement. In contrast, each of these similar acts of judgement has a phenomenological or psychological content that differs from any other. For Husserl, the ideal meaning is independent of any act of judgement, although we obtain such a notion only by the apprehension of acts of judgement. Whereas for Twardowski the logical judgement, the identical meaning, is explained in terms of individual judgement-products, which themselves are explained as results of judgemental acts, for Husserl, the logical judgement is prior in the order of explanation to the notion of judgemental act: it is as a species in relation to its instances (Husserl 1901, I, § 30–35). Because for Twardowski meaning is ultimately explained as an abstraction upon (products of) individual acts, his semantics does not escape psychologism. In contrast to Husserl’s semantics, Twardowski’s notion of meaning is (partly) explained in naturalistic terms in so far as a

ACTION AND PRODUCT

47

judgement-product is the meaning of a sentence only if that sentence is able to cause a similar judgement in those who perceive that sentence. According to Twardowski, (3) the primary bearer of truth and falsity is the judgement-product, and not the judgemental act. According to Twardowski, the judgement-product does not change its truth-value. In his article “On So-Called Relative Truths”, Twardowski discusses sentences that seem to change their truth-value over time (Twardowski 1900, 149–152). A sentence such as ‘It rains (here and now)’ is incomplete. The circumstances within which the sentence is pronounced help us to specify the judgement made, which is fully determined. Sentences are bearers of truth and falsity only in a secondary sense, whereas judgement-products are such bearers in a primary sense. Is Twardowski able to surmount the problem of psychologism in logic by making the judgement-product the bearer of truth and falsity? Logic is one of the humanities, that is, one of those “sciences whose objects [of study] are either mental products, considered independently of the mental functions that produce them, or psychophysical products, considered as such.” (Twardowski 1912/76, 139). Psychology is fundamental to logic, in the sense that the bearers of truth and falsity are essentially products of mental acts (1912e, 132). But the relation between the act of judgement and the judgement-product is not a natural relation of producing; the act of judgement is not the cause of the judgement-product. Further, the judgement-product’s being true does not depend upon the act of judging. Twardowski gives a variant of the correspondence theory of truth: “An affirmative judgment is true if its object exists, a negative judgment, if its object does not exist. An affirmative judgment is false if its object does not exist; a negative judgment, if its object does exist.” (Twardowski 1925, 208). Truth of a judgement is thus explained neither in terms of evidence of the judgement, as Brentano did, nor in terms of subsistence of objectives, as did Meinong.10 Twardowski’s explanation of the judgement-product and its truth and falsity is not psychologistic. Logical relations are relations between judgement-products as bearers of truth and falsity, and are explicitly distinguished from causality relations that obtain between judgemental acts. Because consequence relations between judgement-products are nothing but truth relations holding between these products that can a priori be determined, Twardowski is able to guarantee the objectivity of logic, for

48

MARIA VAN DER SCHAAR

these truth relations may be founded upon existence relations between objects that hold necessarily. Whether Twardowski acknowledges such existence relations I have not been able to determine, but I do not see any reason to accuse Twardowski of psychologism in logic.

4. The Distinction between Action and Product and that between Act and State Applied to the Concept of Knowledge In 1925 Twardowski gave a lecture course, entitled ‘Theory of Knowledge’, whose first lecture concerns terminological elucidations.11 In this lecture, Twardowski gives an explanation of the Polish terms ‘znać’ and ‘poznać’. These terms correspond to the German ‘kennen’ and ‘erkennen’ (and to the Latin ‘noscere’ and ‘cognoscere’). We may translate these terms into English by, respectively, ‘to know’ and ‘to cognize’ (or ‘to acknowledge’). It should be noted, though, that the English verb ‘to know’ does not exactly correspond to the German ‘kennen’. The German ‘kennen’ stands in contrast to the term ‘wissen’. ‘Wissen’ is derived from the perfect form belonging to a verb indicating a cognitive activity; it means to have knowledge. In general, it indicates possession of propositional knowledge. The Polish term is ‘wiedzieć’ (Latin has ‘scire’, but it also uses the perfect form of ‘nosco’, namely ‘novi’, meaning I have knowledge). The distinction between kennen and wissen can be found in most Indo-European languages. In the English language there used to be such a distinction, namely between ‘to ken’ and ‘to wit’, but these terms have become obsolete. The English verb ‘to know’ has taken over the meaning of both ‘to ken’ and ‘to wit’. Thus, ‘to know’ means to have knowledge, especially to have knowledge that. And, ‘to know’ can have the same meaning as the German ‘kennen’: to have a certain capacity to recognize or to understand (cf. ‘to know John’, ‘to know Latin’). Thirdly, the verb ‘to know’, and the German ‘kennen’ as well, originally has a more active meaning, which has now become obsolete. We no longer use ‘to know’ in the sense of ‘to become cognizant’ or ‘to get or learn to know’.12 The verb ‘to cognize’, being a comparatively modern word, can be used for this activity of coming to know in general, although the English language would prefer to use more specific terms, such as ‘to understand’, ‘to perceive’, or ‘to notice’. In gen-

ACTION AND PRODUCT

49

eral, there is a tendency in all European languages to use a term with the prefix ‘co-’, or an equivalent, for to get to know. Twardowski explicates the meaning of ‘to know’ (‘znać’) as to have knowledge, and the meaning of ‘to cognize’ (‘poznać’) as to acquire knowledge or to enter into a state of knowing (Twardowski 1925, 184). To summarize: English to cognize (to perceive etc.) to know, in the sense of ‘to ken’ to know, in the sense of ‘to wit’

German erkennen

Meaning to acquire knowledge

kennen

to have a certain capacity to recognize or understand to have knowledge (that)

wissen

Twardowski takes knowledge in a subjective sense, that is, as someone’s mental state of knowledge. If someone has the knowledge that S, he is capable of making a correct judgement that S. According to Twardowski, a further condition is needed for the explanation of knowledge. The judgement that S has to depend on memory. Not every capability to make a correct judgement counts as knowledge. I do not know at this moment how much money is in my pocket, as I have not counted the money. Although I have a capability to count the money, and thus to make a correct judgement about the sum, my capacity to do so does not count as knowledge. As soon as I have counted the money, and have determined that it is 35 złoty, I have knowledge, because my capacity to make the correct judgement is based on memory (idem, 185). It strikes the modern reader that Twardowski does not explain knowledge in terms of justification or warrant. Twardowski’s explanation rather improves upon Bolzano’s explanation of a piece of knowledge (eine Erkenntnis) as a judgement that contains a true proposition (Bolzano 1837, § 36). In Actions and Products, Twardowski makes a distinction between cognizing and cognition. Psychology is concerned with cognizing, whereas cognition, the product of cognizing, pertains to the theory of knowledge (Twardowski 1912e, 114, note 29; cf. the German ‘erkennen’ and ‘Erkenntnis’, 1912d, 166). Because cognizing is a mental act, cognition cannot be an enduring product, according to Twardowski’s explanation of

50

MARIA VAN DER SCHAAR

mental products. Cognition as product, as Twardowski explains it in Actions and Products, can thus not be identified with having knowledge, as Twardowski explains it in the 1925 lectures. It is possible, though, to conceive of the distinction between knowledge as act (cognizing) and knowledge as capacity (having knowledge) as a distinction between act and disposition. As we have seen, in Actions and Products Twardowski explains the notion of conviction or belief as a disposition to judge. Twardowski is thus able to explain an enduring state in terms of acts of judgements. The explanation of knowledge as capacity is similarly explained in terms of its actualisations, namely as a disposition to make the relevant correct judgemental acts. Twardowski thus explains the important notions for logic and theory of knowledge, that is, the notion of judgement-product, the notion of conviction or belief, and the notion of knowledge as mental state in terms of the act of judgement. In this sense, Twardowski is faithful to the Aristotelian explanation of the concept of knowledge as disposition (as hexis, or, as Thomas Aquinas has it, as habitus) as we may find it in De Anima (chapter 5, 430a10–26), and, in general, to the Aristotelian order of explanation as we may find it in the Metaphysics (book IX, 1049b13–14): the actual is prior in the order of explanation to the potential “for it is because it can be actualized that the potential, in the primary sense, is potential”.13 In modern explanations of the concept of knowledge, the focus is on the mental state of knowing. The act through which we obtain knowledge is not clearly distinguished from knowledge as state, or it is neglected, which neglect is reinforced by the lack of a common verb for the activity of getting to know. After all, ‘to cognize’ is a rather artificial term. The idea behind this neglect is most aptly expressed by Twardowski: such acts are the object of psychology, not of a theory of knowledge. But philosophy cannot neglect the act by which knowledge is obtained. A philosophical account of error, belief-revision and knowledge-acquisition, cannot be given without the notion of cognitive act. If it turns out that the knowledge I have obtained about the amount of money in my pocket through an act of counting, is not in agreement with my expenses, I reconsider my act of counting. I might have made a counting mistake. What seemed to be an act of knowing, may be nothing but an apparent act of knowing. Twardowski

ACTION AND PRODUCT

51

would be the first to point out that the word ‘apparent’ is a modifying adjective.

Notes 1

“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 zustande, 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. Twardowski consulted Brugmann’s Griechische Grammatik, cf. Twardowski 1912d, 171. His explanation of the distinction is also given in semantic terms. 2 J. Bergmann used the term Gebilde already in 1879, see Twardowski 1912e, 108, n. The German term ‘Funktion’ is better suited than the English term ‘action’ to cover, besides physical actions, both active and passive psychical processes. 3 The term ‘cognate object’ I took from A Comprehensive Grammar of the English Language, Longman, 1985. 4 Twardowski 1912d, 160. Cf. Historisches Wörterbuch der Rhetorik, ed. by G. Ueding, Darmstadt: Wissenschaftliche Buchgesellschaft, 1996. The term ‘etymon’ is used in the meaning of stem. 5 The distinction between act and product is acknowledged in the scholastic tradition, especially the Thomistic one. Jacques Maritain’s Petite Logique, summarizing the latter tradition, gives: “La distinction essentielle entre l’acte de l’esprit (jugement) et l’ouvrage logique construit par lui (proposition ou énonciation) ” (§ 35). The traditional concept of proposition (or enunciation) differs from the modern concept of proposition: it comprises the two notions of judgement-candidate and judgement made introduced in the first section. In Aquinas the product is called the terminus (ad quem). The terminus of an act of knowing, the ‘inner word’, is distinguished from the external object of knowledge. 6 An extensive treatment of Twardowski’s theory of meaning is given in Smith 1989; cf. also Buczyńska-Garewicz 1980 and Brandl 1996. 7 More details on Twardowski’s theory of judgement in Betti and Schaar 2004. 8 Twardowski 1894, p. 8 (cf. Brentano 1899, 27). On the same page, Twardowski seems also to defend the thesis that both judgements do have the same content: in the affirmation, the existence of the object is affirmed; in the denial, the existence of the object is denied. 9 Another problem in Brentano’s explanation of the objectivity of logic concerns an ambiguity of his notion of evidence, see Schaar 2003. 10 Concerning this point, there is a similarity between Twardowski’s ideas in his logic lectures from 1895 and Meinong’s theory, cf. Betti and Schaar 2004.

52

MARIA VAN DER SCHAAR

11

The text of this lecture is based on the notes that Izydora Dąmbska made when she attended the course in 1925, cf. her note: Twardowski 1925, 182. The course is published by I. Dąmbska in Polish in 1975. 12 The Oxford English Dictionary, as well as the Duden, is a Fundgrube for such terms. A philosopher has to deal with these linguistic data with care. 13 Twardowski’s reading of Aristotle was mediated by the interpretation of Brentano and Thomas. Cf. Twardowski’s Selfportrait, Twardowski 1999, 21. Brentano’s treatment of this aspect of Aristotle’s concept of knowledge can be found in Brentano 1862, 43.

References Betti, Arianna and Schaar, Maria van der 2004. ‘The Road from Vienna to Lvov: Twardowski’s Theory of Judgement between 1894 and 1897’, Grazer Philosophische Studien, 67, pp. 1–20. Bolzano, Bernard 1837. Wissenschaftslehre, in: Bernard Bolzano Gesamtausgabe I/II, 1–9, J. Berg (ed.) Stuttgart – Bad Canstatt: Friedrich Frommann Verlag – Günther Holzboog, 1985–2001. Brandl, Johannes L. 1996. ‘Kazimierz Twardowski über Funktionen und Gebilde’, Conceptus, 39, pp. 145–156. Brentano, Franz 1862. Von der mannigfachen Bedeutung des Seienden nach Aristoteles, Hildesheim: Georg Olms, 1960. Brentano Franz 1899. ‘Über den Begriff der Wahrheit’ in: Wahrheit und Evidenz, Hamburg: Felix Meiner Verlag, 1930, pp. 3–29. Buczyńska-Garewicz, Hanna 1980. ‘Twardowski’s Idea of Act and Meaning’, Dialectics and Humanism, pp. 153–164. Dummett, Michael 1973. ‘Assertion’, Chapter 10 in his Frege: Philosophy of Language, sec. ed., London: Duckworth, 1992, pp. 295–363. Frege, Gottlob 1918. ‘Der Gedanke’, in: Gottlob Frege, Kleine Schriften, ed. by I. Angelelli, Hildesheim: Georg Olms, 1967, pp. 58–77. Husserl, Edmund 1901. Logische Untersuchungen I in: Husserliana XVIII, The Hague, Nijhoff, 1975, and XIX/1–2, The Hague: Nijhoff, 1984. Reinach, Adolf 1911. ‘On the Theory of the Negative Judgement’ in: Parts and Moments, ed. by B. Smith, München-Wien: Philosophia Verlag, 1982, pp. 315–400. Schaar, M. van der 2003. ‘Brentano on Logic, Truth and Evidence’, Brentano Studien X, 2002/2003, pp. 119–150. Smith, Barry 1989. ‘Kasimir Twardowski’ in: The Vienna Circle and the Philosophy of the Lvov-Warsaw School, ed. by K. Szaniawski, Dordrecht, Boston, Lancaster: Kluwer, pp. 313–373. Sundholm, Göran 1999. ‘MacColl on Judgement and Inference’, Nordic Journal of Philosophical Logic, 3, pp. 119–132.

ACTION AND PRODUCT

53

Twardowski, Kasimir 1894. Zur Lehre vom Inhalt und Gegenstand der VorstellungenEine psychologische Untersuchung, Wien: Hölder (anastatic reprint: MünchenWien: Philosophia Verlag, 1982). Twardowski, Kasimir 1900. ‘On So-Called Relative Truths’ in: Twardowski 1999, pp. 147–169. Twardowski, Kasimir 1912d. ‘Funktionen und Gebilde’ ed. by J. Brandl, Conceptus, 39, 1996, pp. 157–189. This German version of Actions and Products, made by Twardowski, deviates from 1912e, which is a translation of the Polish paper. A shorter German version is published in 1911. Twardowski, Kasimir 1912e. ‘Actions and Products. Some Remarks from the Borderline of Psychology, Grammar and Logic’ in: Twardowski 1999, pp. 103–132. Twardowski, Kasimir 1912/76. ‘The Humanities and Psychology’ in: Twardowski 1999, pp. 133–140. Twardowski, Kasimir 1924. ‘The Essence of Concepts’ in: Twardowski 1999, pp. 73– 97. The article originally appeared in 1903. The parts I refer to additions of the 1924 edition. Twardowski, Kasimir 1925. ‘Theory of Knowledge. A Lecture Course’ in: Twardowski 1999, pp. 181–239. Twardowski, Kasimir 1999. On Actions, Products and Other Topics in Philosophy, ed. by J. Brandl and J. Woleński, Amsterdam, Atlanta: Rodopi.

The Strange Case of Savonarola and the Painted Fish On the Bolzanization of Polish Thought* ARIANNA BETTI

Introduction When Twardowski came to Lvov in 1895, his aim was to establish a philosophical trend heavily inspired by Brentanism, although peppered with Bolzanian ideas. As shows from a comparison between the lectures in logic he gave in Vienna in 1894/5 and in Lvov in 1895/6, Twardowski’s teaching activity in Poland was even more Brentanian than in Vienna.1 There is little doubt that the reason for this is that Brentano’s thought was unknown in late 19th Century Poland.2 It is well-known that Twardowski’s own thought was also significantly influenced by Bolzano, and that he played also a major role in disseminating some of Bolzano’s ideas in Poland, the most important being perhaps the notion of time-independent truth. I have previously discussed in several papers specific Bolzanian elements present in the Polish tradition. This paper will not, for the most part, add anything in particular to that. The new – and rather blunt – hypothesis to be put forward here is that, despite appearances, Twardowski also contributed de facto to slowing down the reception of Bolzano’s most modern logical discoveries. For in Poland Bolzano was to remain one logician among many for rather long. It was chiefly thanks to two factors that Bolzano’s star could, slowly, begin to rise in Poland, or, at least, that the fundamental achievements of his logic could be known. One factor is antipsychologistic (more precisely Platonistic) influence coming from Husserl and from Twardowski’s student Łukasiewicz. The other factor is the change in the conception of logic which took Polish logic from, say, Sigwart, to Tarski through Leśniewski and Łukasiewicz. What I am going to say is meant to have impact on the standard picture of the all-Brentanian background of the Lvov-Warsaw school, though my account will be limited to two pupils of Twardowski’s of the first generation, Łukasiewicz and Leśniewski. I hope this paper will contribute

Actions, Products, and Things. Brentano and Polish Philosophy. Arkadiusz Chrudzimski and Dariusz Łukasiewicz (eds.), Frankfurt: ontos, 2006, 55–81.

56

ARIANNA BETTI

both to the debate on the scope of Polish Brentanism, and to prompting further investigation into the reception of Bolzano’s thought in Poland.

I. Bolzano and the Standard Picture of the Background of the Lvov-Warsaw School The following is, roughly put, what for some time has been suggested in reconstructions of the philosophical heritage of the Lvov-Warsaw School. Twardowski was a direct pupil of Brentano’s; Łukasiewicz, Leśniewski and Kotarbiński were direct pupils of Twardowski’s, a sort of third generation of Brentanians, as it were.3 So, if the Lvov-Warsaw School has a grandfather, this was Brentano. Although more recently attempts have been made to include in the cast of characters also Bolzano,4 the idea that Brentano’s influence has been enormous and perhaps incomparable to that of any other contemporary philosopher in Poland is, as well-testified by the title of this collection, little disputed. The problem with this picture is its monolithic character: zoom in further, and you no longer see Brentanian homogeneity. By now it is common knowledge that particular Polish issues come from Bolzano. These include Twardowski’s and Leśniewski’s (versions of) sempiternal truth (the view that a truth-bearer does not change its truth value over time), Łukasiewicz’s logical probability, and Tarski’s logical consequence.5 More recently, Bolzano’s influence on Ajdukiewicz’s conception of analyticity, on Ajdukiewicz’s 1923 notion of consequence, and on the analysis of truth-bearers and their ontological counterparts in the early Leśniewski, Łukasiewicz and Kotarbiński have been added to the record.6 Telling the full story of Bolzano’s influence on the Lvov-Warsaw School may seem – details aside – a relatively easy historical task. For we seem to have a clear explanation for how the issues mentioned above were passed on, namely Twardowski’s mediation. This is certainly correct as to the sempiternity of truth, for that position is distinctively anti-Brentanian and Twardowski took it up explicitly in 1900. But for all other notions things are not this simple, and one cannot but suspect that the story of Bolzano’s influence has not yet been told in its entirety.

ON THE BOLZANIZATION OF POLISH THOUGHT

57

It is far from the aim of this paper to suggest that we should substitute Bolzano for Brentano in the role of grandfather of the Lvov-Warsaw School. This would be just naïve. I also do not intend to dispute Brentano’s general influence on choice of topics and methods in Polish analytic philosophy. There is truth in the picture, obviously, as Twardowski’s initial target was the exportation of Brentanism onto Polish soil. There is also indisputably much more Brentano than Bolzano in some of Twardowski’s students, one clear example being Tadeusz Czeżowski. But that’s exactly the point. There is little to object as to the Brentanian mark on choice of topics and methods insofar as it was Twardowski’s initial choice. The Lvov-Warsaw School in its entirety, however, brought together individuals with diverging philosophical and logical tendencies, and underwent internal developments which should not be underestimated. The variety and the change over time were due, among others, to the influence of Husserl, Frege and Bolzano, which in turn contributed to at least two Generationsstreite inside the Lvov-Warsaw School. Sure, Husserl was himself a Brentanian, but he was Husserl, not Brentano: variety reigned in the ‘loose association’ – to use an expression of Kevin Mulligan’s – of the Brentanians as well. The revolt against Brentano in the theory of judgement – which resulted, among others, in Husserl’s, Meinong’s and Twardowski’s introduction of states of affairs in their ontologies – makes it in fact three Generationsstreite.7 All this should make us wary of looking at Leśniewski, Łukasiewicz, Kotarbiński, let alone Tarski, as ‘Brentanians’, flat.8 Therefore, the reason why the picture of the Brentanian roots of the Lvov-Warsaw School is so homogeneous is, it seems, that it is taken from the perspective of Twardowski’s Brentanian background, peppered by isolated, specific Bolzanian points. Can we say that Twardowski misunderstood Bolzano? As to the specific issues just mentioned – locally, one might say – and against what some scholars have claimed – we cannot. But we can say that as to Twardowski’s global consideration of Bolzano’s thought. It is the latter I am concerned with in this paper. Since the former issue is controversial, however, I’ll say first something about Twardowski’s construal of particular Bolzanian ideas.

58

ARIANNA BETTI

Twardowski is said to have seriously misunderstood Bolzano in his major work, On the Content and Object of Presentations (1894). True, Twardowski exploited in his favour Bolzano’s two-fold notion of the object of a (subjective) Vorstellung, but I think we should take this as part of the philosophical game, and not as a misrepresentation of Bolzano’s thought, pace Husserl.9 To see the fundamental fairness with which Twardowski treats Bolzano, one need only compare the account of Bolzano’s ideas given in his book with Marty’s or Meinong’s later misunderstandings. Take for instance Marty’s claim in 1908 that Bolzano did not have false propositions-in-themselves alongside true ones. Or take Marty’s personal Bolzanologist Shmuel Hugo Bergmann, who did concede that Bolzano had false propositions-in-themselves, but thought he would have been better off without. And, finally, take Meinong’s identification of his Objektive with propositions-in-themselves, the reason being, as in the case of Marty and Bergmann, that propositions-in-themselves were mistaken for truth-makers.10 Such mistakes Twardowski never made. Twardowski took over a series of distinctively Bolzanian traits without perverting their role or their significance, and this, as one may expect, was of no little importance for the Polish reception of specific Bolzanian issues.11 These comprise, first of all, Bolzano’s distinction between object and content of presentations which, as is well-known, Twardowski rescued in 1894; the consequent distinction and discussion of the relationship between parts of the content (of a Vorstellung an sich in Bolzano) and parts of the object of that Vorstellung; the difference between content and extension of an idea; and, as already mentioned, Bolzano’s view that truth is sempiternal. A very Bolzanian trait in Twardowski’s thought is also that he resisted – unlike Husserl and Meinong – the introduction of Annahmen in favour of a Bolzano-flavoured semantic ascent: in Twardowski you can just present a judgement without judging, as in Bolzano you can just present a proposition-in-itself without judging. All these aspects left a long-lasting mark on the Lvov-Warsaw School. Nevertheless, Twardowski seems to have misunderstood Bolzano in a global sense, that is to say, he failed to see his greatness as a logician. So did most others, but this is no excuse. Twardowski’s global misunderstanding of Bolzano is relevant for a correct assessment of the dissemina-

ON THE BOLZANIZATION OF POLISH THOUGHT

59

tion and the popularity of Bolzano’s thought in Poland, and of Twardowski’s role as a Vermittler of Bolzanism. For it is the hiatus between the two aspects, the local and the global, that makes historical pictures like the one at issue here difficult to take. In the following I will restrict my discussion to the relationship – from the global point of view just sketched, Twardowski’s grasp of Bolzano’s stature – between Bolzano on the one hand, and Leśniewski and Łukasiewicz on the other hand. Philosophers by training turned logicians, Leśniewski and Łukasiewicz were the pupils of Twardowski’s with more distinctly logical interests. Their case seems particularly interesting not only because they were to be teachers, together with Kotarbiński, of Tarski, but because they had been actors of primary role in a crucial moment of transition for logic as a discipline.

II. Leśniewski, Savonarola and The Painted Fish As far as I know, the sole remark about Bolzano we have in Leśniewski’s small oeuvre is from 1927, refers to the 1914 edition of Wissenschaftslehre and is quite incidental.12 For I did not consider it to be out of the question that Mr. Fraenkel uses the expression “improper set” in the fashion of expressions known from ordinary language – “dead man”, “false diamond”, “painted fish” and so on (see for instance Hauptwerke der Philosophie in originalgetreuen Neudrucken. Band IV. Werke Bernard Bolzanos I. Wissenschaftslehre in vier Bänden. Erster Band. Leipzig 1914. p. 257 and also the quotation from Savonarola on p. 79).13

Leśniewski mentions here the notion of modifying adjectives and he refers to the §§ 19 and 59 of the Wissenschaftslehre. Although the distinction between modifying and determining adjectives had been known since Aristotle, it was very popular among Brentanians, and it played a key role in Twardowski’s 1894 theory of intentionality. It was not at all customary to associate this notion with Bolzano, nor is it today: why does Leśniewski here quote the Great Bohemian where one would have expected him to see Brentano, or at least his master Twardowski? If the picture showing a pervasive presence of Brentanian heritage in the Lvov-Warsaw School would

60

ARIANNA BETTI

be entirely correct, this should not have happened, one might argue. Leśniewski should have quoted Brentano or Twardowski. So why does Leśniewski quote Bolzano? Consider the following hypothesis. The logicians among Twardowski’s students became aware at a certain point of a tension between a Bolzanian (or Bolzanian/Fregean) and a Brentanian trend in their philosophical milieu, being consequently driven towards the former. Actually, Leśniewski’s choice to quote Bolzano in this particular connection may give the impression that he was siding with a crucial figure of the Bolzano-Rezeption, namely Benno Kerry. Kerry was the first among Brentano’s pupils to study Bolzano seriously. It was he who attracted Twardowski’s attention to Bolzano, and he was the only ‘organic’ Brentanian to proclaim that Brentanians could have learned a great deal from Bolzano. Instead, the Brentanians were ignoring him. Kerry writes: Brentano […] declared that the relation of predicates to their subjects according to which the latter are not so much determined as modified by the former has been “hitherto overlooked by logicians”: he could have found this point in §23 (I, S. 92) and §29 (I, S. 138) of B’s Wissenschaftslehre. Ibid. in §19 (I, S. 79) even Brentano’s example of the “dead man” that is no man is adduced from Savonarola’s Logic.14

Juxtaposing quotations might yield witty results. But isn’t my case too thin? First of all, one might say that if Leśniewski’s passage quoted above was an attempt to express appreciation for Bolzano, it was a twisted one indeed. Secondly, however remarkable might be the concordance on Savonarola – by far more known for his burning reformatory zeal and the bonfires of vanities in the 15th Century Republic of Florence than for his logic, except to a careful reader of Bolzano –, I would still need to show that Leśniewski read Kerry’s 1885–1891 series of papers, if what I say about Kerry should support my cause. Unfortunately, there is no such evidence available; furthermore, Leśniewski’s reference to Bolzano is too occasional. Is my scenario – the preference of Twardowski’s more logically oriented students for the Bolzanian over the Brentanian trend – unlikely, then? Before nodding, think first of Twardowski’s chronicle of the major

ON THE BOLZANIZATION OF POLISH THOUGHT

61

influences on his development, to wit, Brentano and Bolzano, whose Wissenschaftslehre he said he studied “eifrigst”, after Kerry had drawn his attention to it.15 But note also that, despite appearances, Twardowski did not realize, at the beginning of his activity in Poland, that Bolzano’s logical achievements surpassed those of all his contemporaries. Very positive and explicit judgements on Bolzano’s greatness in this sense and on the Wissenschaftslehre are to be found only in Twardowski’s Actions and Products (published as separatum in 1911 and in 1912 in a collection), a witness of Twardowski’s anti-psychologistic turn, influenced by Husserl.16 In fact, it is a fairly established matter that at the beginning of the 20th Century Husserl had perhaps been the sole to grasp the greatness of Bolzano’s work.17 Now, Actions and Products is a work that Leśniewski read for sure. So, one could think that it was chiefly thank to Twardowski that Leśniewski (if at all, in 1911) got to know about the Wissenschaftslehre. Things, actually, appear to be more complicated than this. Actions and Products is a later work by Twardowski. It was published 16 after Twardowski’s arrival in Lvov, 17 years after Zur Lehre, and 10 after Husserl’s Logische Untersuchungen. That is to say, something must have happened in the meantime that had to do, at the very least, with Bolzano, Husserl and psychologism. But what exactly? When Twardowski, in his early Polish years, was asked to recommend a logic textbook to one of his first pupils, his choice did not fall on Bolzano’s Wissenschaftslehre, but on Höfler and Meinong’s Logik. The year was 1898, the pupil Łukasiewicz.18 Recalling those times fifty years later, Łukasiewicz saw the things as follows: Twardowski held in high esteem the writings of another priest who lived in the first half of the 19th century, Bernard Bolzano. Bolzano was professor of science of religion at the University of Prague and was an eminent mathematician and logician. His works in the field of philosophy have an incomparably higher scientific level than the philosophical prattle of Kant or Hegel. If Twardowski had realized the difference between the scientific method applied by Bolzano and the confused and often giddy prattle of German philosophers he would have been able to create a new trend of scientific philosophy, surpassing in worth the views of the Vienna Circle. Twardowski was instead […] fascinated […] by later philosophical writings, infected with psychologism.19

62

ARIANNA BETTI

This passage is taken from a page of Łukasiewicz’s unpublished memoirs in Polish written in 1949. The context is the author’s memoirs from 1904 regarding the foundation of the Polish Philosophical Society by Twardowski on the centenary of Immanuel Kant’s death. It is understandable that Łukasiewicz should use such unkind words for German idealism, but it may seem incredible that he should have come close to charging Twardowski with obscurantism. Łukasiewicz’s animosity against Twardowski is less puzzling than it may seem at first, though. Although it is conceivable that after the war, having emigrated to Dublin, Łukasiewicz was not well-disposed towards Poland and Polish matters while writing his memoirs,20 his dissatisfaction towards Twardowski in his Lvov years had several other reasons, both personal and scientific. One scientific reason was Łukasiewicz’s antipsychologism, another a Generationsstreit over the conception, role and significance of logic. Both are relevant for my Bolzanian purposes here, and to them I shall devote the following two sections before coming back to Leśniewski and Savonarola.

III. Łukasiewicz, Twardowski and Psychologism Husserl’s Logische Untersuchungen, as we know, contained a keen appraisal of Bolzano, both in words and in facts, to such an extent that Husserl felt compelled to tell his readers that he was not just glossing the Wissenschaftslehre.21 Thanks to Husserl’s work Łukasiewicz began to develop a deep aversion to psychologism early on. It is also reasonably certain that Łukasiewicz played a primary role in Twardowski’s own departure from psychologism.22 The exchanges between Husserl and Twardowski were insignificant, and Twardowski did not have any opportunity to discuss with Husserl in person. Łukasiewicz, Husserl’s Lvov advocate, was, on the other hand, a constant interlocutor of Twardowski. The first volume of Husserl’s Logical Investigations made a big impression in Lvov, especially on me. I had been disliking the psychologism cultivated by Twardowski already for long time, now I had detached myself completely from it.23

ON THE BOLZANIZATION OF POLISH THOUGHT

63

This passage is backed up by a lecture Łukasiewicz held in 1904, “Husserl’s thesis on the relationship between logic and psychology”, in which he claimed, following Husserl, that the logical laws cannot be based on psychological ones, because the latter are just probable, while the former are certain.24 Note, however, that in the passage above Łukasiewicz meant only the first volume of the Investigations, the Prolegomena zur reinen Logik (1900). The second volume (1901) he severely reproached for being too imbibed with psychologism “to succeed in some brave reform of pure formal logic”.25 […] the second volume of Husserl’s logical investigations disappointed me. It contained again that certain murky prattle which was driving me away from all German philosophers. I was astonished that there could be such a difference between two volumes of the same work. Later I convinced myself that in the first volume of the logical investigations it was not Husserl talking to me, but only someone much greater than he, whom Husserl exploited in his book, and this was Gottlob Frege.26

Łukasiewicz was not the only one to be astonished by the second volume of the Investigations. The astonishment was due to the two notions of psychology employed by Husserl and common among Brentanians: genetic psychology, an empirical science like biology, and descriptive psychology, or phenomenology, which was what interested Husserl and was considered an a priori science, like mathematics.27 Łukasiewicz accepted only the first understanding of psychology, and it is on this basis that we must understand the disagreement between him and Twardowski which emerged during the discussion of a lecture from 1904 on Husserl.28 In his lecture Łukasiewicz claimed that psychological laws are probable, to which Twardowski objected that if they are evident judgements, as in the case of inner perception, then they are as certain as mathematical laws. Łukasiewicz replied that general psychological judgements are probable even if the particular judgements on which they are based are certain: far from deriving deductively from definitions, these judgements derive inductively from observations about real objects, because psychology is an empirical science, and empirically are its laws obtained. Finally, Łukasiewicz explained that evidence could not be considered a criterion of truth.29

64

ARIANNA BETTI

In short, Łukasiewicz harboured a hostile attitude against what he took to be Twardowski’s conservatism in questions related to psychologism in logic. In his Memoirs he went as far as writing this: The apparatus of ideas and problems which Twardowski brought with him from Vienna to Lvov was incredibly poor and sterile. Whether a conviction was a mental phenomenon of a separate kind or a connexion of concepts was incessantly under discussion, intuitions, presentations, concepts, their content and object were incessantly under discussion, and no one knew whether the analyses carried out […] belonged to psychology, logic or grammar.30

And driven by his antipsychologism and Platonism, in his memoirs Łukasiewicz never misses an occasion to heap bitter words on works of Brentanian inspiration whenever such works seemed to him excessively psychologistic.31 There were aspects of Brentanian thought that Łukasiewicz did appreciate, but those were limited to neo-Aristotelian metaphysics and its formal treatment, like Twardowski’s mereology in Zur Lehre.32 One might well object that these are not particularly distinctive Brentanian traits, but Leibnizian and Bolzanian as well. Łukasiewicz’s attitude towards the Brentanians seems to have undergone a change in 1907, when he became fascinated with the ideas of Alexius Meinong, who had also studied with Brentano, contained in his (then brand-new) work on Gegenstandstheorie.33 Meinong’s theory of objects had essentially two features dear to Łukasiewicz: it was a modern revival of Aristotle’s metaphysics (a good Brentanian element),34 and, with its plethora of non-existents and possibilia, it dovetailed with Łukasiewicz’s surge towards a non-Aristotelian logic. One reason why Łukasiewicz preferred Meinong’s ideas to Twardowski’s (even if behind Meinong there was Twardowski’s pioneering and influential talk of contradictory objects) is probably the circumstance that Twardowski, unlike Meinong, never rejected either the principle of contradiction or the principle of the excluded middle. Still, Łukasiewicz’s fascination with Meinong was not to last long. Moreover, more typically Brentanian traits of Meinong’s thought, like the role of evidence, failed to win any favours with Łukasiewicz:

ON THE BOLZANIZATION OF POLISH THOUGHT

65

To exploit the concept of evidence as a criterion of truth is a relic of ‘psychologism’ which led philosophical logic into non-viable directions. Psychologism maintains strict links with subjectivism and skepticism. If evidence is a criterion of truth, then every judgement seeming evident to someone is true. […] Every truth becomes, thus, something subjective and relative, and absolute and objective truth ceases to exist.35

This passage comes from Łukasiewicz’s book on Aristotle, well-known in the Polish environment at the time. These were provocative words for Twardowski, who ten years before had fiercely defended the absolutism of truth from relativism and skepticism, on the basis of unequivocally Bolzanian – though not antipsychologistic – arguments.36 Łukasiewicz’s insistence was successful, and he eventually managed to influence Twardowski’s antipsychologistic turn significantly. In a manuscript called Psychology of Thinking, containing the lectures of a course given by Twardowski in Lvov in 1908/09, we read: The relationship between psychology of thinking and Logic has not always been accepted and presented in the right way. For some define Logic as the science of thinking or the science of correct thinking, are inclined to construe logic directly as the psychology of thinking, and, therefore, as part of psychology, or at least as some application of psychology. In this view Logic, although it is not a part of it, should be based on psychology. This conception, known as psychologism, is not tenable.37

This all-Polish evidence shows that Husserl was not the sole stimulus Twardowski had when taking his ideas into an anti-psychologistic direction. One might be tempted to think the contrary, since in Twardowski’s Actions and Products, where the exact separation of products from acts is said to have contributed in a decisive way to free logic from the influence of psychology,38 Husserl’s ideal in-specie meanings play a prominent role. But we can be confident enough that in the manuscript passage just quoted it is Łukasiewicz’s anti-psychologistic stance on the relationship between logic and psychology that Twardowski echoed.

66

ARIANNA BETTI

IV. Already Discovered Americas, that is, Old and New Logic Łukasiewicz was a philosopher with a mission: reforming logic. There is no work by him that cannot be read as an attempt to show that traditional logic was deficient and that it was in need of profound reform. This never meant a Brentanian reform, though. Reform would, in his view, pass via Platonism – a strong Bolzanian element. Leśniewski (publicly and solemnly) and Łukasiewicz (privately) both rejected their own early writings. Although their conceptions of logic were to radically differ, one of the strongest reasons for their recantations was the fact that at that time their logical tools were too poor. Here’s Leśniewski’s recantation, in his characteristic style: Steeped in the influence of John Stuart Mill’s logic in which I mainly grew up, and ‘conditioned’ by the ‘universal-grammatical’ and logico-semantical problems in the style of Edmund Husserl and of the exponents of the so-called Austrian School, I attacked unsuccessfully the foundations of ‘logistic’ from this point of view. […] Living intellectually beyond the sphere of the valuable scientific achievements of the exponents of ‘Mathematical Logic’ and yielding to many destructive habits resulting from an one-sided, ‘philosophico’-grammatical culture, I struggled in the works mentioned with a number of problems which were beyond my powers at that time, discovering already discovered Americas on the way. I have mentioned those works wishing to point out that I regret very much that they have appeared in print at all, to formally ‘repudiate’ them herewith – though I have already done this long ago from my university chair –, and to affirm the bankruptcy of the ‘philosophico’-grammatical enterprises of the first period of my activity.39

We do not have equally interesting quotes from the mature logician Łukasiewicz, but we find similar passages in tone and content in his memoirs, where he seems to be almost embarrassed by the poor knowledge of logic he had when lecturing to students on the algebra of logic (1906). Łukasiewicz regretted that his logical training under Twardowski was modest. He was almost incapable of hiding his contempt when recalling Twardowski’s course on “Reformatory trends in logic” (1898/99) on Hamilton, Boole and others, since, as it later turned out, it was based on Liard’s “very poor dissertation on English logicians”.40

ON THE BOLZANIZATION OF POLISH THOUGHT

67

If we follow Łukasiewicz’s own account, it was no thanks to Twardowski that he discovered mathematical logic in 1906: he became interested in Russell through one of his colleagues, Marian Borowski, and was immediately fascinated by the exactness of Russell’s method “without philosophical drivel”. He then ordered Russell’s Principles of Mathematics through the University library in Lvov. He studied the book for months on end, although he devoted himself to mathematical logic only later, during the First World War.41 The irony is that, objectively speaking, Łukasiewicz was, in his early writings, less of a sharp philosopher than Twardowski. In his early writings Łukasiewicz would often confuse use and mention of words, and there are lethal type-confusions of objects, their concepts, classes, elements of those classes, etc... Few managed to avoid this at the time, but Twardowski and Leśniewski were among these few. And it must be said that Łukasiewicz’s craving for exactness and clarity in writing and thought was a result of Twardowski’s teaching, and for these ideals Twardowski had to thank Bolzano and Brentano in equal measure. Yet Łukasiewicz was not ready to credit Twardowski with exactness in thinking, but only with the ability to systematise and clarify even the most intricate problems.42 When Łukasiewicz met Leśniewski, moreover, it became harder to regard Twardowski as an example of clarity and exactness, as the comparison with the obsessive formal perfectionism of Leśniewski was certainly to Twardowski’s disfavour. Twardowski was not only too old to become an enthusiastic follower of formal logic himself, he was also not pleased by the progressive ‘logicization’ of his students, Łukasiewicz in primis. And Łukasiewicz, then in full swing with his logical research on polyvalent logics, was almost certainly the primary target of Twardowski’s reproach in “Symbolomania and pragmatophobia” (1921/22).43 What about Bolzano’s influence upon Łukasiewicz as to some specific logical issues? The most extensive discussion of Bolzano’s ideas we find in Łukasiewicz’s Die logischen Grundlagen der Wahrscheinlichkeitsrechnung (1913). According to Łukasiewicz it was Twardowski who attracted his attention to Bolzano’s notion of validity: interestingly, Łukasiewicz says that he had already known the Wissenschaftslehre “for a long time”, although he

68

ARIANNA BETTI

did not pay attention to the notion of validity.44 The core of Łukasiewicz’s piece on probability was already ripe in 1909.45 Twardowski had mentioned Gültigkeit – calling it ‘logical value’ – in 1911, in Actions and Product, but no doubt Twardowski knew already Bolzano’s theory of variation, because an example involving the theory is quoted in his paper on relative truths (with no mention of Bolzano). An important aspect linked with the fact that Łukasiewicz ‘discovered’ Bolzano’s notion of validity, is his subsequent spreading of Bolzano’s ideas on the subject. Ajdukiewicz, who formulated a rather Bolzanian, preTarskian, definition of logical consequence in 1923, quoted Bolzano in 1913, in his Polish work “On the convertibility of the relation of consequence”, thanking Łukasiewicz for attracting his attention to Bolzano. Note, en passant, that this circumstance does not seem to have been noticed in all its significance: in its light, do we really still want to believe that Tarski did not know Bolzano’s notion of consequence when formulating his own? In conclusion, for some important aspects Łukasiewicz seems to have been, in practice, a bigger ‘Bolzanizer’ in the Polish tradition than Twardowski. To reinforce this claim I shall now turn to Łukasiewicz’s possible role in directing Leśniewski’s interest towards Bolzano. I should warn the reader, though, that the whole matter of the relationship between Leśniewski and Bolzano remains hypothetical.

V. Leśniewski, Bolzano and the Painted Fish Back to Leśniewski and Savonarola. Apart from siding with Benno Kerry, there is another possible reason why Leśniewski should have referred to the passage about modifying adjectives from the Wissenschaftslehre. Leśniewski’s quote is from On the Foundations of Mathematics and the context is his criticism of set theory. As is known, this work has an “autobiographico-synoptical character” and the quote corresponds to the years 1912–14. It is likely that Łukasiewicz and Leśniewski discussed the Bolzano/Savonarola passage precisely in those years, and that Łukasiewicz, again, played a role in attracting Leśniewski’s attention to Bolzano.

ON THE BOLZANIZATION OF POLISH THOUGHT

69

How did Leśniewski enter into contact with Bolzano’s thought? An anecdote told by Kotarbiński has Leśniewski arrive in Lvov in 1910 carrying Marty’s Untersuchungen under his arm.46 As I have shown elsewhere, Leśniewski ceased to be a follower of Marty’s ideas in 1911, and arguably had come into contact with Bolzano’s ideas at least by that time (recall that Marty was an orthodox Brentanian). There are striking similarities (mutatis mutandis) between Bolzano’s theory of truth and Leśniewski’s theory in 1911–12, not to mention their common, overall classical, approach, which Leśniewski never abandoned.47 Another strong element of similarity between Leśniewski and Bolzano is the adherence of both thinkers to what has been called the Aristotelian Model of Science, and in particular to the idea that a science is a collection of truths (objective and independent of time!) about the objects of that science, with a precisely determined structure, an idea to be found everywhere in Leśniewski’s work.48 By 1910 Leśniewski had access to two sources pointing explicitly towards Bolzano: Twardowski’s Zur Lehre, and, most importantly, Husserl’s Logische Untersuchungen, again, like in Łukasiewicz’s case. Note that Leśniewski could have known the Logische Untersuchungen even before knowing Zur Lehre through Alexander Pfänder, with whom he had studied in Munich (1909/10).49 And we know that Leśniewski even translated into Polish at least part of Husserl’s Untersuchungen.50 The other book he was keen to translate at the very beginning of his Lvov days was none other than Marty’s Untersuchungen. And since he was so keen on Marty’s book – if Kotarbiński is right – he may have read Husserl’s Bolzano-flavoured 1910 review of it. Leśniewski’s main interest at that time (and in fact for many years to come, up to 1920) was the regimentation of natural language in order to turn natural language into an adequate tool for scientific argumentation: both Husserl, especially in his Fourth Investigation, and Marty dealt with this. Interestingly, the Fourth Investigation, the one in which Leśniewski found the ancestor of his semantic categories (Husserl’s Bedeutungskategorie) is also the one in which Bolzano is most quoted. §2 of Husserl’s “Fourth Investigation” discusses Bolzano and Twardowski’s Zur Lehre on the ‘Land ohne Berge’ example; §4 is on syncategorematic expressions and Bolzano’s ‘mistakes’; §12 is on the distinction Unsinn/Widersinn, Bolzano and the ‘round square’. All considered Bolzano must have been already at least a household name for Leśniewski when he joined the group of Twar-

70

ARIANNA BETTI

dowski’s students. Then, in 1911, Actions and Products came out, so Leśniewski was exposed also to Twardowski’s explicit appraisal of Bolzano, and in the same year Leśniewski and Łukasiewicz met in Lvov. All these elements, however, are significant only in the light of the fact that by 1911 Leśniewski had a specific reason to consider Bolzano, namely his disagreement with Łukasiewicz on the principle of contradiction, made public in his 1912 article “An attempt at proving the principle of contradiction”. On that occasion he defended the Bolzanian and anti-Twardowskian, anti-Meinongian, anti-Łukasiewiczan view that the name “round square”, though having meaning, does not denote anything.51 This was a good reason to look carefully at Twardowski’s fight against Bolzano’s gegenstandslose Vorstellungen in Zur Lehre, §5. Moreover, someone else also attracted Leśniewski’s attention to this point, namely Leon Chwistek. In a work from 1912 which Leśniewski quoted in 1913, Chwistek claimed that Bolzano stood up against the argument that expressions like “round square” are nonsense, distinguishing “the concept of nonsense from what is contrary to sense”, again referring to §12 of Husserl’s “Fourth Investigation”.52 Chwistek’s work, like Łukasiewicz’s, deals with the principle of contradiction and centres on various paradoxes and antinomies. Russell’s antinomy, mentioned in Łukasiewicz’s book on Aristotle, on the one hand awoke an interest in all other known paradoxes and antinomies, and on the other brought up – at least in Poland – a discussion of the universal applicability of the principle of contradiction. For if even mathematics is contradictory, why keep the principle of contradiction? That’s why Leśniewski, who could never bring himself to believe that there was anything wrong with mathematics,53 was so interested in proving the principle of contradiction: he was convinced that there could never be contradiction in reality or in our reconstruction of it, and particularly not in mathematics. Enter Savonarola. In Leśniewski’s philosophy at the time, as was standard in the Lvov philosophical milieu, Russell’s paradox and three other themes, the principle of contradiction, the principle of excluded middle and the liar paradox, were brought together under the same heading: the semantics of empty names and in particular the semantics of what Bolzano would call ‘objectless propositions’. Both Leśniewski and Łukasiewicz worked on their respective solutions to the Liar in the period to which

ON THE BOLZANIZATION OF POLISH THOUGHT

71

Leśniewski’s passage from On the Foundations of Mathematics corresponds. Leśniewski’s solution, published in 1913, comprises a ban on selfreference, which also Bolzano discusses, and he thinks, like Bolzano, that the Liar is false;54 Łukasiewicz maintained in 1915 that Liar-sentences are outside the scope of logic: they aren’t well-formed sentences and they cannot be employed in logical transformations.55 Now Łukasiewicz’s stand is exactly the same as Savonarola’s as quoted by Bolzano: the Liar is no proposition, exactly like a dead man is no man and a painted fish is not a fish.

Conclusion Suppose that what I said to back up my case is convincing, that indeed Twardowski blindness to Bolzano’s stature and his overall Brentanian teaching was a regrettable thing to some of his pupils, and that it was in fact rather Łukasiewicz the one who played the main role in adjusting Bolzano’s image and in spreading some of the novelties of Bolzano’s logic. Well, then, there is still a good number of questions which the above considerations leave wide open. We may agree that you have to be a logician, or at least logically-minded, possibly an anti-psychologist, and perhaps a Platonist, to appreciate Bolzano. We do not have much material, though, let alone published or in vehicular languages, that might help us disentangle the as yet intricate matter of Bolzano’s influence in Poland and how this related to Brentanian inputs, even if we limit the scope, as I have done, to some of Twardowski’s students. The conjectures I have advanced, including Twardowski’s negative role, need more support than I have been able to provide. What I said does not go far towards proving, for instance, that Leśniewski did read the Wissenschaftslehre, or, if so, read it before the 1914 edition appeared, if this is what we aim at. Could he? Was a pre-1914 edition of Wissenschaftslehre available to Twardowski’s students? The kind of philosophy Twardowski was to teach was terra incognita when Twardowski moved to Lvov, so there is no guarantee that a copy of Wissenschaftslehre was around before 1914. Unless, of course, Twardowski brought a copy with him, for he put his personal library at his students’

72

ARIANNA BETTI

disposal. Was the copy he used in Vienna his own? To quote Peter Simons,56 “answers on a postcard, please”.57 My aim has been to show that a critical in depth assessment of the real meaning of “Polish Brentanism”, independently of what we can infer from Twardowski’s own story, and with particular attention to Bolzano’s role, is an urgent matter, though not an easy one.58 To make headway with this we will have to publish editions, search for manuscript material, train young scholars interested in digging things out. Do pracy! Notes *

A version of this paper has been presented at the Bolzano Atelier in Geneva in September/October 2001 under the title “Twardowski’s Misunderstanding of Bolzano and its Polish Consequences”, which I owed to Kevin Mulligan, and under which other versions have circulated so far. I am indebted to Arkadiusz Chrudzimski and Jan Woleński for comments, to Bjørn Jespersen, Göran Sundholm for suggesting language improvements, and to Marije Martijn for comments, language and stylistic improvements on a previous version. Thanks also to Jan Woleński and Jan Siek, director of the library of the Institute of Philosophy and Sociology of Warsaw University, for having provided me with copies of the valuable manuscripts I use in the text, and to Jacek Juliusz Jadacki, custodian of Twardowski’s Archives, for allowing me to quote from this material. Work on this paper has been funded by the Netherlands Scientific Research Organization (NWO), project no. 275-80-001. 1 Cf. Twardowski (1894/5) and Twardowski (1895/6). The edition of these manuscripts is in progress. A 50-page draft English résumé of the latter can be downloaded from the Polish Philosophy Page at http://www.fmag.unict.it/PolPhil/Tward/TwardLog.html The manuscripts and the letters quoted in this paper are all housed in the Library of the Institute of Philosophy and Sociology at the University of Warsaw. 2 Cf. Ingarden (1938), p. 25. 3 See the influential Woleński & Simons (1989). 4 See for instance Woleński (1999b). 5 See Betti (2005a), Künne (2003), pp. 180–4, Woleński & Simons (1989), pp. 430 fn 24. Siebel (1996) challenges the vulgate on the Bolzano-Tarski consequence relation, thereby confirming the existence of standard lore on this issue. 6 Cf. on the first point Künne (1997), p. 74 and (2003) pp. 186–9. On the second point, Batóg (1995). On the third point, Betti (2005c) in which I suggest that non-Brentanian stances in the analysis of truth-bearers and their ontological counterparts in the early Leśniewski, Łukasiewicz and Kotarbiński come, via Twardowski, from Bolzano. Other Polish works dealing with Bolzano or mentioning him are: Biegański (1912),

ON THE BOLZANIZATION OF POLISH THOUGHT

73

Fränklówna (1914), Smolka (1927), Kotarbiński (1929), Wiegner (1930), as reported by Batóg (1995), and most of all Jan Franciszek Drewnowski’s doctoral thesis under Kotarbiński, Podstawy logiki Bernarda Bolzana (1927), entirely devoted to Bolzano – but destroyed during the Second World War. A Polish translation of Bolzano’s Paradoxes appeared, with Kotarbiński’s preface, in 1966. On Fränklówna (1914) see Künne (2003), pp. 184–5. 7 On Twardowski’s criticism of Brentano’s theory of judgement, see Betti & van der Schaar (2004). On Meinong’s see Rollinger (2005). On Meinong’s and Husserl’s, Rollinger (2004), especially pp. 270 and ff. 8 The meaning of expressions like ‘Brentanism’ or ‘Brentanian tradition’ when more is meant than a genealogical line of pupils and students starting from Brentano, seems to me already rather difficult to define in a satisfactory way. Though I will use freely these and similar expressions in this paper, I am aware that this issue is more relevant to the main thesis of this paper than I will be able to account for here. 9 And pace Künne (1997), pp. 41 and ff., (2003), p. 180; cf. Husserl (1896). By “twofold notion of the object of a (subjective) Vorstellung”, I mean that there is a sense in which all Bolzanian subjective presentations (Twardowski’s presentations) can be taken to have an object, this object being a presentation-in-itself, and being what is thinkable in a subjective presentation, cf. Bolzano WL §99 (I 461), §67 (I 305) and cf. also Bolzano-Exner p. 11 ll. 26–34. Although it’s clear that for Bolzano the object of a subjective presentation is the object of the presentation-in-itself contained in the subjective presentation, cf. Bolzano-Exner p. 26 ll. 19–22 and 69 ll. 31–4, the point is what stance one takes on the relation between object and thinkable. 10 Cf. also Künne (1997), p. 56. 11 For the record, in Zur Lehre Twardowski quotes, in this order, §289, §49, §67, §89, §103, §108, then again §49, §65, §112, §58, §90, §64, §63, §68 of Bolzano’s Wissenschaftslehre. Bolzano’s name, with approximately thirty occurrences, is the most cited in this work. 12 For Bolzano and the early Leśniewski cf. Betti (1998b), (2005a), (2005c). 13 Leśniewski (1927), p. 196, fn 1. Eng. Transl. p. 214, fn 19, reproduced here with changes. Translations from Polish are mine unless otherwise indicated. 14 Cf. Kerry (1885–1891), VIII, pp. 135–6. The translation from German is mine. The passage is also quoted in Künne (1997), p. 34. 15 Cf. Twardowski (1926), p. 10. 16 Cf. Twardowski (1911), p. 30. Eng. Trans. p. 131. 17 Cf. Sebestik (2003), p. 62. 18 Cf. Łukasiewicz (1949–54), p. 38 verso. 19 Łukasiewicz (1949–54), p. 57. The passage quoted also in Jadczak (1997), p. 46–7. 20 Göran Sundholm has suggested this to me. Actually, this might explain how Łukasiewicz could suggest that the philosophical achievements of the Vienna Circle could be superior to the ones of the Lvov-Warsaw School: he had clearly denied this

74

ARIANNA BETTI

superiority in 1936, at least concerning the Viennese anti-metaphysical stand. Cf. Łukasiewicz (1936). 21 Cf. in Husserl (1900/01), the Anhang to the Prolegomena. 22 By ‘psychologism’ in this paper exclusively psychologism in logic is meant, that is, in all generality the view that logic ought to be based on psychology. 23 Łukasiewicz (1949–54), p. 57 v. 24 Cf. Łukasiewicz (1904a), p. 341. 25 Cf. Łukasiewicz (1905), p. 469. 26 Łukasiewicz (1949–54), p. 57 v. There are 45 years of difference between the quotations I use in this section, but we do not have reasons to think Łukasiewicz changed his mind in this respect. 27 The kind of accusation put forward by Łukasiewicz was so common that (in the third edition of his work) Husserl referred to it, as a groteske Vorwurf. Cf. Husserl (HUA), B2 III. Cf. also Bell (1990), p. 86. 28 Cf. Łukasiewicz (1904a) and the replies contained in it to an anonymous speaker. Some, including myself, have taken the anonymous speaker to be Twardowski (cf. also Rojszczak (1996), p. 140). This does not seem entirely sure, however. Łukasiewicz (1904a) is a report of the Husserl lecture, which was split in two meetings of the Polish Philosophical Society in Lvov, on 11 and 25 May 1904. According to Jadczak (1997), p. 43, in that period Twardowski spent several months abroad, and in a letter dated 17 May 1904 (cf. (1904b)) Łukasiewicz reports to Twardowski on the meeting of 11 May. So, Twardowski was not present at the first meeting. He could have been present at the second one, but I have not been able to check this. 29 See also the passage on the next page. 30 Łukasiewicz (1949–54), p. 57 v. Note the last line: “at the borders of logic, psychology and grammar” was the subtitle of Twardowski (1911). The passage is quoted also in Ingarden (2000), p. 65 and in Jadczak (1997), p. 47. 31 Cf. in Łukasiewicz (1910), p. 28 the footnote against Höfler & Meinong (1890). 32 Cf. Łukasiewicz (1904c), also quoted in Jadczak (1997), pp. 43–4. 33 Cf. Łukasiewicz (1909). 34 Ibidem. 35 Cf. Łukasiewicz (1910), p. 104. In the letter of 19 November 1908 from Graz Łukasiewicz tells Twardowski that he could not reach an agreement with Meinong regarding evidence. The relevant fragment is also quoted in Jadczak (1997), pp. 46–7. 36 Cf. Twardowski (1900). 37 Twardowski (1908/09). Twardowski’s turn could be dated to ca. 1907. Support comes from Rojszczak (1996, p. 140), who quotes as significant the fact that in Łukasiewicz (1907) there was no discussion similar to the one mentioned above regarding Łukasiewicz’s 1904 lecture (but see fn 28 above). An opponent of my thesis might argue that Twardowski himself in his Selbstdarstellung (cf. Twardowski (1926), pp. 19–20) states to have been convinced by Husserl’s Logische Untersuchungen to aban-

ON THE BOLZANIZATION OF POLISH THOUGHT

75

don psychologism, and does not mention Łukasiewicz. To this I would answer that in the passages in question Twardowski aims at portraying himself as an early follower of the antipsychologistic trend. All other evidence points however in the direction on a later adoption of this stance. 38 Cf. Twardowski (1911), p. 31; Eng. trans. p. 132. 39 I juxtapose here two passages: the piece before ‘[...]’ is from Chapter I of Leśniewski (1927), p. 169, that thereafter is from Chapter II of the same work, p. 182– 3. The latter, rather famous, passage is easier to interpret in the context of the theme of this section. It has been read as a pure and simple recantation of Leśniewski’s early writings, but the difficulty with this is that there is continuity in Leśniewski’s oeuvre. When the passage is placed in the context of the theme of this paper it becomes clearer that the recantation is less about Leśniewski’s philosophical convictions in those works than about his research tools at the time. I reproduced the English translation (p. 181 and 197–8) with changes. 40 Łukasiewicz (1949–54), pp. 59–60. Despite what other sources report, it is uncertain whether Twardowski dealt with Bolzano in that course. 41 Cf. Łukasiewicz (1949/54), p. 60. I think Woleński is absolutely right in saying that the level of Łukasiewicz’s teaching at the time probably did not exceed the level of Couturat’s L’Algèbre de la Logique (1905), on which the Appendix to Łukasiewicz (1910) was based. Cf. Woleński (1999a), p. 65. 42 Łukasiewicz (1949/54), 61r. See also Jadczak (1997), p. 49. 43 Cf. Twardowski (1921/22). See also Jadczak (1997), p. 49. Twardowski’s attitude in this piece would deserve extensive treatment in connection with the changes in the conception of logic in Poland in those years, rather than the few scattered remarks I offer here. 44 Cf. Łukasiewicz (1913), §24. Despite different accounts (Simons (1989)), Łukasiewicz’s own theory of probability seems to have been influenced not by Meinong, but by Bolzano. True, Meinong’s ideas on probability were mentioned by Stumpf in a footnote of his 1892 work on probability which Łukasiewicz also quotes, where Bolzano is completely absent, and, moreover, Łukasiewicz spent some time with Stumpf in Berlin in 1905. Yet Łukasiewicz himself says explicitly that “ ideas were not after Meinong” cf. Łukasiewicz (1949–54), pp. 61r, 61v. The influence of Bolzano seems to be evident when we examine the notion of undetermined sentence (unbestimmte Aussage), cf. also Childers & Majer (1998), Niiniluoto (1998) and Künne (2003), p. 180–4. 45 The first time Bolzano’s name appears in Łukasiewicz’s works – next to Cantor’s and Dedekind’s – is as early as 1905, in Łukasiewicz’s lecture on the concept of infinity. This is linked to the Paradoxes of the Infinite, but not to the Wissenschaftslehre (as far as I know, Twardowski never dealt with the Paradoxes). 46 Cf. Kotarbiński (1966), p. 157.

76

ARIANNA BETTI

47

For Marty’s and Bolzano’s influence on Leśniewski see Betti (1998a), p. 95, 99– 101 and (1998b), in particular p. 131. 48 Cf. Betti (2005d). 49 Cf. the list of courses in Avé-Lallemant (1982), p. 373. Pfänder used Husserl’s Investigations in his exercises in logic. 50 As one would expect, this seems to have happened before 1915, cf. Głombik (1999), p. 98 and fn. 51 On Leśniewski and contradiction see my (2005b). 52 Chwistek (1912), p. 16. 53 Leśniewski was to be hardly convinced at first that Gödel’s incompleteness theorem was correct. 54 The solution for the rest is quite different, cf. Betti (2004). 55 Cf. Łukasiewicz (1915), pp. 29–30. His formulation of the liar sentence is famous because Tarski took it over in 1933 in his epoch-making work on truth. 56 Simons (1987), pp. 15. 57 Added in proof. Twardowski’s home library housed at present at the Institute of Philosophy and Sociology of Warsaw University includes an annotated copy of the Wissenschaftslehre from 1837. 58 This is not to say that a parallel and thorough investigation of Frege’s influence on the development of Polish logic would be a less urgent matter, quite the contrary. For a first recognition see Woleński (2004).

References Avé-Lallemant, Eberhard 1982. ‘Nachlaß-Uebersicht’, in H. Spiegelberg & E. AvéLallemant (eds.), Pfänder-Studien, The Hague/Boston/London, M.Nijhoff, pp. 371–379. Batóg, Tadeusz 1995. ‘Ajdukiewicz and the Development of Formal Logic’, in V. Sinisi & J. Woleński (eds.), The Heritage of Kazimierz Ajdukiewicz, Amsterdam/Atlanta, Rodopi, pp. 53–67. Bell, David 1990. Husserl, London & New York, Routledge. Betti, Arianna 1998a. ‘Il rasoio di Leśniewski’, Rivista di Filosofia LXXIX, pp. 89– 117. Betti, Arianna 1998b. ‘De Veritate: another Chapter – The Bolzano Leśniewski Connection’, in K. Kijania-Placek & J.Woleński (eds.), The Lvov-Warsaw School and Contemporary Philosophy, Dordrecht, Kluwer Academic Publishers, pp.115–137. Betti, Arianna 2002. ‘The Incomplete Story of Łukasiewicz and Bivalence’, in T. Childers (ed.), The Logica 200 Yearbook, Prague, The Czech Academy of Sciences – Filosofia, pp. 21–36.

ON THE BOLZANIZATION OF POLISH THOUGHT

77

Betti, Arianna 2004. ‘Leśniewski’s Early Liar, Tarski and Natural Language’, Annals of Pure and Applied Logic 127, pp. 267–287. Betti, Arianna 2005a. ‘Sempiternal Truth. The Bolzano Leśniewski Twardowski Axis’, in J. J. Jadacki & J. Paśniczek (eds.), The Lvov-Warsaw School. The Second Generation, Amsterdam, Rodopi (Poznań Studies in the Philosophy of the Sciences and the Humanities), forthcoming. Betti, Arianna 2005b. ‘Łukasiewicz and Leśniewski on Contradiction’, Reports on Philosophy 22/2004, pp. 247–271. Betti, Arianna 2005c. ‘A Note on Early Polish Semantics, Bolzano and the Woleński Thesis’, in J. Hartman (ed.), Logics and Philosophy. Towards Jan Woleński, Amsterdam, Rodopi, forthcoming. Betti, Arianna 2005d. ‘Logic as Universal Medium? Leśniewski’s Systems and the Aristotelian Model of Science’, to appear in S. Lapointe and M. Marion (eds.), Logic, Ontology, and Aesthetics in Poland, Berlin, Springer. Betti, Arianna & Van der Schaar, Maria 2004. ‘The Road from Vienna to Lvov: Twardowski’s Theory of Judgement 1894–1897’, Grazer Philosophische Studien 67, pp. 1–20. Biegański, Władysław 1912. Teorya Logiki, Warszawa, Skład główny w księgarni E. Wende i spółki. Bolzano, Bernard WL, Wissenschaftslehre, hrsg. von J. Berg Bernard Bolzano-Gesamtausgabe Reihe I, Bande 11–14 (1985–2000). Bolzano-Exner, E. Winter (ed.) 1935. Spisy Bernarda Bolzana, Korespondence B. Bolzana s F. Exnerem / Bernard Bolzano’s Schriften, Band 4 – Der Briefwechsel B. Bolzanos mit F. Exner, Praha / Prag, Königliche böhmischen Gesellschaft der Wissenschaften / Královská česká společnost nauk. Childers, Timothy & Majer, Ondrej 1998. ‘On Łukasiewicz’s Theory of Probability’, in K. Kijania-Placek & J. Woleński (eds.), The Lvov-Warsaw School and Contemporary Philosophy, Dordrecht, Kluwer Academic Publishers, pp. 303–312. Chwistek, Leon 1912. ‘Zasada sprzeczności w świetle nowszych badań Bertranda Russella’, Rozprawy Akademii Umiejętności, Wydzial Historyczno-Filozoficzny 2, vol. LV, series 30, Kraków, 1912, pp. 270–334. Fränklówna, Maria 1914. ‘O pewnym paradoksie w logice Bolzana’, Przegląd Filozoficzny 17, pp. 315–323. Głombik, Czesław 1999. ‘O niedoszłych polskich przekładach Logische Untersuchungen’, in Polska filozofia analityczna w kręgu Szkoły Lwowsko Warszawskiej, Toruń, Wydawnictwo Uniwersytetu Mikołaja Kopernika, pp. 89–106. Höfler, Alois & Meinong, Alexius (1890) Philosophische Propädeutik – I Theil Logik, Prag/Wien/Leipzig,Tempsky/Freytag. Husserl, Edmund 1896. ‘ K. Twardowski, Zur Lehre vom Inhalt und Gegenstand der Vorstellungen – Eine psychologische Untersuchung, Wien

78

ARIANNA BETTI

1894’, in Aufsätze und Rezensionen (1890–1910), Husserliana XXII, M. Nijhoff publ., The Hague/Boston/London, 1979, pp. 357–80. Husserl, Edmund 1900/01. Logische Untersuchungen, Max Niemeyer: Halle a.d.S., 1900–1901; Husserliana XVIII, XIX/1,XIX/2, M. Nijhoff publ., The Hague / Boston / London. Ingarden, Roman 1938. ‘Działalność naukowa Twardowskiego’ in Kazimierz Twardowski. Nauczyciel-Uczony-Obywatel, Lwów, Nakładem Polskiego Towarzystwa Filozoficznego – Skład główny w księgarni “Książka” (A. Mazzuccato), pp. 13– 30. Ingarden, Roman Stanisław 2000. Roman Ingarden – Życie filozofa w okresie toruńskim (1921–1926), Toruń, Uniwersytet Mikołaja Kopernika. Jadczak, Ryszard 1997. Mistrz i jego uczniowe, Warszawa, SCHOLAR. Kerry, Benno 1885–1891. ‘Über Anschauung und ihre Psychische Verarbeitung’, Vierteljahrsschrift für wissenschaftliche Philosophie, IX,1885, pp. 433–493 (I); X, 1886, pp. 419–467 (II); XI, 1887, pp. 53–116 (III); 4, XI, 1887, pp. 249–307 (IV); 5 XIII 1889 pp. 71–124 (V), pp. 392–419 (VI); XIV, 1890, pp. 317–353 (VII); XV, 1891, pp. 127–167 (VIII). Kotarbiński, Tadeusz 1913. ‘Zagadnienie istnienia przyszłości’, Przegląd filozoficzny XVI, pp. 74–92. Kotarbiński, Tadeusz 1929. Elementy teorii poznania, logiki formalnej i metodologii nauk, Lvov, Ossolineum, 1929; Eng. Transl. Gnosiology. The Scientific Approach to Theory of Knowledge, Oxford, Pergamon Press, 1966. Kotarbiński, Tadeusz 1966. ‘Garstka wspomnień o Stanisławie Leśniewskim’, Ruch Filozoficzny n. 3–4, 1966, pp. 155–63. Künne, Wolfgang 1997. ‘Die Ernte wird erscheinen – Die Geschichte der BolzanoRezeption (1849-1939)’, in Ganthaler, H. & Neumaier O. (eds.) Bolzano und die österreichische Geistesgeschichte, Sankt Augustin, Academia Verlag, pp. 9–82. Künne, Wolfgang 2003. ‘Bernard Bolzano’s Wissenschaftslehre and Polish Analytical Philosophy Between 1894 and 1935’ in: J. Hintikka, T. Czarnecki, K. KijaniaPlacek, T. Placek, A. Rojszczak (eds.) Philosophy and Logic – In Search of the Polish Tradition – Essays in honour of Jan Woleński on the Occasion of his 60th Birthday, Dordrecht, Kluwer Academic Publishers, pp. 179–192. Leśniewski, Stanisław 1927. O Podstawach Matematyki (On the Foundations of Mathematics), Chs. I–III, Przegląd Filozoficzny XXX, pp. 164–206, 1927. Eng. Trans. in Stanisław Leśniewski, Collected Works, 1 vol., ed. by S. J. Surma, J. T. Srzednicki, D. I. Barnett e V. F. Rickey, Dordrecht, Kluwer Academic Publishers, 1992, pp. 181–382. Łukasiewicz, Jan 1904a. ‘Teza Husserla o stosunku logiki do psychologii’, Przegląd Filozoficzny 7, 1904, pp. 476–7. Page numbers refer to the reprint in Łukasiewicz (1998), pp. 341–2.

ON THE BOLZANIZATION OF POLISH THOUGHT

79

Łukasiewicz, Jan 1904b. Letter to Kazimierz Twardowski (Lvov, 17.5.1905), in Łukasiewicz (1998), pp. 459–60. Łukasiewicz, Jan 1904c. Letter to Kazimierz Twardowski (Charlottenburg, 12.12.1904), in Łukasiewicz (1998), pp. 467–8. Łukasiewicz, Jan 1905. Letter to Kazimierz Twardowski (Charlottenburg, 6.2.1905), in Łukasiewicz (1998), pp. 468–71. Łukasiewicz, Jan 1906. ‘Analiza i konstrukcja pojęcia przyczyny’, Przegląd Filozoficzny 9, pp. 105–179. Page numbers refer to the reprint in Łukasiewicz (1961), pp. 9–62. Łukasiewicz, Jan 1907. ‘Logika a psychologia’, Przegląd Filozoficzny 10, 1907, pp. 489–91. Page numbers refer to the reprint in Łukasiewicz (1961), pp. 63–5. Łukasiewicz, Jan 1909. ‘O poglądach filozoficznych Meinonga’, Przegląd Filozoficzny 12, p. 559. Page numbers refer to the reprint in Łukasiewicz (1998), pp. 344–5. Łukasiewicz, Jan 1910. O zasadzie sprzeczności u Arystotelesa – Studium Krytyczne, Kraków, Polska Akademia Umiejętności. Łukasiewicz, Jan 1913. Die logische Grundlagen der Wahrscheinlichkeitsrechnung, Krakau, Akademie der Wissenschaften. Łukasiewicz, Jan 1915. ‘O nauce’, in Poradnik dla samouków (wyd. nowe), Tom 1, Warszawa, Wydawnictwo A. Heflicha i S. Michalskiego, pp. XV–XXXIX. Page numbers refer to the reprint in Łukasiewicz (1998), pp. 9–32. Łukasiewicz, Jan 1936. ‘Logistyka a filozofia’ in Łukasiewicz (1961), pp. 195–209. Łukasiewicz, Jan 1949/54. Pamiętnik (Memoirs), unpublished typescript, housed by the library of the Institute of the Philosophy and Sociology of the University of Warsaw. Łukasiewicz, Jan 1961. Z zagadnień logiki i filozofii. Pisma wybrane, Warszawa, PWN. Łukasiewicz, Jan 1998. Łukasiewicz – Logika i metafizyka, ed. by J. J. Jadacki, Warszawa, Wydział Filozofii i Socjologii Uniwersytetu Warszawskiego. Meinong, Alexius 1907. Über die Stellung der Gegendstandtheorie im System der Wissenschaften, Leipzig, Voigtländer, 1907. Repr. in Alexius Meinong Gesaumtausgabe V, R. M. Chisholm ed., Graz, Akademische Druck- u. Verlagsanstalt, 1973. Niiniluoto, Ilkka 1998. ‘Induction and probability in the Lvov-Warsaw School’, in K. Kijania-Placek & J. Woleński (eds.), The Lvov-Warsaw School and Contemporary Philosophy, Dordrecht, Kluwer Academic Publishers, pp. 323–335. Rojszczak, Artur 1997. Vom Urteil zum Satz. Das Wahrheitsträgerproblem von Bolzano zu Tarski, PhD thesis, Jagiellonian University, Krakow, 1997. Rollinger, Robin D. 2004. ‘Austrian Theories of Judgement’ in A. Chrudzimski & W. Huemer, Phenomenology and Analysis. Essays on Central European Philosophy, Frankfurt/Lancaster, Ontos Verlag, pp. 257–84.

80

ARIANNA BETTI

Rollinger, Robin D. 2005. ‘Meinong and Brentano’ Meinong Studies 1, pp. 159–197. Sebestik, Jan 2003. ‘Husserl Reader of Bolzano’ in D. Fisette (ed.), Husserl’s Logische Untersuchungen Reconsidered, Dordrecht, Kluwer Academic Publishers, 2003. Siebel, Mark 1996. Der Begriff der Ableitbarkeit bei Bolzano, Sankt Augustin, Academia Verlag. Simons, Peter 1987. ‘Meinong, Łukasiewicz and Many-Valued Logics ‘, in id., Philosophy and Logic in Central Europe from Bolzano to Tarski, Dordrecht, Kluwer Academic Publishers, 1992. Simons, Peter 1989. ‘Bolzano, Tarski and the Limits of Logic’, in id., Philosophy and Logic in Central Europe from Bolzano to Tarski, Dordrecht, Kluwer Academic Publishers, 1992. Smolka, Fr. 1927. ‘Pojęcie zmiany u Bolzana i u Russella’, Przegląd Filozoficzny 30, p. 291. Twardowski, Kazimierz 1894. Zur Lehre vom Inhalt und Gegenstand der Vorstellungen – Eine psychologische Untersuchung, Wien, Hölder (anast. repr., MünchenWien, Philosophia Verlag, 1982). Twardowski, Kazimierz 1894/5. Logik – Wintersemester 1894/5, manuscript containing the lecture notes in logic given in the A.A. 1894/5 at Vienna University, pp. 272. Edition by Arianna Betti and Venanzio Raspa in progress. Twardowski, Kazimierz 1895/6. Logika – Kurs zimowy 1895/6, manuscript containing the lecture notes in logic given in the A.A. 1895/6 at Lvov University, pp. 264. Edition by Arianna Betti in progress. Partial English (working) translation on the Polish Philosophy Page at http://www.fmag.unict.it/PolPhil/Tward/TwardLog.html. Twardowski, Kazimierz 1900. ‘O tak zwanych prawdach względnych’, Księga pamiątkowa Uniwersytetu Lwowskiego ku uczczeniu pięćsetnej rocznicy fundacji Jagiellońskiego Uniwersytetu Krakowskiego, Lwów, Nakładem Senatu Akademickiego Uniwersytetu Lwowskiego. Eng. trans. in On Actions, Products and Other Topics in philosophy, Johannes Brandl and Jan Woleński (eds.), Amsterdam, Rodopi, 1999, pp. 147–169. Twardowski, Kazimierz 1908/9. Psychologia myślenia, lecture notes, Archiwum Kazimierza Twardowskiego, Biblioteka IFiS PAN, Warsaw, pp. 31. Twardowski, Kazimierz 1911. ‘O czynnościach i wytworach – Kilka uwag z pogranicza psychologii, gramatyki i logiki’, Gubrynowicz i syn, Kraków, 1911, pp. 33 (extract from Księga pamiątkowa ku uczczeniu 250-tej rocznicy założenia Uniwersytetu Lwowskiego przez króla Jana Kazimierza vol. II, Lwów, Nakładem Uniwersytetu Lwowskiego, 1912) pp. 33. Eng. trans. in On Actions, Products and Other Topics in philosophy, Johannes Brandl and Jan Woleński (eds.), Amsterdam, Rodopi, 1999, pp. 103–132.

ON THE BOLZANIZATION OF POLISH THOUGHT

81

Twardowski, Kazimierz 1921/22. ‘Symbolomania i pragmatofobia’, Ruch filozoficzny VI, pp. 1–10. Twardowski, Kazimierz 1926. ‘Kazimierz Twardowski: Selbstdarstellung’, hrsg. Jan Woleński & Thomas Binder, Grazer Philosophische Studien 39, 1991, pp. 1–24. Wiegner, Adam 1930. W sprawie zasady odwrotności między treścią a zakresem pojęć, Poznań, Nakładem Poznańskiego Towarzystwa Przyjaciół Nauk z zasiłku Ministerstwa Wyznań Religijnych i Oświecenia Publicznego. Woleński, Jan 1999a. ‘Mathematical Logic in Poland 1900–1939 People, Circles, Institutions, Ideas’, in id., Essays in the History of Logic and Logical Philosophy, Jagiellonian University Press, Dialogikon, pp. 59–84. Woleński, Jan 1999b. ‘Theories of Truth in Austrian Philosophy’ in id., Essays in the History of Logic and Logical Philosophy, Kraków, Jagiellonian University Press, Dialogikon, pp. 150–175. Woleński, Jan 2004. ‘The reception of Frege in Poland’, History and Philosophy of Logic 25, pp. 37–51. Woleński, Jan & Simons, Peter 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, Dordrecht, Kluwer Academic Publishers, pp. 391–442.

Things and Truths: Brentano and Leśniewski, Ontology and Logic PETER SIMONS

When Stanisław Leśniewski went to Lwów in 1910 to study with Kazimierz Twardowski, he was already acquainted with the philosophy of what he later called “the Austrian School”, by which he meant Franz Brentano and his students, especially Anton Marty and Edmund Husserl, whose works initially captivated Leśniewski. This knowledge had been acquired during Leśniewski’s philosophical education in Germany, Russia and Switzerland, but it is very likely that his doctoral supervisor Twardowski would have made further aspects of Brentano’s work known to him. Despite this, and despite the often remarkable convergences between the philosophical views of Brentano and Leśniewski, there is little evidence of direct influence of the former on the latter. We have the testimony of Leśniewski’s friend and colleague Tadeusz Kotarbiński that the latter’s even closer philosophical parallels to Brentano’s work were acquired independently.1 Leśniewski did interact with Brentano’s views, but only at the very beginning of his career, and unsuccessfully. Later parallels are just that: parallels, and we are unlikely to come upon significant new evidence as to whether there was or was not any direct influence, positive or negative. Brentano and his philosophy were part of the philosophical wallpaper in Twardowski’s Lwów, but Leśniewski’s interests soon came to settle on logic and the foundations of mathematics, and such figures as Russell, Frege, Cantor and Zermelo quickly outranked Brentano in their importance for his views. Nevertheless, the parallels are real, interesting, and revealing.

1 Existential Propositions Reviled: Leśniewski’s Early Criticism of Brentano Leśniewski’s first published paper is called ‘Przyczynek do analizy zdań egzystencjalnych’, ‘Contributions to the Analysis of Existential ProposiActions, Products, and Things. Brentano and Polish Philosophy. Arkadiusz Chrudzimski and Dariusz Łukasiewicz (eds.), Frankfurt: ontos, 2006, 83–105.

84

PETER SIMONS

tions’. It appeared in Przegląd Filozoficzny in 1913, and contains the gist of his doctoral dissertation. It includes a criticism of Brentano’s existential analysis of propositions. The work seems to have impressed Twardowski enough for the latter to have urged Leśniewski to publish it as a paper before the dissertation appeared. If so, one has to question Twardowski’s judgement. It is commonplace for connoisseurs of Leśniewski to praise the paper’s clear distinction between object language and metalanguage. While this distinction is indeed present, it is there by instinct and is not made into a leading methodological point, which is one of several reasons why Tarski tends to get the credit for his teacher’s innovation, despite Tarski’s scrupulous attribution of priority to Leśniewski. It is much less commonly observed that the paper has two glaring defects: it is amazingly and ineptly pedantic in style, and it seriously defends one of the silliest theses ever put forward in twentieth century philosophy. As to the style, the paper’s main structure is that of a numbered series of mostly short pithy statements, each followed by strings of much longer “remarks” which amplify the short statements. The first such statement is about what the author understands by ‘existential proposition’: it is five lines long, whereas its ensuing remarks are 43 lines long. Section 3, ‘All positive existential propositions are analytic’ is a page long, followed by six pages of “remarks”. It is several pages before the wood begins to become visible for the trees. Rarely can a major philosopher-logician’s literary career have had a less auspicious inception. Not that Leśniewski ever lost his instinctive pedantry, and nor is pedantry wholly reprehensible: it is one of the things which make him a great logician. But pedantry in details of logic is one thing, pedantry in presentation and literary style is another. They need not go together, as the case of Frege testifies. Frege is one of several counterexamples to those who claim that Germans cannot write fine philosophical prose, but until Leśniewski came along there was no one more careful than Frege about the apparently trivial but in fact important “technical” details of his logical systems, such as how to formulate rules of substitution, or what conditions a definition has to satisfy. That indeed was the reason why Leśniewski came to adulate Frege, and quite rightly so. However, whereas Twardowski wrote with limpid simplicity, Łukasiewicz with flair and elegance, Kotarbiński with old-fashioned clarity, and Ajdukiewicz with dogged determination, Leśniewski’s prose style fluctu-

THINGS AND TRUTHS

85

ates throughout his career between wild emotionalism in some philosophical passages and rigid schoolmasterly punctiliousness in other places, especially in any commentary on technical matters. The result is an excellent cure for insomnia and an often unappetising way to get to know his – after all – important ideas. As to content, the paper puts forward, with, one might say, a wholly straight logical face, the amazing thesis that any positive existential proposition is analytically and hence necessarily true, and contrapositively that any negative existential is self-contradictory and hence necessarily false. This is based on a subject-predicate analysis of propositions as follows. A proposition is always properly expressed in subject-predicate form with positive copula, which may be singular or plural.2 Such a proposition of the form ‘A is a b/are bs’ says that every property connoted by the predicate term ‘b’ is one of the properties possessed by every object falling under the subject term ‘A’;3 it is true if so, and false if some object falling under the subject term does not have one of the properties connoted by the predicate term. If every property connoted by the predicate term is further connoted by the subject term, the proposition is analytic, whereas if some property connoted by the predicate term is not among those connoted by the subject term, the proposition is synthetic. Leśniewski considers ‘existent’ or ‘being’ to connote either nothing or at most plurality, so that any positive existential claim such as ‘A is existent’ or ‘A are beings’ is analytically true, since no property connoted by the predicate is not connoted by the subject, since all terms connote being. The only property connoted by a negative existential is the property of non-existence, and this contradicts any subject term, which connotes being. Hence ‘Unicorns exist’ is necessarily true and ‘Square circles are non-existent’ is necessarily false. According to Leśniewski, the former means ‘Existing objects which are unicorns are existing’ while the latter means ‘Existent square circles are non-existent’, from which we can see why Leśniewski takes the propositions to have the truth-values he ascribes to them. Before criticising this view let us note a parallel. The historically perceptive may be already thinking of Alexius Meinong. While Leśniewski critically discusses Meinong’s contradictory objects in a later paper,4 in this one he appears not to be aware of it. When Russell mentioned Meinong’s views in ‘On Denoting’ only a few years before, he noted that it

86

PETER SIMONS

appears natural to interpret Meinong’s views as committing one to the view that it is self-contradictory to deny the existence of anything.5 But note: not only does Russell think this view is patently absurd, so, as a matter of fact, does Meinong, and the Meinong to whom Russell ascribes this view is a straw man of his own concoction.6 As far as I can see only Leśniewski seriously held the absurd view that Russell wrongly attributes to Meinong. Leśniewski notes that it is a position of “some exponents of Franz Brentano’s so-called ‘Austrian school’”7 to the effect that every proposition is reducible to a positive or negative existential proposition. Brentano’s views were well known from his exposition of them in the Psychologie vom empirischen Standpunkt. There Brentano criticised Mill’s insistence that all propositions have the subject-predicate form. In effect, Leśniewski is reinstating Mill’s view, and he rejects the Brentanian analysis as wrong. According to Leśniewski, the Brentanian view – and indeed the common view – that ‘Unicorns exist’ is false and ‘Square circles do not exist’ is true arise from inadequate verbal expression of the contents in question. According to Leśniewski the proper way to express the intended content of the false proposition is ‘Some beings are unicorns’ while the true proposition should be expressed as ‘No being is a square circle’.8 While it is plausible that these formulations say what is intended, or at least something logically equivalent to what is intended, they are also in the mouth of the average speaker equivalent to the existential formulations that Leśniewski castigates. The only reason for the non-equivalence is Leśniewski’s legislation that only the subject-predicate analysis of a certain form is “adequate”. There is no external reason to accept such an arbitrary norm and an excellent reason not to do so, namely that it is a distortion of actual usage. Also the sudden introduction of the words ‘no’ and ‘some’, which had been conspicuously absent from previous critical discussion, is neither commented on, nor is their logical significance discussed by Leśniewski. It is a clear and blatant case of double standards being applied. If we treat ‘No A is a b’ as a subject–predicate proposition in Leśniewski’s style, we get the counterintuitive consequence that ‘No cat is a mammal’ is an analytically true proposition. So either it cannot be a subject-predicate proposition of the kind Leśniewski uses to define truth and analyticity, and so some propositions have other forms, or it has the

THINGS AND TRUTHS

87

counterintuitive truth-value and cannot be used as the “adequate” expression of the content it seems to disclose. Leśniewski cannot have it both ways. If ‘No’ propositions are not subject-predicate ones, despite there being a subject and a predicate term, there is no barrier to accepting propositions with only one term, such as ‘Nothing is an A’ or ‘There is an A’, which are just the forms Brentano invokes in his reduction strategy. Finally, the contingently false proposition ‘Nothing exists’ would have to be rendered by Leśniewski as ‘No being is a being’, which according to some intuitions sounds as though it should be necessarily false, namely when the subject term has existential import, though with sensible reading can indeed be interpreted as equivalent to ‘Nothing is a being’, i.e., ‘Nothing exists’, or ‘There is no being’, which brings us once again full circle to Brentano’s suggestion. It is indeed either disingenuous or a glaring oversight of Leśniewski not to mention that ‘No A is an A’, which he could accept, is equivalent to ‘There is no A’, which he does not.

2 Existential Propositions Revived The paper on existential propositions is thus, apart from some of its incidental features, a complete mess, and in no way constitutes a refutation of Brentano’s views on the logical adequacy of positive and negative existentials. Whether this is correct or not,9 Leśniewski does not refute it. It is then perhaps not surprising that Leśniewski later dropped and indeed repudiated his bizarre views, and his later papers are logically much more sober and reasonable. He dropped the notions of connotation and property, and with them the distinction between analytic and synthetic propositions. This is I think in some ways a loss, but I will not return to the issue. Truth of a subject-predicate proposition is thenceforth defined purely in terms of denotation. A singular subject-predicate proposition ‘A is (a) b’ is true if ‘A’ names a single object and ‘b’ names one or more objects of which A is one. A non-overtly quantified general inclusion ‘As are bs’ is true iff every individual which is an A is also a b, the notions of ‘no’ and ‘some’ are adequately defined, and finally existential propositions, positive or negative, are propositions with only one nominal term, not two.

88

PETER SIMONS

In his ‘ontology’, the logical system of names, predicates and higherorder functors which he began axiomatizing in 1920, Leśniewski based the system on a singular copula ‘is (a)’ with existential import. He could then define the existence functor using quantification as There are b iff for some A, A is a b However, Brentano used nominal conjunction and nominal negation to define the categorical propositions. All b are c iff there are no b non-c No b are c iff there are no bc Some b are c iff there are bc Some b are not c iff there are b non-c In Leśniewski, nominal conjunction and nominal negation are definable as follows A is a bc iff A is A and A is a b and A is a c A is non-b iff A is (an) A and it is not the case that A is a b To ensure that Brentano’s and Leśniewski’s respective primitives are interdefinable we need to define the uniqueness that characterizes the singular copula’s subject. This turns out not to be definable in the elementary logic Brentano uses, but to require term-binding quantification:10 There is at most one b iff for all c and d: if there is a bc and there is a bd then there is a cd so Leśniewski’s primitive is definable in terms of Brentano’s primitives using quantification: A is b iff there is an A and there is at most one A and all A are b.

THINGS AND TRUTHS

89

3 Reism Notwithstanding the near equivalence of Leśniewski’s and Brentano’s views on elementary logic, they start from quite different positions. Leśniewski started from Mill’s analysis of categorical propositions, which Brentano rejected in favour of his existential analysis of propositions. This in turn was based on his theory of judgement, according to which a judgement consists not in a combinatory act of putting ideas together or a decombinatory act of taking them apart, as it was then commonplace to assert, but in a unitary act of acceptance or rejection of an idea, whether simple or compound. The history of subsequent philosophy and phenomenology has tended to confirm Brentano’s view that judgement is a unitary stance, though most commentators have tended to side with Frege’s view that denial consists in the affirmation of a negation. Phenomenologically this is by no means evident, but for logical as distinct from phenomenological purposes the difference is not important. More important is the recognition of Frege, Husserl, Meinong and others that judgement (affirmation or denial) is not the only “propositional attitude”, that assumption, conjecture, doubt, uncertainty, and refraining from a verdict are all possible attitudes, some of which are involved in any attitude to a complex proposition. All of this is of little or no interest to Leśniewski, whose focus was always on the sentence (zdanie) publicly spoken or written rather than on the private mental act of judgement (sąd) preferred by the previous generation.11 Whether for simplicity, or because he was too unsure about psychology, or more probably to avoid the danger of psychologism, Leśniewski left the mental bearers of meaning out of the picture altogether. Most twentieth century philosophy followed suit in one way or another, arguably to its detriment. Because Leśniewski focussed relentlessly on logic and the philosophy of mathematics, the explicit study of language and meaning is not pursued in his writing after the early “grammatical” essays. By contrast, Brentano’s theory of judgement was part of a much more wide-ranging philosophy encompassing mental philosophy and metaphysics as well as practical philosophy and the history of philosophy. All the same, where the two share an interest, as in issues of truth and inference, their views, despite Leśniewski’s radical adherence to mathematical logic and Bren-

90

PETER SIMONS

tano’s equally radical rejection of it, converge remarkably, and nowhere more so than on the issue of reism. Reism is the view, variously anticipated but first propounded clearly by Brentano after about 1904, that the only items that exist are individual things (res). Brentano has an argument for reism. The argument is extraordinarily bad. It goes as follows. To think is to think of something (intentionality of the mental). ‘Think’ is univocal. Therefore ‘think of something’ is univocal. Therefore ‘something’ is univocal. Any thing, such as the dog Snowy, is something. Therefore, since everything is something, everything is a thing. There are holes in this argument at crucial stages. Firstly, ‘to think’ can occur in the construction ‘think that’, as in ‘Benita thought that iron is a conductor’, which is quite different from the ‘think of’ construction, as in ‘Maria thought of her credit card’. Secondly, even if ‘think of something’ is univocal, which is not certain (how can one tell?), it does not follow that everything that can be thought of falls into a single category, because ‘something’ (German etwas) is a catch-all word which connotes no category whatsoever. To a multicategorial ontologist, thinking of the dog Snowy, thinking of Snowy’s white colour,12 thinking of a bark by Snowy, and thinking of the idea Dog are all instances of thinking of something, although the objects of thought are from different categories in each case: a substance, a quality, an event, an idea. A different kind of argument for reism is the ontological economy argument. There are certainly individual things, such as Snowy and Benita. We can think and talk about them, and some of the things we think and say are true, others are false. Let us try and see if we can account for the truth and falsity of every proposition in terms of either the laws of logic or on what individuals there are. A giant leap for mankind was taken in this direction by William of Ockham,13 who claimed that all simple propositions (excepting a few about revealed theology) are true or false in virtue of what substances there are and what qualities (tropes) they have, and that the apparent need for other categories is an illusion fostered by an inadequate appreciation of the nature of language. Reism goes one step further than Ockham in outlawing qualities and claiming that everything we can think or say can be explained in terms of things alone. Reism thus has an ontological dimension, which is easy to state, namely, that there are only things. It also has a semantic dimension, which is hard to state and even

THINGS AND TRUTHS

91

harder to defend, namely that all thought and talk ostensibly about anything other than things is in some way reducible to or explicable in terms of things alone. Now as a matter of fact Leśniewski denied that he was a reist, because he did not know how to explain certain mental phenomena such as afterimages in terms of things. But this is partly because the only things he allows himself are material bodies: the reism of his friend Kotarbiński was a materialist reism of bodies only. Since nothing in reism rules out mental phenomena as being things, this is not a crucial objection to comparing him with Brentano. Brentano qua reist believed in two kinds of substance, namely physical things and mental things or souls. He was a dualist whereas Kotarbiński was a monist. One of the things a reist has to account for is simple qualitative change, like a change of colour or shape. Simply having a wire is not enough to account for the fact that at one time the wire is straight, and later is bent; having a shirt is not enough to account for the fact that at one time it is bright blue and then later is a pale faded grey. Some changes can be accounted for by what things exist at different times, such as there being now three mushrooms on the lawn where before there were none. But with only things – the wire, the shirt – it seems we are not in a position to explain the way the things change which do not involve them coming into or going out of existence. Both Brentano and Leśniewski cope with this problem by expanding the range of what we normally mean by ‘thing’. The differences are instructive. In Brentano’s case things include not only substances like the wire but what he calls ‘accidents’, which are objects containing the substance as a part without containing any supplementary part.14 For example the straight wire is an accident of which the wire itself is a part: it comes into being when – to speak with the vulgar – the wire straightens out, and ceases to exist when the wire becomes bent. The straight wire is replaced as an accident of the wire by a bent wire: what appears to be qualitative change is in fact the replacement of one thing by another, while continuity is provided by the wire which exists throughout and is part in turn of the straight wire and the bent wire. It is similar for the blue shirt which is replaced by the grey shirt. In each case there is in fact not simply a pair of objects but a long sequence of short-lived objects, as the changes are con-

92

PETER SIMONS

tinuous rather than abrupt; one for each shape and each shade of colour. If the thing in question changes simultaneously in more than one way at once, as for instance the wire getting more and more bent while changing temperature, then there is a sequence of shape accidents containing the wire and another sequence of temperature accidents also containing the wire, but neither sort of accident contains one of the other sort: they merely overlap in the wire. Brentano’s view is ingenious but surely mistaken. For one thing, it contradicts the essential signification of the concept part that one thing may have a proper part without having any supplementary part to make up the difference. For another thing, the accidents need to be gerrymandered ad hoc precisely in order to explain the phenomenon in question. If a wire stays straight, why does the straight wire continue to exist as long as the wire stays straight, rather than being replaced part-way through, or at every new instant, by another straight wire? The answer can only be because the wire does not change in shape in that time. What if a wire’s shape changes in synchronization with its temperature: are there two accidents, or one composite one, or all three? Is there a nested sequence of increasingly specific accidents, or is there in each case only one layer of accidents per type of change (quality)? Can there be determinable accidents in addition to determinate ones? Also, relational changes, such as the continuous change in relative position of the sun, earth and moon, seem to call not only for composite substances – the sum of the sun, earth and moon – to be subjects of the change, but also for continuous sequences of instantaneous composite accidents with these subjects as parts. While these imponderables are not sufficient to convict the view of outright contradiction, a simpler and more traditional one lies at hand: the difference between the shirt and the blue shirt lies in a small supplement to the shirt-minus-colour, viz. a colour accident in Aristotle’s sense, that is, a dependent moment or trope which inheres for a while in the shirt. A trope, though it is not a substance, is just as much an individual thing as is a Brentanian accident. Brentano in his early philosophy had maintained a view of tropes as “metaphysical parts” not dissimilar from Aristotle’s accidents, but seen from a more conceptualist point of view. The later view is by contrast fully realist.15 Brentano’s accidents, like his substances, are continuants, that is, they are in time but not extended in time like an event or process. Brentano was

THINGS AND TRUTHS

93

even a presentist, that is, he regarded only present things as real, and past and future things as not real.16 Here is not the place to go into the many grave and conclusive reasons why presentism is false; I simply note and regret Brentano’s view. Leśniewski’s solution to the problem of change was different. Taking the following propositions: Warsaw of 1830 is smaller than Warsaw of 1930 Warsaw of 1830 is Warsaw Warsaw of 1930 is Warsaw simple inference steps bring us to Warsaw of 1930 is smaller than Warsaw of 1930 which is absurd. Whereas the presentist would try to reinterpret or even deny the meaningfulness of the first premiss, Leśniewski prefers to regard the three names as designating different things: Warsaw is the city from its beginning to its end, Warsaw-in-1830 is a time-slice of Warsaw, and Warsaw-in-1930 is a different time slice of Warsaw.17 Leśniewski’s dilemma and his solution almost exactly anticipate the problem posed for the metaphysics of change by David Lewis in 1987:18 the difference is that Leśniewski considers neither presentism (part of Brentano’s way out) nor the theory that qualities are relations to times, which are the two other potential solutions that Lewis (rightly) rejects. In view of his stressing the whole as Warsaw it seems that of two possible perdurantist views, whether the slices are prior (the stage view) or the whole is prior (the worm view)19, Leśniewski is best interpreted as subscribing to the latter. But perhaps, if the question had been put to him, he would have regarded the two views as either both equally true, in that both stages and the whole exist, or equally absurd, in that neither is prior, since there are only stages where there is a worm and a worm is a sum of stages of itself. He would not have acknowledged an intensional relation such as ontological priority. To adjudicate in detail between endurantism and perdurantism would be too much to undertake here. I shall content myself with mentioning my own view, which is that both views are partly right and partly wrong. Ontologically, perdurants are more basic, but by stability and continuity in

94

PETER SIMONS

small and large ways they may serve as the ontological ground of endurant continuants which exist just as much as they do.20

4 A World of Particulars Suppose, as both Brentano and Leśniewski believe, that there are no universals or other abstract objects. What does that tell us about the metaphysics of the world? The answer is that it tells us surprisingly little. The world may be a world of particulars, but that leaves much open, even at the level of categories. Here are some distinctions. Particulars may be individuals or they may be masses of stuff of some sort. That is the count/mass distinction at the ontological rather than linguistic level. They may be endurants or perdurants. They may be independent or dependent. Even this incomplete enumeration of ontological features gives in principle eight different possible kinds of particular if all combinations are permissible. Some combinations are reasonably familiar: independent endurant individuals are substances; perdurant masses are processes; perdurant individuals are events; dependent individuals are tropes. Perhaps we can make sense of the other combinations as well. How the various particulars weave together metaphysically is another variable matter. For both Brentano and Leśniewski the main significant ontological tie is said to be part-whole, though by setting the supplementation principle aside Brentano arguably introduces a different kind of compounding, one which is closer to dependence. Neither seems very interested in non-individual particulars, which is unsurprising, since they form the Cinderella category of modern Western metaphysics, despite their historical importance in the Aristotelian-Scholastic notion of matter. If stuff, whether endurant or continuant, is admitted, then a principal ontological relation must be constitution: a material individual is constituted (at a time) by material stuff; an event is constituted by one or more processes. Among tropes we can distinguish endurant from perdurant: for example the shape of a body is endurant, while the melody of a concretely played tune is perdurant, since it unfolds in time. There may also be tropes that are neither endurant nor perdurant, but simply are, such as the contour of the melody or the intensity of an earthquake.

THINGS AND TRUTHS

95

The aim of these remarks is not to set out the beginnings of a particularist metaphysics, but to indicate by comparison how relatively impoverished reism is as a basis for a particularist metaphysics by comparison with even slightly more discriminating assays of reality. Metaphysics will progress by looking not at the relatively jejune categories which can be read off the surface syntax of ordinary language, but by struggling to cope with describing and explaining the plethora of things which are thrown at us in natural and scientific experience by the real world. In this struggle however, it is most unlikely that the ways to articulate the language we need will vary from those familiar from everyday and the sciences: there will be sentences, true or false, more or less, names for particulars, predicates true of the particulars named, functors, operators, connectives, quantifiers, and functors, operators etc. of such higher semantic type as might be needed for particular purposes. Simply having this (neutral) linguistic and logical framework will tell us almost nothing about contentual metaphysics, even at the most general level. So it is folly to try to read metaphysics off logical semantics, as happened through much of the twentieth century. Frege, Russell, Leśniewski, Wittgenstein, Carnap, Quine and many others took this route, in one variant or another. But giving up the metaphysical pretensions of logic, while it seemingly limits one aspect of logic, sets it free in another respect. If all we know about a singular name by virtue of its being a singular name is that it is the name of (if anything) an individual, not that it names a body, or a soul, or a substance, or an event, or a simple, then there is no temptation to compromise the metaphysical neutrality of logic by trying to line up the syntactic categories of a language with the ontological categories of a metaphysic. That leaves the semantics of a language to be given with greater freedom from metaphysical baggage. However, the freedom is not absolute: metaphysics constrains semantics. Obviously a semantics given by a nominalist should not employ universals as an explanatory category, because that would be inconsistent with the metaphysical view the person holds. Likewise a sceptic about possible worlds will not wish to introduce them as an explanatory category into the semantics of modality, while a set-theoretic sceptic will not wish her ultimate semantics to involve sets. It may be acceptable pro tem to invoke universals or possible worlds or sets as a heuristic device or an interim so-

PETER SIMONS

96

lution to giving a semantic account of some part of language, but in the long run the semantics of a philosopher has to conform to her metaphysics. Of course this can go both ways: if she becomes convinced that such and such a kind of talk is (a) indispensable to a proper view of the world and (b) cannot be properly explained semantically without invoking items X, then she may well introduce items X into her metaphysical view. This has happened many times in the history of philosophical logic. Conversely, a convinced anti-realist about category X will do her utmost to give a semantics which avoids invoking them. It rarely happens that someone uses items for semantics which she does not consider metaphysically kosher.21 The point of these remarks is to indicate that a preference for a logic such as that of Leśniewski’s ontology has nothing intrinsically to do with accepting reism, whether of the materialistic variety propounded by Leśniewski22, Kotarbiński and Lejewski, or of the dualistic variety propounded by Brentano. It is true that Leśniewski’s and Brentano’s understandings of the way in which their logics’ basic terms were to be understood is in each case compatible with reism and indeed perhaps even suggestive of reism. But the individuals to which reference is made need be neither Brentanian individuals nor Kotarbińskian bodies: they could be tropes, or states of affairs, or events, or other namables. Indeed for a realist about universals they could be universals: the sentence Pride is a mortal sin can have the usual Leśniewskian form A is a b and the difference between pride and some concrete individual, say the proud man X, or his individual trope of pride Y, is a metaphysical one not reflected in the grammar of the words. The job (for a non-reist) of joining up the various categories of individuals by the appropriate metaphysical relations such as participation or foundation is one to be dealt with in a proper metaphysical theory and not reflected through grammar. Logic can be decommissioned from its cooption as a servant to metaphysics and returned to its rightful position as the arbiter of valid inference, no matter

THINGS AND TRUTHS

97

what the subject, provided only that the subject allows true sentences to be framed about it.

5 Truth and Semantics in a World of Particulars For a nominalist, or rather, as I prefer to call her, a particularist, the world is a world of particulars. Arithmetically, these come in three sorts: masses, individuals, and pluralities. Few logical systems deal with the first: mass logic is very undeveloped.23 Even pluralities are not much considered: following Frege, most modern logical systems tolerate only names for individuals, though plurals and – though they are not the same thing – general terms may figure from time to time. Brentano lived and wrote largely before the Fregean watershed which saw general terms ousted in favour of singular terms and predicates: Leśniewski lived and worked after it, but despite his admiration for Frege did not follow him, but retained among his terms ones which could denote several things, as well as singular and empty terms. In this he was deliberately retaining aspects of traditional logic, especially as filtered through Schröder’s Boolean algebra of classes.24 For proximity to natural languages it is probably best to add to the categories of sentence and name that of general term or common noun, as a few have advocated. The first to have suggested this seems to have been Ajdukiewicz.25 To date the only fully worked-out proposal is due to Lejewski.26 But for many logical purposes having common nouns when plurals are available (or vice versa) is unnecessarily complicated. A particularist needs to be particularist through and through. She cannot accept any abstract entities whatever in her account of the world, and that includes her account of logic, language, truth and validity. Of course we all know that ordinary language and scientific language are replete with terms apparently referring to abstract entities of the most multifarious kinds, so a particularist cannot take such talk at face value, but that does not mean she is forced to abandon it. Pending a successful account of how talk apparently about abstract entities manages to be meaningful and true, the particularist may continue speaking with the vulgar and using abstract terms. If particularism is wrong, this simply goes with the necessary grain. If not, then the day of revelation is simply put off until later. But when a

98

PETER SIMONS

particularist gives an account of truth that employs some abstract terminology, it is judicious to select forms of abstract parlance which look unlikely ultimately to incur ontological commitment, and to keep one’s reliance on such parlance to a minimum. Thus a standard account of truth for a formal logical language will typically employ abstract expression types as meaning-bearers, and will rely on a set-theoretic semantics to represent their meanings. Neither types nor sets are acceptable to a particularist, so a particularist account of truth ought to rely on expression tokens and avoid sets. However there is one crucial problem that any reasonable particularist semantics for a formal language needs to overcome, and that may be called the paucity of objects problem. In any semantics we need in principle to take account of a number of meanings that is provably larger than the number of objects in the universe. The most direct way to see this is to consider a simple first-order language with singular names and monadic predicates. A standard account of the truth of sentences of monadic predicate logic needs to consider all the different extensions for a monadic predicate. If there are k objects in the world, there are 2k such extensions, and this is always larger than k. For relational and for higher-order predicates we need correspondingly more extensions. For monadic functions from objects to objects we need kk extensions, and so on up the type hierarchy. If our semantic resources are confined to objects in the world, we are always short. This, more than any other single factor, explains why logicians giving semantic accounts of languages for logic always employ abstract theories such as set theory or function theory to give enough objects to be extensions. Take the general terms of Brentano and Leśniewski. A term from this category can have any one of 2k extensions if there are k individuals, so the number of ways a general term may mean or denote objects outruns the number of objects! Even when the only logic we are considering is propositional, there will be more meanings than just the two truth-values for sentences, and the arithmetic gets worse as we ascend the hierarchy of types. Yet it is clearly bizarre to suppose that simply because there is a difference between truth and falsity, or, to put it in a less committing way, because some things are true and others are false, that we should immediately invest in a large Platonic ontology of functions to truth-values.27 However,

THINGS AND TRUTHS

99

help is at hand. Suppose I have a modest library of 100 books. There are then 100! or, according to my calculator, approximately 9.33 × 10157 ways to arrange them on my shelf in order. This is vastly larger than the number of fundamental particles in the universe, which is estimated as somewhere between 1072 and 1087. When we do the arithmetic on permutations and combinations, the numbers quickly get staggeringly large, yet hardly anyone supposes that in addition to the 100 books on the shelf there are alongside them 9.33 × 10157 ways of arranging them. The number represents not actual thing like the books, but possible arrangements, very few of which will be actually realised: we are counting possibilities, not real things. Also, no one but the most die-hard Platonist supposes we actually have 1087 names, one for each individual particle. Almost none of the many objects in the universe actually get named individually. It suffices that by being there they could in principle be named. If we allow general terms we know for sure that every object does get denoted by at least one name, viz. ‘individual’, which denotes each individual. There is nothing magical about its universal denoting ability: being the term it is, that is what it is there for, and which individuals it ends up actually denoting depends not on it but on them. The denoting relation is here referent-determined and it is coextensional with the relation whose first term is any such universal general term and whose second term is any individual at all. Contingency attaches to this at both ends: what universal terms exist, and what individuals exist. The arithmetic of semantic values for expressions is an arithmetic of possibilities whose identities and number are determined by which things actually exist, and nothing else. We do not count possible worlds and counterparts to find out there are 100! ways to arrange my 100 books: the arithmetic tells us the answer a priori, given that there are 100 books in consideration. Likewise, given k individuals, how many different possible ways can a monadic predicate be true or false of them? The answer is 2k. For dyadic predicates the answer is 22k. The arithmetic does not get essentially more complicated than this. If a given compound expression F(b1… bn) is of category α and can mean in one of #(α) ways and the expressions bi are of categories βi and can mean in #(βi) ways respectively for each i then the number of ways F can mean is

100

PETER SIMONS

#(α) = #(α)^(#(β1) ×…× #(βn)) which gets big quickly but is not theoretically difficult to understand, even if k is infinite.28 For all but the most inveterate modal realists the account of how variable expressions may mean extensionally is one which carries no ontological commitment beyond that to the actual objects, which fix the limits of what is possible. The clue as to which modes of combination need to be drawn upon is taken from the categories of expression actually employed in a given formal language, and that varies from one language to another. In a growing logical system such as Leśniewski’s ontology, where the categories of expression used may be extended by definition, the range of combinations coming into play likewise increases. Consider the axiom of Leśniewski’s system of ontology. This runs for all A and c: A is a c iff for some B, A is a B and B is a c The easiest way to see that this is true is not by running through all possible meanings (which is impossible if there are too many of them, even if finite), but by reductio: suppose it false and infer a contradiction. Since it is a universal quantification assuming it false amounts to assuming there are values of ‘A’ and ‘c’ such that the two sides have opposite truth-values. Let ‘A’ and ‘c’ have such values that the left-hand side is true and the right-hand side is false. For the left-hand side ‘A is a c’ to be true the subject term must denote a single object X and the predicate term ‘c’ must demote one or more objects such that X is among them. For the right-hand side to be false if must be the case that for all B, if A is a B then B is not a c. Now A is an A because the left-hand side is true so since the case where ‘B’ designates just X is included among those quantified over, it follows that X is not one of the things designated by ‘c’, in contradiction to what we found for the left-hand side. So suppose ‘A’ and ‘c’ have such values that the right-hand side is true and the left-hand side is false. Then for some B, ‘A is a B and B is a c’ is true. Let the B in question have the semantic value of denoting just the individual Y, as it must denote a single individual for the second clause to be true; then Y is also one of the one or more individuals denoted by ‘c’. For the first clause of the right-hand side also to be true when ‘B’ denotes Y, ‘A’ must also denote Y and Y only.

THINGS AND TRUTHS

101

Hence ‘A’ denotes just one thing Y and this is one of the things denoted by ‘c’, which contradicts the assumption that the antecedent of the biconditional is false. Hence there are no values for ‘A’ and ‘c’ falsifying the matrix and the axiom is true. Each of the 633 theorems that are listed in Leśniewski’s lecture notes on ontology from 1929–3029 could be shown to be true in a similar way. So the logic is intuitively sound. It may be noted that at no point in the demonstration that the axiom is true did we make any unconditional assertions about the existence of anything: we were assuming of course that we had the axiom and its literal parts, but an axiom is not true if there is no axiom there to be true. It is a desirable feature, common to both Brentano and Leśniewski in their respective logical systems, that no logical axiom entails the existence of any individual, and this feature is preserved under the inference rules. Logic is, for each of them, ontologically neutral,30 and that neutrality extends from individuals to all other categories. The only hint of non-neutrality is that both accept classical bivalence: that all truth-bearers are true or false and none is both; but this is not to accept two objects as it would do if we followed Frege in assuming sentential clauses denoted their truth-values.31 The task of defining truth in Tarskian fashion without benefit of sets or other abstract objects is a complex one and I cannot go through it here, but let us muster the resources needed, which I have indicated in separate places. The first thing to note is that combinatorial semantics for functors32 can be extended to variable-binding operators33, thus obviating the need to bring in satisfaction and sequences as in Tarski.34 Tarski’s definition avoids expressions whose reference is context-dependent, such as indexicals. They can be accommodated by suitably parametrizing the T-schemata for particular cases, and truth remains an absolute concept without there being a need to resort to abstract propositions or other particularistically objectionable abstract entities.35 Additionally, a theory of truth should take account of the paucity of actual truth-bearers, and not expect closure under formation rules or logical consequence, as is generally required, and which in turn requires a Platonistic conception of truth-bearers.36 Invoking the modal side of our thinking about meaning or language will help again here to reduce the cognitive tension. Overall, the outlook for a deflationary account of extension and truth is therefore not bad, and the apparently extreme and trenchant views of Brentano and Leśniewski can be afforded a

102

PETER SIMONS

partial justification. On the other hand intensional aspects of meaning have not been considered. Bringing them within the scope of a particularistically acceptable theory is a much harder task. Leśniewski definitely rejected and disapproved of anything intensional. In Brentano’s case there are aspects of intensionality in his account of logic,37 but they are not worked out. In any case, since there is more to meaning than extension, the way is pointed for further inquiry.

Notes 1

Cf. Kotarbiński 1976, 195, where he points out that his reism, elaborated by him in 1929, owed nothing to Brentano despite Twardowski having been Brentano’s student and Kotarbiński being Twardowski‘s student. The explanation was the late change of mind of Brentano, which took him out of the orbit Twardowski knew into a reism which was not widely read or appreciated when it first appeared around 1914. 2 I am not stressing the differences between singular and plural here. 3 Cf. Leśniewski 1992, 56 f. 4 ‘The Critique of the Logical Principle of the Excluded Middle’, Leśniewski 1992, 62 f. 5 Russell 1994, 421: “it would appear, it must always be self-contradictory to deny the being of anything”. 6 Simons forthcoming (1). 7 Leśniewski 1992, 14. 8 Ibid., 16. 9 I actually think Brentano’s idea works for categorical propositions. See Simons 1987, 2004. 10 Cf. Simons 2004, 60. 11 In Polish ‘sąd’ can also mean a court or trial, so carries similar legal connotations to ‘judgement’. 12 Thinking of Snowy’s white colour is different from thinking that Snowy is white (which thought is of course true if and only if Snowy is white). 13 In Book II of his Summa Logicae. 14 Brentano’s theory of accidents is developed in The Theory of Categories, Brentano 1981, a compilation of manuscripts by Alfred Kastil. 15 On the transition see Chrudzimski and Smith 2004. 16 See the Editors’ Introduction to Brentano 1988, xx. 17 Leśniewski 1992, 380 ff. 18 Lewis 1987, 202 ff. 19 The terms are due to Sider 2001, ch. 5.

THINGS AND TRUTHS

103

20

Simons 2000. An exception it seems is Tarski: see Suppes 1988. Apparently Tarski described himself with a self-irony that is only half-humorous as a “tortured nominalist”. 22 Ignoring Leśniewski’s worries about qualia, which could after all be individuals. 23 But see Simons 2005. 24 Leśniewski 1992, 176 f., 412. 25 Ajdukiewicz 1934; see also Lewis 1972. On plurals, see e.g. Simons 1982, Boolos 1984. 26 Lejewski 1979. 27 Cf. Simons 1995. 28 Simons 1985. 29 Leśniewski 1988, 29–58. 30 For the case of Brentano, see Simons 1987. 31 Cf. Simons 1995. 32 As outlined in Simons 1985. 33 See Simons 2006. 34 Milne 1998 points out that Tarski’s use of satisfaction enables him not only to sidestep the issue whether meaning is a semantic notion, but also to prove the metalogical principles of contradiction and excluded middle. Bivalence results will however only be forthcoming for a language if the range of values for expressions is (possibly restrictively) crafted so that bivalence results. Truth-value and reference gaps, multivalence and vagueness are all phenomena with which a semantic framework should be able to cope. Combinatorial semantics can cope. 35 Simons 1993, 2003. 36 Simons forthcoming (2). 37 For example in taking the complex names ‘a and non-a’ and ‘b and non-b’ to be logically distinct despite their necessary equivalence. See Simons 1987. 21

References Ajdukiewicz, K. 1934. W sprawie uniwersaliów. Przegląd Filozoficzny 37, 219–34. Reprinted in his Języki i poznanie, Warsaw, 1960, 196–210. Boolos, G. 1984. To Be is to Be the Value of a Variable (Or to Be Some Values of Some Variables). The Journal of Philosophy 81, 430–49. Brentano, F. 1981. The Theory of Categories. The Hague: Nijhoff. Chrudzimski, A. and Smith, B. 2004. Brentano’s Ontology: From Conceptualism to Reism. In: D. Jacquette, ed., The Cambridge Companion to Brentano. Cambridge: Cambridge University Press, 197–219. Geach, P. T. 1972. A Program for Syntax. In: D. Davidson & G. Harman, eds., Semantics for Natural Language. Dordrecht: Reidel, 483–97.

104

PETER SIMONS

Kotarbiński, T. 1976. Franz Brentano as Reist. In: L. McAlister, ed., The Philosophy of Brentano. London: Duckworth, 194–203. Lejewski, C. 1979. Idealization of Ordinary Language for the Purposes of Logic. In: D. J. Allerton, E. Carney & D. Holdcroft, eds., Function and Context in Linguistic Analysis. Cambridge: Cambridge University Press, 94–110. Leśniewski, S. 1988. S. Leśniewski’s Lecture Notes on Logic. Dordrecht: Kluwer. Leśniewski, S. 1992. Collected Works. 2 vols. Dordrecht: Kluwer. Lewis, D. 1972. General Semantics. In: D. Davidson & G. Harman, eds., Semantics for Natural Language. Dordrecht: Reidel, 169–218. Lewis, D. 1987. On the Plurality of Worlds. Oxford: Blackwell. Milne, P. 1998. Tarski on Truth and its Definition. In: P. Childers, P. Kolář and V. Svoboda, eds., Logica ’96. Proceedings of the 10th International Symposium. Prague: Philosophia, 189–210. Russell, B. 1994. On Denoting. In: Foundations of Logic, 1903–05. Collected Papers of Bertrand Russell, Vol. 4. London: Allen & Unwin, 414–430. Sider, T. 2001. Four-Dimensionalism. An Ontology of Persistence and Time. Oxford: Clarendon Press. Simons, P. M. 1982. Plural Reference and Set Theory. In B. Smith, ed. Parts and Moments. Munich: Philosophia, 199–256. Simons, P. M. 1985. A Semantics for Ontology. Dialectica 39, 193–216. Simons, P. M. 1987. Brentano’s Reform of Logic. Topoi 6, 23–63. Simons, P. M. 1992. Verità atemporale senza portatori di verità atemporali. Discipline filosofiche 2, 33–47. Simons, P. M. 1995. Leśniewski and Ontological Commitment. In D. Miéville & D. Vernant, eds., Stanisław Leśniewski Aujourd’hui. Université de Grenoble: Recherches Philosophie, Langages et Cognition No. 16, 103–119. Simons, P. M. 2003. Absolute Truth in a Changing World. In J. Hintikka, T. Czarnecki, K. Kijania-Placek, T. Placek and A. Rojszczak, eds. Philosophy and Logic. In Search of the Polish Tradition. Essays in Honour of Jan Woleński on the Occasion of his 60th Birthday. Dordrecht: Kluwer, 37–54. Simons, P. M. 2004. Judging Correctly: Brentano and the Reform of Elementary Logic. In: D. Jacquette, ed., The Cambridge Companion to Brentano. Cambridge: Cambridge University Press, 45–65. Simons, P. M. 2005. Mass Logic. In: K. Trzęsicki, ed. Ratione et Studio. Profesorowi Witoldowi Marciszewskiemu w darze. Białystok: Wydawnictwo Uniwersytetu w Białymstoku, 35–46. Simons, P. M. 2006. Languages with Variable-Binding Operators: Categorial Syntax and Combinatorial Semantics. In J. J. Jadacki and J. Paśniczek, eds., Lvov–Warsaw School. The New Generation. Amsterdam: Rodopi, 239–268. Simons, P. M. forthcoming (1). Straw Meinong and the Present Kings of France. In N. Griffin, ed., Proceedings of the Centenary Conference on Russell’s ‘On Denoting’.

THINGS AND TRUTHS

105

Simons, P. M. forthcoming (2). Truth on a Tight Budget: How to Emulate Tarski but Remain a Nominalist. In S. Lapointe and D. Patterson, eds., The Golden Age of Polish Philosophy. Suppes, P. 1988, Philosophical Implications of Tarski’s Work, Journal of Symbolic Logic 53, 80–91.

The Young Leśniewski on Existential Propositions ARKADIUSZ CHRUDZIMSKI

It was one of Brentano’s central ideas that all judgements are at bottom existential. In his Psychology from an Empirical Standpoint he tried to show how all traditionally acknowledged judgement forms could be reinterpreted as existential statements. Existential propositions, therefore, were a central concern for the whole Brentano School. Kazimierz Twardowski, who also accepted this program (Twardowski 1894, 15f., 25), introduced the problem of the existential reduction to his Polish students, but not all of them found this idea plausible. In 1911 Stanisław Leśniewski published a paper under the title “A Contribution to the Analysis of Existential Propositions” where he criticised Brentano’s translation. According to Leśniewski the consequences of Brentano’s program would be absurd because according to Leśniewski all positive existential propositions are analytically true and all negative ones are contradictory. In his later works Leśniewski repudiated all his early writings (1911–1914) as philosophically immature and formally imprecise. “[…] I regret that they have appeared in print,” he writes, “and formally ‘repudiate’ them herewith […]”. (Leśniewski 1927– 31, 198) But in spite of this severe assessment, these early papers are worth considering not only from a historical standpoint. As we will see, Leśniewski’s critique of Brentano is unsound, but it casts an interesting light on his understanding of certain basic metaphysical concepts.

1. Brentano’s existential reduction In his Psychology from an Empirical Standpoint (1874) Brentano claimed that all four traditional categorical judgement forms that have been investigated since Aristotle could be reformulated as existential forms. This allowed Brentano to construct a very simple theory of intentionality. According to his view the basic form of an intentional state is a presentation (German: Vorstellung) of a nominal object. In a presentation an object is simply “put before subject’s mind” without any claim concerning its exisActions, Products, and Things. Brentano and Polish Philosophy. Arkadiusz Chrudzimski and Dariusz Łukasiewicz (eds.), Frankfurt: ontos, 2006, 107–120.

ARKADIUSZ CHRUDZIMSKI

108

tence.1 Such a claim can be issued only in a judgement which is a higherorder mental state. It assumes a presentation as its basis and consists in a mental accepting or rejecting of the presented object. The success or failure of the existential translation is thus crucial for Brentano’s entire philosophy. The translation goes as follows (cf. Brentano 1911/1925, 164–168): Aristotelian Forms AaB All A are B AiB Some A are B AeB No A are B AoB Some A are not B

Brentanian Forms There is no A which is non-B There is an A which is B There is no A which is B There is A which is non-B

If we symbolise the existential accepting / rejecting as “+/–” and use a negation operator “*” allowing us to build the negative counterpart of any given term (so that “*A” means “non-A”), then the four Brentanian forms will look like that: –A*B +AB –AB +A*B (We assume that the operator “*”always has the minimal scope, so that e.g. “*AB” means “non-A which is B” and not “non-A which is non-B”.) Brentano’s reduction looks very elegant and attractive, but also controversial. It was criticised by Meinong, who argued that within the framework of Brentano’s theory neither the ontological mechanism of the concatenation of terms nor their negation could be satisfactory explained, which led him eventually to the introduction of propositional entities (objectives) as target entities for judgements. Also Leśniewski criticised the Brentanian approach but his reasons were different. In his paper “A Contribution to the Analysis of Existential Propositions” (1911) he argued that Brentano’s existential reduction cannot work, because it makes certain intuitively true judgements into analytically false ones. (Leśniewski 1911, 14) We will see that this criticism is based on a misunderstanding of

LEŚNIEWSKI ON EXISTENTIAL PROPOSITIONS

109

Brentano’s theory. Nonetheless it casts an interesting light on Leśniewski’s understanding of the concepts of object and being.

2. The general form of proposition According to the young Leśniewski all propositions have a subject-predicate form that employs the positive copula. What this means is that all predicates apparently not involving the copula could be split into predicative and copulative part (so that e.g. “barks” becomes “is barking”) and that all propositions of the form “A is not B” could be rephrased as “A is non-B”, in other words: that the negative copula could be generally replaced by negative terms. (Cf. Leśniewski 1911, 10; Leśniewski 1912, 23) If we symbolise the copula as “ε”, then all propositions will take the form “A ε B”,2 sometimes with the negative term in place of “B”. Following our earlier convention, we build the negative counterpart of any given term with help of the operator “*” so that “*A” means “non-A”. If we assume further, that the negation operator can be applied also to the copula, then Leśniewski’s thesis of the reducibility of the negative copula to negative terms could be formulated as follows: A *ε B ≡ A ε *B Also existential propositions have, according to Leśniewski the subjectpredicate forms. If we symbolise the existence-predicate as “E”, they take the following forms: There is an A There is no A

AεE A ε *E

3. The character of the existence-predicate The most important tool employed by Leśniewski in his analysis of existential propositions is the concept of connotation, as defined by John Stuart

110

ARKADIUSZ CHRUDZIMSKI

Mill. The existence-predicate, claims Leśniewski, connotes no property; and precisely because of that it denotes everything. (Leśniewski 1911, 4) According to Leśniewski there is thus the existence-predicate but there is no existence-property. Leśniewski criticises the view that the existence predicate connotes the property of existing. He writes: [S]hould the word ‘being’ really connote the property of existing, we could define this word as ‘that which had the property of existing’, or in other words ‘being which had the property of existing’ (since the definition must indicate not only differentiae specificae, but also the genus); this would, then, give rise to an inevitable regressus in infinitum. The world ‘being’ cannot be in fact defined at all; the statement that this word does not connote anything is fully in keeping with this fact. (Leśniewski 1911, 5)

Leśniewski does not explicate how exactly the vicious regress starts, but it is obvious that he must have in mind the following argument: Should we assume, that the word “exists” (which he equates with “being”) has any non-empty connotation, it had to be contrasted with another word (genus) with a wider extension, say “shmexists”, so that it would be possible to define: An existent is a shmexistent that exists. But for the word “shmexist” the same argumentation figure can be repeated. We have to decide whether it is already connotatively empty or has some still higher genus. If “shmexist” has a higher genus, it will be the beginning of an infinite regress, and if we decide that it is connotatively empty, then we have here after all a word which has the feature that Leśniewski believed to be constitutive of the word “exist”. So why not assume that our “shmexist” is simply “exist” in Leśniewski’s sense? This kind of argument is very instructive of how to understand Leśniewski’s view properly. It resembles Meinong’s argumentation to the effect that the “Außersein” of a pure object is nothing which could be being-like. (Meinong 1904, S. 492–4) According to Meinong the existencepredicate connotes a genuine (although in his terminology extra-nuclear – “außerkonstitutorische”) property, which certain objects do not exemplify.

LEŚNIEWSKI ON EXISTENTIAL PROPOSITIONS

111

Neither triangular circles nor golden mountains exist on Meinongs view. Nonetheless they are objects in Meinong’s sense. They can be referred to in mental acts and some properties can be truly predicated of them. It means that the existence-predicate is on Meinong’s view a differentia which has a higher genus: namely “being an object” or “being außerseind”. But since everything is an object, for the later predicate there exists no such higher genus. This is Meinong’s reason for believing that “being an object” or “being außerseind” cannot be compared with having of any ontological status. Being an object (or being außerseiend) means, therefore, primarily a lack of any ontological status. It is clear that the predicate of existing/being as understood by the young Leśniewski is in this respect very similar to Meinong’s predicate of being an object. In both cases we have a predicate which is true of everything that could be referred to and which specifies no property which would be characteristic for a certain definite group of items. We will return to this point in the last section of this paper.

4. Predicates and terms In our analysis of Leśniewski’s views it will be convenient to use some technical devices which allow us to make perspicuous the inner structure of predicates. What first comes to mind is the so-called lambda abstraction. With help of the lambda operator “λ” we can build from any open formula “…x…” a predicate “λx(…x…)” which is true precisely of those objects which fulfil the formula “…x…”. However, according the standard notation, lambda-terms are syntactically complete predicates, so they are to be directly concatenated with subject terms. The following form (with “A” standing for a singular term) is therefore a well formed proposition: λx(…x…) A But we must remember that Leśniewski uses other syntax. His propositions are composed from the copula, which is flanked by subject and predicate symbols. It means that Leśniewski’s predicates are to be construed rather

ARKADIUSZ CHRUDZIMSKI

112

as general terms like “a red thing” or “a horse” which, when concatenated with the copula, build expressions like “is a red thing” or “is a horse”, which are complete predicates in the Fregean sense. For our purposes it will instead be convenient to use the operator of the indefinite description “η” which allows us to build from any open sentence “…x…” a general term “ηx(…x…)” meaning: “an x which is such that …x…”.3 Here is an example of a well formed proposition in Leśniewski’s style: A ε ηx(…x…) which means: “A is an x which is such that …x…”. Leśniewski’s thesis that each predicate implicitly involves the copula means, in our notation, that each lambda term can be split into an indefinite description term and the copula, according to the following scheme: λx(…x…) A ≡ A ε ηx(…x…)

5. The descriptive theory of the subject terms One point that is very important for Leśniewski’s early treatment of existential propositions is his further conviction that all subject terms have a descriptive content. He does not accept Mill’s non-connotative names and says that each name connotes at least the property of bearing this very name. “Instead of ‘Paul’” – we read – “we can say ‘a being which has the name ‘Paul’’.” (Leśniewski 1911, 6; cf. also Leśniewski 1912, 22) It means that for the young Leśniewski, as for the majority of pre-Fregeans, all terms become in the end general terms, in that they refer to their objects by means of some identifying properties that can in principle be exemplified by a plurality of objects. It means further that each subjectterm can be interpreted as an indefinite description; and that it can become “singular” either by chance (if it specifies an identifying property that “happens” to be exemplified by only one object) or in a “Scottish” way (if

LEŚNIEWSKI ON EXISTENTIAL PROPOSITIONS

113

it specifies an identifying property which is a haecceity, i.e. a property which by definition could be had by at most one object).4 If we remember that we already placed an indefinite description at the right side of the copula, we see now that in truth both flanking terms are of the same syntactic nature, so that the following is a well formed formula: ηx(__x__) ε ηx(…x…), meaning: “An x which is such that __x__ is an x which is such that …x…”.5

6. All positive existential propositions are analytically true After this introduction it shouldn’t be hard to imagine why Leśniewski believed that all positive existential propositions have to be construed as analytically true. As each subject term takes the form: “a being which is…” and the whole content of the existence-predicate is nothing over and above the content “is a being”, it is clear that each positive existential proposition will be analytic in the Kantian sense. (Leśniewski 1911, 6) As we have seen, Leśniewski’s claim is that the existence- or beingpredicate is indefinable, but if we had to ignore his conviction, then the best candidate for such a connotatively empty existence-predicate (or strictly speaking: existence-term) would probably be: ηx(x = x) In what follows we will use this term in place of Leśniewski’s “existing” or “being”, but nothing of substance will depend on this stipulation. We could as well assume that being is really primitive and symbolise the being-term as “ηx(there is an x)” or “ηx(Ex)”. Under the assumption that being is to be construed as self-identity, the general form of the positive existential proposition will be: ηx(x = x ∧ …) ε ηx(x = x)

114

ARKADIUSZ CHRUDZIMSKI

We see that on the right side of the copula there is nothing which couldn’t be already found on its left side, which is the defining feature of analyticity in the Kantian sense.

7. All (standard) negative existential propositions are synthetic and contradictory Leśniewski’s conviction that all standard negative existential propositions are synthetic and contradictory (and a fortiori false) should now be understandable. If each subject term has a form “a being which is…” and a negation of the existence-predicate can only be understood as “is a non-being” then it follows that such a proposition (i) contains in its predicate something which is definitely not contained in the subject (Leśniewski 1911, 10) and therefore is synthetic in the Kantian sense; and (ii) this “something new” contained in the predicate term contradicts something which is contained in the subject term. (Leśniewski 1911, 12) The form of the negative existential proposition will be: ηx(x = x ∧ …) ε *ηx(x = x) This form could be translated as: ηx(x = x ∧ …) ε ηx(x ≠ x), which makes the contradiction obvious. Surprisingly enough, there is, according to Leśniewski, also a special group of negative existential propositions which are analytic. He means here propositions which describe their objects as non-beings already in their subject terms. As a general form of a subject term is: “a being which is…”, a proposition of this kind involves a contradiction already in its subject term (“a being which is non-being and…”) and will therefore be (like each standard negative existential proposition) contradictory and false. Nonetheless its predicate contains nothing which is not contained already in its subject; and this feature allows us to classify such propositions as analytic in the Kantian sense.

LEŚNIEWSKI ON EXISTENTIAL PROPOSITIONS

115

The general form of such a proposition will be: ηx(x = x ∧ x ≠ x ∧ …) ε ηx(x ≠ x) In any event, all negative existential propositions are, on this account, false. A rather surprising consequence of Leśniewski’s analysis is therefore that the intuitively necessarily true proposition: Non-beings do not exist which Simons termed “Heidegger’s Law” (Simons 2004, 63), but which could be as well called “Parmenides’ law”, turns out false! The reason is that Leśniewski deprived himself of resources which would allow him to refer consistently to non-beings (term “Λ” in Simons 2004). He insists that each subject term implicitly classifies its object(s) as being(s) and he takes this classification to be in a sense “more important” for determining the truth-value of the whole proposition than the subsequent classification as non-being. The proposition “Non-beings do not exist” looks in our notation as follows: ηx(x = x ∧ x ≠ x) ε ηx(x ≠ x) and we see that it is in a sense at the same time analytic and contradictory. The fact that Leśniewski construes it as false explains the sense in which the classification of an object as “being”, involved in the subject term, is for him “more important” for determining the truth-value of the whole proposition than its correlative classification as “non-being”, which is there contained as well. It would be equally possible to reverse this hierarchy and classify negative existential propositions with contradictory subject terms as analytically true.

8. Leśniewski and Brentano According to Leśniewski it follows from his own analysis of the existential propositions that Brentano’s reform of logic cannot function properly.

116

ARKADIUSZ CHRUDZIMSKI

Brentano’s translation would have overtly absurd consequences which Leśniewski illustrates by the following example. The proposition: Paris is not situated in China reads on Brentano’s account: A Paris situated in China does not exist Now, on Leśniewski’s view, this latter proposition must (like all negative existential propositions) turn out false and if it really were equivalent to the proposition that Paris is not situated in China, then this latter proposition couldn’t be true as well. But be honest: it is perfectly true that Paris is not situated in China, which according to Leśniewski demonstrates clearly that Brentano’s translation must be seriously flawed. (Leśniewski 1911, 15 f.) However, what Leśniewski has really demonstrated is rather the fact that there must be certain deep differences between his and Brentano’s view, for according to Brentano the proposition “A Paris situated in China does not exist” is after all not false but true. Indeed, according to Brentano neither positive nor negative existential propositions could be generally regarded as necessarily true or false. In fact there are many philosophically significant differences between Leśniewski’s and Brentano’s treatment of existential propositions, but the one which is relevant for the present issue can be summed up as follows: According to Brentano there is a difference of meaning between “being a presented object” and “being existent”, so that an existential proposition “A exists / does not exist” will be only exceptionally analytically true or false.6 According to Leśniewski there is no such difference. In both cases the putative referent is classified as “a being”.7

9. Two interpretations There are two possible interpretations of this difference. The first interpretation says that (i) Leśniewski’s subject-terms are ontologically loaded, the second that (ii) his existence-predicate is ontologically neutral.

LEŚNIEWSKI ON EXISTENTIAL PROPOSITIONS

117

The first interpretation is at first sight the most natural one. It is a characteristic feature of Brentano’s philosophy that each object x is primarily “given” to a subject’s mind in an ontologically neutral way, so that it makes sense to ask whether x exists or not. Brentano calls this neutral way of being given “presentation” (Vorstellung) and presentations are mental acts which give meaning to the subject terms of any proposition. The first interpretation says that in Leśniewski there is no such ontologically neutral way of presentation. Each subject-term contains an implicit classification of its object as “a being” and Leśniewski’s argumentation to the effect that the existence-predicate simply repeats this classification seems to support the hypothesis that, in contrast to Brentano, his subject-terms are already ontologically loaded. However, we should not forget that the starting point of Leśniewski’s entire argument is his claim that the existence-predicate is connotatively empty.8 This may dispose us to take seriously the second interpretation, according to which neither Leśniewski’s subject terms, nor his existencepredicate contain anything of ontological import. On this reading Leśniewski’s “being a being” would mean nothing more then “being an object”. Indeed, in his paper “An Attempt at a Proof of the Ontological Principle of Contradiction”, published a year later than “A Contribution to the Analysis of Existential Propositions”, Leśniewski writes that “the word ‘object’ is […] the symbol of everything” (Leśniewski 1912, 41) and argues that [t]he word ‘object’, just as the word ‘being’, cannot be defined at all because there is no expression which could be generic with respect to the word ‘object’ […]. (Leśniewski 1912, 43)

It seems that the words “existent”, “being” and “object” have indeed to be understood as synonymous and it is not improbable that none of these concepts was by Leśniewski intended to have any ontological import. Needless to say, according to this construal the young Leśniewski would in a sense be a Meinongian, but the similarities between Leśniewski’s concept of “being” and Meinong’s concept of “a pure object” that were pointed out in section 3 seem to support this interpretation.

118

ARKADIUSZ CHRUDZIMSKI

To close, I believe that it is impossible to give a definitive answer to the question of which interpretation is ultimately the right one, solely on basis of the text. It is even possible that Leśniewski himself was in 1911 not quite clear about these matters. But from the perspective of the later development of his thought (I mean here in particular his liberal treatment of quantifiers) the Meinongian interpretation seems to be much more interesting.

Notes 1

“Wir reden von einem Vorstellen, wo immer uns etwas erscheint.” (Brentano 1874/ 1925, 34) 2 In his later works Leśniewski used a similar notation, namely “ε {A B}” for “A is B”. (Cf. Leśniewski 1930, 609) 3 The η-operator has been introduced in Hilbert/Bernays 1934–39. Its intended meaning could be best explained by the following inference rule: ∃xFx |− F(ηxFx). 4 This neglecting of singularity is best illustrated by the following citation from Lewis Carroll’s Symbolic Logic (1st edition 1896): “A proposition, whose Subject is an Individual, is to be regarded as Universal. Let us take, as an example, the Proposition ‘John is not well’. […] [T]he Class ‘men referred to by the speaker when he mentions ‘John’’ is a one-Member Class, and the Proposition is equivalent to ‘All the men, who are referred to by the speaker when he mentions ‘John’, are not well.” (Carroll 1886/1977, 68) 5 This is a characteristic feature of the ε-predicate as used by the late Leśniewski that neither of the terms flanking “ε” need be singular. (Cf. Simons 1987, 21) 6 According to Brentano there is an important group of analytically true negative existential propositions (the rejections of contradictory objects) and only one case of the analytically true positive existential proposition (the proposition “God exists”, provided the meaning of “God” is here really “clearly and distinctly” grasped, which for “finite” human minds could not be the case). All other acceptations and rejections are synthetic. 7 A further difference which is not important in the present context, but which is absolutely central for the right understanding of Brentano’s philosophy, concerns the “deep grammar” of the existence-predicate. We have seen that according to Leśniewski there is, strictly speaking, no existence-property, but nonetheless there is a (connotatively empty) existence-predicate. Now, on Brentano’s view there is in a sense no existential predicate either. True enough, in expressing existential judgements we use words like “is”, “exists”, or “there is”, and in our symbolism we use the symbols “+/–”, but the

LEŚNIEWSKI ON EXISTENTIAL PROPOSITIONS

119

central point of Brentano’s theory was that in so doing we do not refer to any item that could be sensibly classified as a part of the judgement content. By the existential “+/–” we rather express a mental modus of accepting or rejecting of a presented object. The syntax of Brentanian judging is therefore very simple and this fact has direct consequences for the theory of truth-making and a fortiori for the whole ontology. 8 In Leśniewski 1913 we read even that because of the emptiness of the existence predicate all existential propositions (i.e. also the positive ones which are analytic in Kantian sense) are false. (Leśniewski 1913, 59)

References Brentano, Franz 1874/1924. Psychologie vom empirischen Standpunkt, Bd. I, hrsg. von O. Kraus, Leipzig: Meiner [1st ed. 1874]. Brentano, Franz 1874/1925. Psychologie vom empirischen Standpunkt, Bd. II, hrsg. von O. Kraus, Leipzig: Meiner [1st ed. 1874; 2nd ed. with the „Anhang” 1911]. Brentano, Franz 1911/1925. “Anhang zur ‘Klassifikation der psychischen Phänomene’”, in: Brentano 1874/1925, 131–182. Carroll, Lewis 1886/1977. Symbolic Logic, ed. by W. W. Bartley, Hassocks: Harvester Press. Hilbert, D. / Bernays, P. 1934–39. Die Grundlagen der Mathematik, vols. I/II, Berlin: Springer. Leśniewski, Stanisław 1911. “A Contribution to the Analysis of Existential Propositions”, in: Leśniewski 1992, vol. I, 1–19. Leśniewski, Stanisław 1912. “An Attempt at a Proof of the Ontological Principle of Contradiction”, in: Leśniewski 1992, vol. I, 20–46. Leśniewski, Stanisław 1913. “The Critique of the Logical Principle of the Excluded Middle”, in: Leśniewski 1992, vol. I, 47–85. Leśniewski, Stanisław, 1927–31. “On the Foundations of Mathematics”, in: Leśniewski 1992, vol. I, 174–382. Leśniewski, Stanisław 1930. “On the Foundations of Ontology”, in: Leśniewski 1992, vol. II, 606–628. Leśniewski, Stanisław 1992. Collected Works, vols. I/II, edited by S. J. Surma, J. T. Srzednicki, D. I. Barnett, V. F. Rickey, Warszawa/Dordrecht/Boston/London: PWN / Kluwer Academic Publishers. Meinong, Alexius 1904. “Über Gegenstandstheorie”, in: Gesamtausgabe, Bd. II, 481– 535. Meinong, Alexius 1969–78. Gesamtausgabe, hrsg. von R. Haller et al., Graz: Akademische Druck- und Verlagsanstalt. Simons, Peter 1987. Parts, Oxford: Clarendon Press.

120

ARKADIUSZ CHRUDZIMSKI

Simons, Peter 2004. “Judging Correctly: Brentano and the Reform of Elementary Logic”, in: D. Jacquette, (ed.), The Cambridge Companion to Brentano, Cambridge: Cambridge University Press 2004, 45–65. Twardowski, Kazimierz 1894. Zur Lehre vom Inhalt und Gegenstand der Vorstellungen, Wien: Hölder.

On the Phases of Reism1 BARRY SMITH

§1. Introduction Along with almost all the more important Polish philosophers of the twentieth century, Kotarbiński, too, was a student of Kasimir Twardowski, and it is Twardowski who is more than anyone else responsible for the rigorous thinking and simplicity of expression that is so characteristic of Kotarbiński’s work. Twardowski was of course himself a member of the Brentanist movement, and the influence of Brentanism on Kotarbiński’s writings reveals itself clearly in the fact that the ontological theories which Kotarbiński felt called upon to attack in his writings were in many cases just those theories defended either by Twardowski or by other thinkers within the Brentano tradition. Leśniewski, too, inherited through Twardowski an interest in Brentano and his school, and as a young man he had conceived the project of translating into Polish the Investigations on General Grammar and Philosophy of Language of Anton Marty, one of Brentano’s most intimate disciples. Leśniewski, as he himself expressed it, grew up “tuned to general grammar and logico-semantic problems à la Edmund Husserl and the representatives of the so-called Austrian School”. (1927/31, p. 9) The influence of Brentanism on Polish analytic philosophers such as Kotarbiński and Leśniewski has, however, been largely overlooked – principally as a result of the fact that the writings of the Polish analytic school have been perceived almost exclusively against the background of Viennese positivism or of Anglo-Saxon analytic philosophy. It is hoped that the present paper, following in the footsteps of Jan Woleński’s recent work,2 might do something to help rectify this imbalance. The paper will consist of a critical survey of Kotarbiński’s development, from his early nominalism and ‘pansomatistic reism’ to the later doctrine of ‘temporal phases’. It will be shown that the surface clarity and simplicity of Kotarbiński’s writings mask a number of profound philosophical difficulties, connected above all with the problem of giving an adequate account of the truth of contingent (tensed) predications. The paper will examine in particular the attempts to Actions, Products, and Things. Brentano and Polish Philosophy. Arkadiusz Chrudzimski and Dariusz Łukasiewicz (eds.), Frankfurt: ontos, 2006, 121–182.

122

BARRY SMITH

resolve these difficulties on the part of Leśniewski. It will continue with an account of the relations of Kotarbińskian reism to the ontology of things or entia realia defended by the later Brentano. Kotarbiński’s identification of Brentano as a precursor of reism is, it will be suggested, at least questionable, and the paper will conclude with a more careful attempt to situate the Brentanian and Kotarbińskian ontologies within the spectrum of competing ontological views.

§2. Stages in the Development of Reism We shall be concerned, in the first place, with Kotarbiński’s magnum opus, the Elements of the Theory of Knowledge, Formal Logic and Methodology of the Sciences, first published in 1929 and hereafter referred to as Elementy. The principal doctrine expounded and defended by Kotarbiński in this work is that of ‘reism’, a doctrine according to which all existence is made up entirely of individual things, realia or concreta. A more specialised version of the doctrine is referred to by Kotarbiński as the doctrine of ‘somatism’ (or sometimes also ‘pansomatism’), which results when one adds the thesis that individual things are to be identified in every case as physical bodies – a thesis which Kotarbiński also accepts. In an essay of 1958 appended to the second edition of his Elementy, Kotarbiński speaks retrospectively of seven ‘stages’ in the development of the reistic theory, from his own early acceptance of nominalism – which he himself preferred to call ‘concretism’ 3 – to the working out of a full-blown pansomatist ontology in the 1930s. It is especially in relation to the early stages that Twardowski’s influence is most strongly felt. Stage 1 consists in the rejection of universals, properties, or general objects. All entities are individuals, on this (‘concretist’) view, though it does not thereby follow that they must all be things. Kotarbiński’s adoption of nominalism in this sense may be attributed on the one hand to the effects of his early exposure to the thinking of the British empiricists at the hands of Twardowski. On the other hand however it can be seen as a reaction against Twardowski’s own thesis that there are general objects, objects which result when the features common to the particular objects falling under a given concept are ‘unified into a whole’.4 A discipline like geometry, in Twardowski’s view, is concerned precisely with

ON THE PHASES OF REISM

123

general objects in this sense (Triangle, Circle, Square, and so forth), and a similar thesis may be extended to the other sciences. As Husserl points out in his Logical Investigations, general objects as conceived by Twardowski are subject to all the disadvantages of the Lockean general triangle.5 This suggestion is taken still further by Leśniewski, Kotarbiński’s colleague in Warsaw, who offers a proof that Twardowski’s theory (together with a range of similar theories, including Husserl’s own) is contradictory,6 and Kotarbiński would later claim that it was only concerning universals or general objects that nominalism “succeeded in convincingly proving their non-existence by a reductio ad absurdum.” (1966, p. 55) Stage 2 consists in the rejection of events, processes, states of affairs, and other putative individuals falling outside the category thing. This, too, may be interpreted as a reaction on Kotarbiński’s part to the Brentanist views of his teacher Twardowski, given that the ontology of states of affairs or Sachverhalte was, in the first decades of this century, a quite peculiar preserve of the Brentanist movement.7 Stage 3, the rejection of sets or classes, reflects the influence of Leśniewski, and above all of Leśniewski’s criticisms of the theory of sets.8 The set-theoretical antinomies had resulted, in Leśniewski’s view, not from any inherent contradiction in the notion of set as originally conceived by Cantor, but from a departure from this notion in the direction of a conception of sets as abstract entities. As Cantor puts it: “Every set of well-differentiated things can be conceived as a unitary thing in which these things are constituent parts or constitutive elements.” (Cantor 1887/88, p. 379) Thus for example a musical composition is a set consisting of the sounds which are its constituents, a painting is a set consisting of various patches of colour. A set, therefore, on Leśniewski’s interpretation of Cantor’s views, is a concrete whole made up of concrete parts, not an abstract entity sealed off from changes in the real world of material things.9 According to Leśniewski, it would be correct to say, for example, that the Black Forest is just the set of trees now growing in a certain area, and that this set becomes smaller as trees within it die. Clearly, on this view, there can be no empty set, and a set consisting of just one object as member will be identical with that object. Further, there can be no sets of higher type, which means also that there is no way in which the more usual antinomies may be generated.

124

BARRY SMITH

Frege, too, as Leśniewski points out, attacks those mathematicians who introduce into their theories such arbitrary ‘inventions’ as the empty set, merely because they prove expedient for certain purposes.10 Leśniewski’s own strictures in this respect are directed in particular against axiomatic theories of sets such as were developed by Zermelo. These do not merely lack the sort of naturalness that would dispose one to accept them; they lack also that intrinsic intelligibility which would make their meaning clear, so that Leśniewski can in all honesty assert that he does not understand what is meant by ‘set’ as this term is supposed to be ‘implicitly defined’ by theories like Zermelo’s.11 Leśniewski himself, in contrast, starts not from ‘inventions’ or from axioms or hypotheses selected for pragmatic reasons, but from what he calls intuitions, commonly accepted and meaningful to all, relating to such concepts as whole, part, totality, object, identity, and so on.12 The language of Leśniewski’s theories is therefore an extrapolation of natural language, a making precise of what, in natural language, is left inarticulate or indistinct. His work forms part of that strand in the development of logic, represented also by Frege and by the early Russell, which sees logic as a descriptive enterprise, part and parcel of the attempt to produce formal theories adequate to and true of the actual world.13 Hence he is mistrustful, too, of the model-theoretic semantics that has been built up on an abstract settheoretical basis, and he is opposed also to the work of those formalist logicians who embrace an essentially abstract-algebraic approach to logic, or see logic as having to deal essentially with uninterpreted formal systems.14 Stage 4 in the development of reism, which consists in the rejection by Kotarbiński of mental images and other ‘immanent contents’, again reflects the influence of Twardowski. The status of mental entities was an issue of particular importance to Kotarbiński, since it marked one of the very few areas of disagreement between himself and Leśniewski. For Leśniewski admitted contents and images into his ontology, remaining in this respect faithful to the heritage of Brentano and Twardowski.15 Thus Leśniewski, in this sense and perhaps also in others, is not a reist. Since, however, like Brentano, Marty and Twardowski, he held that contents are concrete items, tied, in effect, to specific mental episodes, his acceptance of contents does not imply a departure from nominalism or ‘concretism’.

ON THE PHASES OF REISM

125

Stage 5, which consists in the awakening of Kotarbiński’s interest in certain precursors of his own way of thinking, was provoked by the discovery of what he took to be reist tendencies in Brentano’s later work – a matter which will be dealt with in more detail below. Stage 6 consists in an amendment to the reist doctrine, provoked by criticisms put forward by Ajdukiewicz in his 1930 review of the Elementy.16 These criticisms concern in particular the question as to how the negative theses of reism (‘properties do not exist’, ‘events do not exist’, and so on) are properly to be treated. Are such formulations to be accepted as they stand as literal renderings? Certainly not, Ajdukiewicz claims, if ‘exists’ is taken in the literal sense – the sense it has in sentences like ‘rabbits exist’, ‘dinosaurs no longer exist’ and so on. For the subjects of such sentences are in every case the names of things, which is ex hypothesi not the case where we have to deal with expressions like ‘property’, ‘event’, and so on. Yet the reist allows no other sense of ‘exists’. Kotarbiński himself initially responded to this criticism by taking up Ajdukiewicz’s suggestion that the negative theses of ontology be reformulated on the level of semantics, as theses to the effect that certain kinds of apparent statement are nonsensical.17 This solution is unsatisfactory for a number of reasons. It implies, first of all, a view of ‘words with negative meanings’ which would seemingly allow them to be employed as a means of constructing well-formed sentences more or less at random from nonsensical strings of words. Further, as Lejewski points out in his paper On the Dramatic Stage in the Development of Kotarbiński’s Pansomatism, it implies that ‘the negative theses of ontological reism fail to say anything about reality’ because they are merely ‘statements about the language of the reist.’ (1979, p. 200) A semantical doctrine in this sense must however, as Lejewski argues, presuppose a prior ontological doctrine: ‘Semantics without ontology is like a house without foundations. It collapses into a set of arbitrary injunctions and prohibitions justified by ad hoc considerations.’ (1979, pp. 205f.) Moreover, how, in the absence of some more deep-seated ontological theory, could the reist be assured of the truth of his semantic claim that all onomatoids will vanish in ultimate formulations? And how could he account for the fact that, as he will want to insist, translation into the language of things is both natural and clarificatory?

126

BARRY SMITH

There is, however, an alternative response to Ajdukiewicz’s criticism. This is to accept that the reist’s negative theses make good sense (are in good grammatical order) as they stand, not, however, in the language of the reist but in the language of his opponent, i.e. of someone who accepts both a multicategorial ontology (accepts categories other than that of thing) and the concomitant multicategorial language. For if, as Lejewski puts it, the multicategorial ontologist’s assertions ‘are made in terms of a multicategorial language, the same language must be used to negate those assertions’. Propositions such as ‘there are no properties’, ‘there are no relations’, ‘there are no events’, etc., are properly to be understood in the light of the multicategorial idealization of natural language. And on this assumption the nouns ‘property’, ‘relation’, ‘event’, etc. belong to different fundamental semantical categories, which in turn determine the semantical category of the expression ‘there are no’ in each of the conjuncts. (Lejewski 1979, pp. 211f.)

The reist, we might say, can accept his opponent’s multicategorial language as a ladder, to be thrown away when once it has served its polemical purpose. Stage 7 sees the re-institution of reism as an ontological doctrine, founded on a recognition of the need to supply non-tautological definitions of notions such as ‘thing’, ‘object’, ‘body’, etc. This development, too, was provoked by a criticism of Ajdukiewicz, a criticism to the effect that, if ‘exists’ has a literal sense only when used in conjunction with names for things, then the positive statement of reism, to the effect that only things exist, is tautological. It is equivalent to the truism: ‘only things are things’. Here, also, Kotarbiński’s initial reaction was one of retreat to semantics. Later, however, he responded to Ajdukiewicz’s objection in a more ontologically-minded fashion, by seeking definitions of concepts such as ‘thing’, ‘object’, ‘body’, etc., in a way which he hoped would render nontautological the fundamental theses of reism and somatism. Since a formal statement of such definitions has been provided by Lejewski in his just-mentioned paper, it will be sufficient here if we examine briefly (and critically) the concepts Kotarbiński here employs.

ON THE PHASES OF REISM

127

All reality, according to Kotarbiński, is composed exclusively of things, and things are in every case bodies. Kotarbiński initially sought to define body as that which is extended in space and time, as that which is ‘bulky and lasting’. Then, however, he saw reason to add the further condition that bodies are ‘inert’. Certainly it would be sufficient, Kotarbiński holds, to define ‘body’ as ‘that which is extensive’: But in order to avoid misunderstandings which might lead someone to suppose, on the strength of that definition, that physics is also concerned with ‘fragments of empty space’ (which in our opinion, do not exist) or ‘immanent coloured patches’ (which also seems to be a hypostasis), we prefer to narrow the definition as to intension – without thereby, as we think, narrowing its extension – by adopting the formula stating that ‘a body is what is extensive and inert’. (1966, p. 330)

In other contexts Kotarbiński preferred to define body as: that which is extensive and such as to offer resistance. All is not quite clear, however, about the application of either formula. Thus Kotarbiński is on the one hand keen to insist that the term ‘body’, as he understands it, embraces not only planets, rocks, etc., but also objects investigated by physics ‘such as electrons, protons, magnetic fields’. (1966, p. 331) On the other hand, however, he stresses that it excludes for example ‘immanent coloured patches’. Consider, however, the example of a glass cube that is uniformly red in colour. Is the transcendent redness of this cube (an individual three-dimensionally extended moment of colour), a body, on Kotarbiński’s view? Certainly it is bulky and lasting and, perforce, such as to offer resistance. Kotarbiński, it would seem, was able to ignore such cases in framing his account of ‘body’ only because his attentions were concentrated on instances of surface colour, entities which fall short of three-dimensionality and can be excluded on this count. In order to rule out examples like the cube of colour, Kotarbiński would have to add the condition that a body is that which exists (is extended and inert) in its own right (has need of no other thing in order to exist).18 As we shall see, a condition of this sort is very much in the spirit of Aristotle. Certainly such a condition would capture the sense in which the given example gives grounds for suspicion – that the cube of colour exists merely as a dependent moment of the cube of glass, and enjoys no separate existence. Yet how are we to formulate the condition in question in such a way that it would not rule out other examples

128

BARRY SMITH

which we would wish to count as bona fide bodies? Does a human being, for example, exist ‘in his own right’, given that he has need, for example, of nourishment, and processes of breathing and metabolising (to say nothing of parents), in order to exist? How, moreover, are we to make precise the sense of ‘other’ in ‘has need of no other thing’? Simple non-identity will not do, since everything may in this sense stand in need of its own proper parts in order to exist. On the other hand spatiotemporal discreteness or disjointness will not serve, either, since the cube and its colour would seem to coincide in space and time. All that can be said here is that considerations such as this have exercised Kotarbiński (and Leśniewski, et. al.) too little, so that the project of a somatist ontology still leaves much to be desired in terms of a clear statement of what is meant by ‘body’.

§3. Reism and Truth Kotarbiński’s reism is, as we have seen, a doctrine according to which all existence is made up entirely of individual things. At the same time he defended in the Elementy a form of the correspondence theory of truth derived from his teacher Twardowski. Twardowski himself had at least at one stage in his career come to the conclusion that a conception of truth as correspondence requires special ‘states of affairs’, unitary entities which would stand to sentences or acts of judgment in something like the way in which things or objects in the narrow sense would stand to names or acts of presentation.19 Not only Twardowski, but also Husserl, Meinong and Marty, as well as the early Moore, Russell and Wittgenstein were drawn to similar views. How, then, was it possible for Kotarbiński to maintain a correspondence theory of truth and at the same time embrace the view that there are no entities other than things? Kotarbiński himself holds that this dilemma may be resolved by rejecting that ontological interpretation of correspondence (derived from the scholastics) which would interpret truth in terms of ‘copies’ of reality existing somehow in the mind of the judging subject. He advances, rather, what might nowadays be called an adverbial theory of truth:

ON THE PHASES OF REISM

129

The point is not that a true thought should be a good copy or simile of the thing of which we are thinking, as a painted copy or a photograph is. A brief reflection suffices to recognize the metaphorical nature of such a comparison. A different interpretation of ‘accordance with reality’ is required. We shall confine ourselves to the following: ‘John thinks truly if and only if John thinks that things are so and so, and things in fact are so and so’. (1966, pp. 106f.)

He came, in other words, to interpret the correspondence theory in the superficially neutral terms of the Aristotelian ‘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 it is not, is true.’20 Conceptions of truth in terms of the copy theory are to be avoided, from Kotarbiński’s point of view, not merely because they involve a hypostatisation of states of affairs or other special entities on the side of the object; they commit us also, on the side of the subject, to ‘immanent contents’, ‘thoughts’, ‘judgments’, ‘propositions’ or ‘meanings’ – and all of these terms are mere façons de parler, to be eliminated from any language adequate to the purposes of philosophy. When I judge truly, then I judge in accordance with the things, and that is all that need be said. Can matters really be so simple, however? Certainly in the case of judgments expressed by positive existential sentences such as ‘John exists’ or ‘cheetahs exist’, it is plausible to account for their truth or falsehood exclusively on the basis of an appeal to things or bodies as commonly understood. ‘John exists’, on a view of this sort, is made true by John himself; ‘cheetahs exist’ by some one or more cheetahs.21 But how, on this basis, are we to deal with negative existential judgments like ‘Baal does not exist’ or ‘there are no unicorns’. It was precisely difficulties in the treatment of judgments such as this which led some Brentanists to the view that what makes a judgment true are special sui generis entities designated by expressions of the form: the existence of A, the non-existence of B, the existence of an A which is B, and so on, where ‘A’ and ‘B’ stand in for expressions like a horse, the redness over there, unicorns, God, Baal, and so on. The consideration of sentences like ‘John is suntanned’, ‘John is eating’, ‘John is a heavy eater’, ‘John’s eating is on the increase’, ‘John has a bad case of dispepsia’, ‘there is a ridge of high pressure over the Atlantic’, suggests moreover that the domain of such special, non-thingly truth-

130

BARRY SMITH

makers must be extended even more widely, to embrace events, processes and conditions, as well as complex states of affairs involving these as parts. How, then, can Kotarbiński cope with cases such as this in a way that will not stretch ontologically beyond the realm of things? Two answers to this question may suggest themselves: (1) that it would be possible to effect a logical or linguistic analysis of the sentences in question, of a sort that would reveal their underlying form as involving a relation only to things; (2) that it would be possible to embrace special sorts of things as truth-makers for the given sentences, so that reism would be saved, though only at the expense of our embracing a notion of ‘thing’ which would depart in some degree from common sense. As we shall see, elements of both solutions are present in Kotarbiński’s work.

§4. Problems of Semantics Let us look, first, at the semantic side of Kotarbiński’s doctrine. Consider the sentence ‘John’s jump cleared the hurdle’. This seems to refer to a certain concrete individual event or process – John’s jump – which occurred at a certain time. And it must therefore surely correspond, if true, to a segment of reality containing this jump as part. We have a strong intuitive disposition to suppose that any account of what makes the given sentence true will be inadequate if it takes no account of this specific jump. According to Kotarbiński, however, this intuition cannot even be properly expressed. For all apparent references to jumps and other events or processes are ‘merely substitutive’. A literal rendering of the intentions of one who utters the sentence in question would be: ‘John jumped clear of the hurdle’, a sentence in which the only names that occur are names for things.22 It is renderings of this sort, Kotarbiński insists, that reproduce ‘the intention of any statement that says something about an event or events.’ For, ‘it is only seemingly (and never in fact) that we can make a true statement about an event, namely if we take that statement in its substitutive, and not literal and fundamental role.’ References to events are mere ‘onomatoids’ or ‘apparent names’. They are terms which merely sound like

ON THE PHASES OF REISM

131

names. When the attempt is made to establish a literal interpretation, then it becomes clear that the expressions in question belong to a category quite different from that of names in the strict and proper sense.23 Physics too, of course, along with many other sciences such as phonology, military history and meteorology, seems to trade largely in sentences of greater or lesser generality about events. Such disciplines are, accordingly, in need of radical linguistic reform, so that, as Kotarbiński points out, ‘one of the most topical but unperformed tasks of concretism is to work out a dictionary of mathematics and physics [and of other sciences] in the reistic interpretation.’24 It is not difficult to appreciate the obstacles confronting such a project in relation, say, to the physicist’s talk of energy-fields characterising points or regions of spacetime. Kotarbiński in fact copes with the latter not by semantic means, but ontologically. As we have seen, he accepts into his ontology fields and other creatures of physics. These, too, are extended in space and time and are ‘such as to offer resistance’ (or, at least, they are presumably such as to be involved in causal relations of certain sorts). Reality, as the reist conceives it, is not therefore ‘a static conglomerate’ (‘a mere sum’) of ‘rigid and changeless solids’; it is a ‘fabric composed of changing things’, in a new and extended sense of ‘thing’.25 No explicit criterion is provided, however, as to what is ‘thing’ and what ‘event’ on this more liberal dispensation, so that one does not know, for example, whether quarks, neutrinos, or flashes of lightning are to be admitted as (short-lived) things or rejected as events. Moreover, even where we are dealing with non-scientific sentences of the everyday world, the reist’s literal renderings are not in every case so easy to come by. What, for example, is to count as a ‘literal’ rendering of a judgment like: ‘John’s jump impressed the spectators’? Perhaps: ‘John jumped and impressed the spectators’. Yet it is far from clear that this rendering is even roughly adequate. John’s jump, after all, may have impressed the spectators, but not John himself. Or John may have jumped, and impressed the spectators, without it being the case that it was his jump by which they were impressed.26 Kotarbiński’s problem here results from the fact that there is as it were a selectivity of intentional verbs like ‘see’ or ‘think about’ or ‘be impressed by’. It seems that such verbs may relate their subjects to entities such as events, processes, images, contents, meanings, surfaces, boundaries, states

132

BARRY SMITH

of affairs, absences, and so on, in ways not accountable for exclusively in terms of any mere directedness to things. Such selectivity is characteristic especially of memory, which may as it were conceal from our present consciousness the things which serve as supports for events or circumstances remembered. Thus Harry may remember the intonation of Mary’s voice, yet he may have forgotten both Mary herself and the voice that had this certain quite specific intonation.27 Further problems arise for an approach of the sort sketched by Kotarbiński when we consider sentences apparently involving quantification over events or types of event (John danced the same jig twice), or when we consider relational or comparative sentences like Mary’s blush was redder than Susan’s, The beginning of John’s jump was more elegant than the end of Jack’s, and so on.28 Kotarbiński himself, of course, since he believes that somatistic reism is true, can countenance neither a selectivity of mind to non-things, nor the possibility of relations involving apparent non-things in ways which could not be cashed out satisfactorily in terms of corresponding things. Hence he is constrained to hold, for example, that when Harry remembers the intonation of Mary’s voice, then there is of necessity a sense in which he remembers Mary also, and that the precise content of his memory can be accounted for without loss of content in terms of his relation to this and other things. Later reists in the Polish tradition take a more relaxed view of such translation problems, conceiving reistic semantic analysis as of value only when confined to theses purely philosophical in character. Outside ontology, as Lejewski puts it, reistic semantics ‘loses its rationale’: There is not much point in avoiding abstract noun-expressions in disciplines of lesser generality. Elimination of onomatoids from final pronouncements is of paramount importance only if these final pronouncements are meant to be used in ontological arguments.29

Now there is, certainly, some justice to this, if it means that the reist is restrained from embarking on gratuitous attempts to reform the language of his fellows, language which must surely be in order as it is. What Lejewski has to say should not, however, be interpretated as implying that we may

ON THE PHASES OF REISM

133

properly ignore those forms of everyday and scientific language which pose prima facie problems for the would-be reist translator.

§5. Kotarbińskian Psychology As we have seen, Kotarbiński rejects the doctrine of mental contents propounded by his teacher Twardowski. Contents and images are, as Kotarbiński points out, commonly held to come into being when someone recalls something or dreams of something. The subject who dreams or remembers is then ‘ready to formulate various true judgments, allegedly pertaining to those images’. (1966, pp. 430f.) Brentanists such as Twardowski had defended the thesis that such contents or mental images enjoy an immanent existence ‘in the subject’ or ‘in a person’s head’.30 Kotarbiński, however, could not see how this ‘in’ is properly to be interpreted. Surely not spatially, ‘as though it referred to the nervous tissue in the brain?’ (And where, for example, would we locate such mental phenomena as the pain in a phantom limb?) Yet it seems equally inappropriate to regard mental images as located outside the brain, for instance where imagined external objects seem to be located. Leśniewski had been prepared to conclude from these difficulties that contents and images exist ‘nowhere’, a conclusion perhaps in the spirit of Descartes, with his view of res cogitans as unextended, and accepted also among the Brentanists.31 Kotarbiński, however, could permit himself no such radical departure from somatistic realism and concluded that immanent contents and images are not to be accepted as bona fide things at all. This conclusion he saw as being supported further by the fact that such putative entities are not three-dimensional. Thus they cannot count as ‘bodies’ as ordinarily understood, and this, for Kotarbiński, rules out their counting as things in any sense.32 But how, then, are we to cope semantically with our apparent references to images and other like phenomena? Here, again, Kotarbiński’s attack is two-pronged, both semantic and ontological. On the one hand he hopes, with Leśniewski, to ‘de-intensionalise’ psychological statements, to find means of converting such statements to extensional forms. On the other hand, however, he hopes to develop a reistic conception of the discipline of psychology itself, a conception according to which psychology would deal

134

BARRY SMITH

not with mental acts of hearing or thinking or desiring and with the contents of such acts, but rather with things of certain sorts – with the sentient person, the hearer, thinker, or desirer. That which sees and hears and desires is, Kotarbiński holds, identical with a certain organism (or at least with some part of the organism such as the brain or the system of nerve receptors).33 To think, then, is to be a thinking brain or body, a brain or body which, in non-reistic language, enjoys certain special states or processes of thinking. As Kotarbiński is himself careful to stress, this is not a materialist or behaviourist doctrine. For while he certainly holds that physics investigates all that there is, Kotarbiński does not suppose that all scientific statements about what there is will turn out to be statements of physics. As for Spinoza, so also for Kotarbiński, it is as if, in the case of sentient beings, one single substance is able to support two different systems of mutually incommensurable modifications. Physics describes how sentient organisms (and other bodies) move and how their particles are located. Psychology describes how sentient organisms think and feel.34 Suppose, however, that during some given period of time one and the same sentient organism is both thinking and jumping. The same thing, in such circumstances, is both a thinker and a jumper. Is not the reist left in such circumstances with no means in his ontology to distinguish between what are, surely, activities of different sorts? Clearly, he cannot solve this problem by appealing to the fact that different (mental and physical) predicates are applied to the thing in question, for the issue here is precisely that of establishing in virtue of what such different predications count as true, and to this end the reist has only things to which he can appeal. The problem cannot be solved, either, by appealing to any special understanding of the material ‘thing that thinks’ (which had been left indeterminate by Kotarbiński himself). For whichever concrete thing is fixed upon by the reist as that which thinks, be it the brain, the central nervous system, or some other proper or improper part of the organism as a whole, there will always be physical truths about the thing selected in relation to which the given problem will recur. Moreover, whatever the nature of the material thing that the reist puts forward as his candidate ‘thing that thinks’, it seems not logically excluded that two parallel consciousnesses should be realised simultaneously within it, after the manner of Siamese twins. We might then

ON THE PHASES OF REISM

135

have occasion to assert that consciousness1 is thinking this, while consciousness2 is thinking that, and then it seems that the reist – short of assuming special immaterial things – would have no way of doing justice to the presence of parallel thinking processes in the given case, for there are ex hypothesi no separate bodies which might here serve as subjects of the respective predications. What is not logically excluded seems thereby to be ruled out by the fiat of the reist’s linguistic predilections. Reism has consequences not only for the subjects of mental experiences, however, but for the objects of such experiences also. As already stated, the reist insists that that to which our experiences are related is in every case a thing. In everyday perception, as also in hallucinations, dreams and memories, we are typically presented with external things which seem to us to be coloured and shaped in this or that particular way. And in dreams and memories, as Kotarbiński puts it, we as it were ‘observe, though somehow in a secondary manner, things from our past environment, which seem to us to be such or another’. (1966, p. 431) This account will clearly face problems in connection with iterated reference to what is mental – dreams about dreams, for example – as also in connection with that peculiar selectivity of memory and other acts discussed in §4 above. Kotarbiński’s view, nevertheless, is that our mental experiences are in every case a matter of our being related in special ways to things. From this it follows that all (third person) psychological statements must have literal readings of one or other of the forms: A feels this: B, A experiences this: B, A thinks this: B, and so on – where ‘A’ stands in for the name of some sentient body and ‘B’ for words or phrases which answer the question ‘what?’: ‘What does John imagine?’, ‘What does John think?’, ‘What does John want?’, and so on. ‘B’ will stand, typically, for a ‘summary description of [A’s] surroundings made in extrospective terms’, and is supposed in every case to involve reference exclusively to things.35 A slightly different analysis may be required for statements like ‘my tooth aches’ or ‘I feel sick’. These may on the one hand be compared to

136

BARRY SMITH

statements like ‘my shoe is pinching’. Taken in this sense, ‘I feel sick’ would be formulated as ‘This is sickening’, ‘where the indicative pronoun would point to a certain region of the alimentary tract and adjacent parts of the body.’ (1966, p. 348) Quite often, however, the sense of ‘I feel sick’ is to signify ‘I experience a feeling of sickness’ and this is a statement which complies with the original Kotarbińskian scheme. It means ‘I experience this: it (my body) is sickening’ – where again, reference is made exclusively to things. But in virtue of what are sentences of the form ‘A feels...’, and so on, true? Perhaps we can express Kotarbiński’s view as follows. It is as if there are certain sui generis determinations of sentient bodies in virtue of which such bodies are directed in a quite specific way to things. It is of course not the case that the determinations in question could be somehow isolated, whether actually or in thought, in such a way that they could be examined in their own right. Yet they are not simply unknowable, either; for there exists the possibility of imitation, in the sense that one subject can think in a way which duplicates the thought-determinations of another. Such imitation is possible because our mental determinations characteristically express themselves physically in a range of familiar ways. Above all, there is an organic relationship between a subject’s feelings or thoughts and the kinds of things he says. Hence we can come to know the former indirectly, by coming to an understanding of the latter in a way which amounts to a (more or less perfect) duplication of those mental determinations which they characteristically bring to expression. Here Kotarbiński draws on the work on meaning and expression of his teacher Twardowski,36 as also on Gestalt psychological ideas concerning our knowledge of other minds.37 Strangely, he applies these ideas even to reflexive self-knowledge. We acquire knowledge of our own experiences, he holds, only by ‘self-imitation’,38 so that there may be a sense in which we do not know what we think until we hear what we say. Kotarbiński’s claim, then, is that we may come to know what another person experiences by allowing ourselves to be guided by his statements or by other overt behaviour in such a way that we come to imitate his experiences within ourselves. Reistically expressed, we can make ourselves think or feel (more or less) as the other person thinks or feels, by allowing ourselves to be determined psychically by the things he says. (The question-

ON THE PHASES OF REISM

137

begging nature of this idea becomes clear immediately if we ask ourselves in what respect we can speak of ‘likeness’ or ‘similarity’ here.) We try to interpret the word ‘experiencing’ as follows. It is merely an announcement of the imitation of the individual spoken of by the speaker, and it informs in a summary way in what respect he will be imitated; thus, that the individual spoken of will be imitated as looking, or listening, or exploring tactually, and so on (1935, p. 499).

For this to make sense in reist terms, therefore, it must be that our utterances themselves are in some extended sense imitations of the very psychic determinations they bring to expression.39 Thus in the general formula of the psychological statement ‘A experiences this: B’, the ‘B’ may be seen as an imitation in this extended sense by the one who makes the given statement of the relevant experience on the part of A. When I say, ‘John thinks this: 2 + 2 = 4’, ‘John feels this: they are playing badly’, ‘John doubts this: do angels exist?’, ‘John desires this: to be happy’, then I become a samesayer with the way John thinks or feels. And we can even generalize this formula so that not only a sentence, but any phrase referring to how a given person experiences, could be substituted for ‘B’. It might even be an inarticulate exclamation, so that a given psychological statement would be: ‘John experiences so: Oh!’ (1966, p. 428)

Kotarbiński’s remarks here will remind us of Davidson’s analysis of indirect discourse in his paper “On Saying That” of 1968. My assertion of ‘Galileo said that the earth moves’, on this analysis, is an assertion to the effect that Galileo said something, and my immediately succeeding utterance of ‘the earth moves’ makes Galileo and me samesayers: Galileo said that. The earth moves. Here it is only the first sentence, consisting of the name of a speaker, a two place predicate ‘said’ and a demonstrative pronoun, that is asserted. The second sentence is, as it were, merely exhibited or named. For Davidson, too, therefore, there is a sense in which the best we can do is to imitate

138

BARRY SMITH

(make ourself samesayers with) the speaker whose words we are reporting.40 A similar idea was incidentally advanced already by Leśniewski41 who considers an interpretation of expressions of the type ‘|— p’ in the language of Principia Mathematica as meaning: that which follows is asserted p. As Küng points out, it is an important feature of such devices that they allow us to talk about a sentence while employing to this end not a name of the sentence but (a token of) the sentence itself; that is, they allow us to avoid an ascent into the metalanguage ‘while at the same time obtaining benefits usually associated with such an ascent.’ (Küng 1974, pp. 243f.) A similar device can be used also to avoid an ascent into set-theoretical language: instead of ‘The set of men is identical with the set of featherless bipeds’, we can say: ‘The following two items are extensionally identical: man, featherless biped’. As Küng argues, it is a device of this sort that lies at the basis of Leśniewski’s understanding of the quantifiers.42

§6. The Aristotelian Concept of Thing Kotarbiński started out, in the Elementy, from the common-sense idea of thing as physical body. He drew, in particular, on the clarification of this idea that was set forth by Aristotle in his treatment of ‘first substance’ in the Categories and in the Metaphysics. Thus at the beginning of the Elementy we read: it is in Aristotle that we can trace the distinction, within the category of things, namely, of first and second substances. Those second substances, universals, are the first to fall victim to eliminating analysis as carried out by nominalism... On the other hand, first substances, things in the primary sense of the word, and for us simply things, fared in exactly the opposite way, since the entire reduction of categories [takes] place precisely to their benefit. (1966, p. 55)

What, then, are the marks of things or first substances as Aristotle conceives them?43

ON THE PHASES OF REISM

139

(i) They are, first of all, individual. A substance is a ‘this’, it is ‘one in number’. (Cat., 3 b 10) (ii) They are not ‘predicable of a subject’ nor ‘present in a subject’. (Cat., 2 a 11–13, Met., 1017 b 10–14, 1028 b 35f., 1029 a 1) (iii) They are that which can exist on their own, where accidents require a support from things or substances in order to exist. First substances are prior in all senses: in definition, in order of knowledge, and in time. (Met., 1028 a 30ff.) (iv) They are that which serves to individuate the accident, to make it the entity that it is – a feature seen by Brentano as the most crucial element of the Aristotelian theory. (Anal. post., 83 a 25, Met., 1030 b 10ff., Cat., 2 b 1ff.)44 (v) They are that which, while remaining numerically one and the same, can admit contrary accidents at different times. (Cat., 4 a 10) (vi) They are able to stand in causal relations. (Met., 1041 a 9) (vii) They are ‘one by a process of nature’. A substance has the unity of a living thing. Hence it enjoys a certain natural completeness or rounded-offness, both in contrast to parts of things and in contrast to heaps or masses of things. (Met., 1040 b 5–16, 1041 b 12, 1041 b 28–31, 1052 a 22ff., 1070 b 36–1071 a 4, Cat., 1 b 5)45 Hence also, for Aristotle, a thing is that which has no actual but only possible parts. (Met., 1054 a 20ff.) A part of a thing, for as long as it remains a part, is not itself a thing, but only possibly so; it becomes an actual thing only when it is somehow isolated from its environing whole. In this sense (and also in others) the substance is the bearer of potentiality, and it is at this point that we should have to list those marks of substance which flow from Aristotle’s hylomorphic theory, and from his theory of act and potency. There are further marks of substance, less easily documented in Aristotle’s texts since they were taken entirely for granted in Aristotle’s day. These are above all: (viii) A substance is independent of thinking, a part of nature – where no Greek would have understood what is meant by ‘independent of thinking’. (ix) A substance is that which endures through some interval of time, however small. This means, firstly, that things exist continuously in time (their existence is never intermittent). But it means also that there are no punctually existing things, as there are punctual processes or events (for

140

BARRY SMITH

example beginnings, endings, judgings, decidings, and instantaneous changes of other sorts).46 A thing is also typically such as to endure for such a length of time that it may acquire a proper name for purposes of reidentification.47 (x) A substance is that which has no temporal parts: the first ten years of my life are a part of my life and not a part of me. As our ordinary forms of language suggest, it is events and processes, not things, that have temporal parts. Even leaving aside, now, the passages where Kotarbiński explicitly allies himself with Aristotle,48 the focal instances of the concept of thing made prominent in the Elementy make it clear that he had intended to follow Aristotle in almost all of the above. A body, as we have seen, is bulky and lasting and such as to offer resistance. Further marks of bodies distinguished by Kotarbiński are: ● They are three-dimensional. ● They are all and only those entities that can be investigated by science; every object is ‘knowable in principle’. ● They enjoy essential perceptibility, and are further characterised by the fact that they all exert influence upon perceptible objects. ● They are at a definite place (that is, they are at a specified spatial distance from certain perceptible objects), and at a specified time (that is, they are at a specified temporal distance from certain perceptible objects).49 Each of these marks is perfectly in conformity with the Aristotelian view expressed above. Kotarbiński’s most important departure from Aristotle, in the Elementy at least, is in regard to (vii). For Kotarbiński – almost certainly under the influence of the Leśniewskian conception of sets as concrete wholes – rejected the thesis that things must in every case be unitary, so that he counted as things also masses and quantities of things and even non-detached thingly parts. Bodies of air, swarms of bees, the solar system, are ‘compound bodies’, in Kotarbiński’s terms, as also are society, nation, social class and all other institutions.50 A lack of sensitivity to the distinction between things, masses, and parts seems to have been shared by a number of Polish philosophers, and may in some degree reflect the fact that the Polish language, with its lack of articles, distinguishes less clearly than other languages between mass and count

ON THE PHASES OF REISM

141

nouns or between mass and count uses of the same noun (‘Polish logicians eat little orange during War’). One (Austro-)Polish philosopher who did not allow himself to be swayed in this way by what is, after all, a peculiarity of language, was Twardowski, who follows Aristotle in insisting that what he calls ‘objects of presentation’ are characterised in every case by the fact that they are integrated wholes, a thesis he extends even to the objects of general presentations such as Triangle, Square, Lion, etc.51 Leśniewski, in contrast, goes so far in rejecting the idea that to be a thing an object must in some sense form a natural unity, that he accepts a principle of the arbitrariness of thingly boundaries. This principle is built into the axioms of his system of Mereology, which includes a theorem to the effect that if a and b are objects, then so also is their sum, irrespective of whether a and b are connected or contiguous or materially related in any way.52 Leśniewski does not, of course, deny that some objects (in his highly general sense) have a natural unity. It is merely that he does not see the need to introduce this concept of natural unity into his theories of Ontology or Mereology. The latter are theories dealing with what he holds to be more primitive notions – notions which would in any case have to be clarified before a rigorous treatment of ‘natural unit’ could be attempted.

§7. Truth, Correspondence and Leśniewski’s Ontology How, now, given his essentially Aristotelian ontology of things, does Kotarbiński cope in the Elementy with the problem of accounting in correspondence-theoretic terms for the truth of sentences such as ‘John is jumping’? Sentences of the given sort are analysed, first of all, as what Chisholm has called concrete predications, expressing relations between things.53 Thus ‘John is jumping’ is analysed as a sentence of the form: ‘John is a jumper’, ‘John is red’ as: ‘John is a red thing’, ‘John desires apples’ as: ‘John is an apple-desirer’, and so on. In each case we arrive at a sentence containing two names of things joined together by the copula ‘is’,54 so that the things picked out by the names in question are to have the job of making true the relevant judgment.

BARRY SMITH

142

Remember, in all that follows, that ‘John is a jumper’ is to be understood as an analysis of the sentence ‘John is jumping’ (‘John is at present executing one or more jumps’). Thus we are to resist the natural tendency to understand nominals like ‘jumper’, ‘swimmer’, ‘bouncer’, as relating to a habitual or professional performance of the relevant activity. This tendency derives, of course, from the already mentioned fact that names are in normal circumstances used for purposes of re-identification; thus they presuppose some duration or recurrence on the part of what they name. ‘John is a jumper’ accordingly analyses ‘John is, on this particular occasion, jumping’, where ‘is’ expresses a real continuous present. The thing picked out by ‘John’ seems relatively easy to identify, at least against the background of the broadly Aristotelian conception described above. But what, in the light of this conception, are we to make of the thing picked out by ‘a jumper’? And what is the relation between John and a jumper that is expressed by the copula ‘is’? Our first port of call, given the strong influence exerted by Leśniewski on Kotarbiński’s (formally much less sophisticated) ontological views, is Leśniewski’s own system of Ontology, a theory built up on the basis of Leśniewski’s system of Protothetic or ‘theory of deduction’ by the addition of the new primitive term ‘is’ and the single axiom: ∏ab [a is b ↔ ∑c (a is c & c is b)]. Colloquially: a is b if and only if, for some c, a is c and c is b.55 Here ‘a’, ‘b’ and ‘c’ are any expressions belonging to the category name. This means, as Leśniewski sees it, that they may be either: (1) ordinary singular designating names or nominal expressions like ‘Ronald Reagan’ or ‘the British Prime Minister’; (2) shared or ‘general’ names like ‘philosophers’ or ‘apples in Vermont’. (3) fictitious or empty singular names like ‘Pegasus’ or ‘the largest prime number’; or (4) fictitious or empty general names like ‘sirens’ or ‘fates’.56

ON THE PHASES OF REISM

143

All such expressions belong to a single category, Leśniewski argues, since whether a name like ‘man at the door’ is singular or shared or empty depends on the factually existing state of the world, and so cannot be regarded as basic from the point of view of logic.57 From this, however, it follows also that we must admit as ‘names’ expressions like ‘jumper’, whose number is in a certain sense indeterminate. The axiom of Ontology, now, lays down simply that for ‘a is b’ to be true, it must be the case that every a is b and that exactly one object is a. It is not difficult to show, on these terms, that if both ‘a’ and ‘b’ are singular and designating, then ‘a is b’ is deductively equivalent to ‘a = b’. But now, applied to what has now become the Polish-sounding sentence ‘John is jumper’, this analysis of ‘is’ tells us only that, if this sentence is true, then ‘John’ must be a singular designating name and ‘jumper’ a designating name designating (possibly inter alia) what ‘John’ designates. From this point of view it becomes clear that the system of Ontology is in fact not an ontology at all (a theory of the different types of being). Rather, it is a theory of names, as is reflected in Kotarbiński’s use of the expression ‘calculus of names’ for what Leśniewski called ‘Ontology’. More precisely, it is a theory of the relations of designation that hold between singular, shared and empty names on the one hand and objects (of whatever variety) on the other. It reflects a concern, therefore, not with problems of ontology, or metaphysics, but with the issues that arise when one allows ‘a’ and ‘b’ to stand in not merely for singular terms as straightforwardly understood but also for any expressions within the wider category thus defined. To put the matter another way, Ontology may be seen simply as an extension of the theory of identity to cope with a somewhat liberal view of what may count as ‘name’, so that the absolute universality of ‘=’ is inherited by the new Ontological ‘is’. This makes it compatible with any ontology formulable by means of expressions belonging to the given category. As Woleński puts it, ‘Ontology is metaphysics free.’ (1987, p. 175) If, now, we return to our sentence ‘John is jumper’ in the light of the Leśniewskian analysis of ‘is’, then it would appear that two alternative readings present themselves, according to whether we take ‘jumper’ as singular or plural. On the first alternative, ‘John is jumper’ will be equivalent to the more English-sounding ‘John is identical with a jumper’, so

144

BARRY SMITH

that the referents of ‘John’ and ‘a jumper’ will be one and the same. Now ‘John’, as we normally suppose, designates an enduring object, subject at different times to contradictory determinations (he is now jumping, now not). If, therefore, the referent of ‘a jumper’ really is to be identical to the referent of ‘John’, then ‘a jumper’, too, must designate something that endures, so that ‘John’ and ‘a jumper’ would be merely two different ways of referring to one and the same ordinary continuant. On this account, however, the truth-maker of ‘John is jumping’ would differ in no wise from the truth-maker of ‘John is John’, and this is an outcome which surely flies in the face of our intuitions to the effect that one or more jumps must somehow be involved in making true the given sentence. If ‘John’ and ‘a jumper’ are two different ways of referring to the same thing, then surely, our intuitions tell us, they refer to this same thing ‘under different aspects’. Leśniewskian Ontology is however so lacking in discrimination in its treatment of ‘things’, that it is able to take no account at all of such ‘different aspects’. What, then, as regards the second alternative, which would make ‘John is jumper’ equivalent to: ‘John is one among the jumpers’. Since from any sentence of the form ‘a is one of the bs’, one can infer within Ontology a sentence which might be rendered colloquially as ‘a is this b’, where ‘this b’ is a singular name for that individual b which a is, this second reading might seem to bring us back once more to the first alternative, which we have seen reason to reject. We may, however, be able to infer from ‘John is one among the jumpers’ also that there are jumpers to which John himself stands in the relation of similarity. And now, since John’s circle of similars qua jumper is different from his circle of similars qua thinker, this may enable the reist to distinguish separate truth-makers for ‘John is thinking’ and ‘John is jumping’ even in those cases where the two activities are performed simultaneously. From this it would follow that the family of jumping things contributes in some way to making it true that John, in particular, is jumping – a consequence which certainly goes beyond what Leśniewski himself had to say on these matters, but which nevertheless has advantages from the reist point of view (to the extent that it has been found acceptable by Lejewski). In particular, it enables the reist to distinguish the truth-makers of ‘John is jumping’ and ‘John is thinking’ even in those circumstances where jumping and thinking are simultaneous: other jumpers contribute to making true the

ON THE PHASES OF REISM

145

former sentence in a way in which they do not contribute to making true the latter.

§8. Time and Tense This is not quite all that can be said on Leśniewski’s behalf, however, and before returning to our discussion of Kotarbiński’s own ontological views it will be useful to look at the Leśniewskian treatment of the phenomena of verbal tense. Recall that the ‘is’ in ‘John is jumping’ is intended to express a real present tense. The Polish ‘jest’, on the other hand, for example in ‘Jan jest skaczący’ (John is jumping) – a form which sounds odd due to the absence in Polish of the continuous aspect – does not express a present tense, and this holds too of ‘Jan skacze’ (John jumps) and ‘Jan jest skoczkiem’ (John is a jumper). In and of itself the Polish ‘is’ is timeless. In order to mark the fact that the jumping is taking place at the moment, the speaker of Polish must add an explicit temporal index and say, for example, ‘Jan teraz skacze’ (John jumps now) or (more stiltedly) ‘Jan jest teraz skaczący’ (John is now a jumper). This timelessness, we now see, must be characteristic also of the ‘is’ of Leśniewski’s Ontology. This is first of all because Leśniewski insisted that the sentences of Ontology should be absolutely true, i.e. true independently of time and occasion of utterance.58 But it is also because, as already noted, the ‘is’ of Ontology is to enjoy absolute universality of scope; it is to be applicable to abstracta as much as to concreta, to objects past and future as much as to objects of the present. It is in fact the same timeless ‘is’ as that which we employ when we say, e.g., ‘Socrates is a philosopher’ or ‘3 is a prime number’ or ‘whales are mammals’. Ontology is not, however, restricted to ‘timeless’ sentences of the given sort. Return, for the moment, to ‘Jan jest teraz skaczący’ (John is now jumping). We should normally interpret the temporal index (‘now’ or ‘teraz’) in such a sentence as governing the verb. Given the universality of scope of Ontology, however, and of the category name with which it deals, it is open to us to allow such temporal indices to govern not the verb but the subject of the sentence.59 This yields sentences of the form ‘Johnteraz jest

146

BARRY SMITH

skaczący’ or ‘Johnnow is a jumper’ – sentences of a sort which make possible a new Leśniewskian reading of our original ‘John is a jumper’. ‘Johnnow’ is a name, like any other; but a name of what? We shall think of it, for the moment, as designating a phase of John, remaining neutral as to what exactly this might mean and recalling only that any view of phases as temporal parts of things would signify a departure from the Aristotelian conception of ‘thing’ outlined above. We shall for the moment presuppose only (1) that some of the phases of John are jumping phases, some not; and (2) that phases exist only for some (normally relatively short) interval (or instant) of time. Someone who asserts that John is a jumper, now, may be seen as asserting that a present phase of John is a jumping phase of John – with a timeless ‘is’, exactly as dictated by the conditions laid down by Leśniewski on the sentences of his Ontology. The notion of a present phase of John may be elucidated in turn as: a phase of John that is simultaneous with the utterance in question, i.e., in reist terms, with the relevant phase of the speaker.60 The advantage of a reading of this sort is that we now have no need to regard ‘a jumper’ as the name of an enduring object. The identity of the referents of ‘John’ and ‘a jumper’ is assured, rather, by the fact that ‘John’ itself has come to refer to an entity which enjoys a merely transient existence.

§9. Phases in the Development of Reism Does this really help, however, in understanding what it is that makes true the sentence ‘John is a jumper’? For what is this ‘phase’ of John that is both John and a jumper? There are, it will turn out, a number of crucial difficulties which we face in establishing what such phases might be. Most importantly, as already remarked, it seems that however this issue is decided the phase ontology will dictate a departure from the broadly Aristotelian conception of ‘thing’, in spite of Kotarbiński’s apparent assumption that this ontology represents a natural and inconsequential extrapolation of the reistic ontology of things.61 There are, be it noted, no phases in Kotarbiński’s Elementy, and that Kotarbiński held to a strictly Aristotelian view in this work is seen above all

ON THE PHASES OF REISM

147

in the fact that – as his examples show – he takes it for granted there that things may change, in the sense that what is true of a given thing at one time may be false of that same thing at another. In his paper of 1935, “The Fundamental Ideas of Pansomatism”, in contrast, Kotarbiński embraces the phase ontology seemingly without a second thought. Every object, he writes, is something corporeal or something sentient (or a whole consisting of such components). An example of something corporeal is: a watch of the trademark Omega No. 3945614 from 1st January, 1934 to 31st December, 1934, inclusive (or any of its parts – for instance, the minute hand from 5th March, 1934 to 7th April, 1934, inclusive). And an example of something sentient: I, from 8 o’clock to 1 o’clock on 20th March, 1935 (or any temporal portion of this object, e.g., I, from 9 o’clock to 10 o’clock inclusive on the same day). (1935, p. 488)

Hence at least one further stage needs to be added to the list of ‘stages’ in the development of reism given above. This consists in the transition from an essentially Aristotelian ontology on Kotarbiński’s part in the period up to 1931, to a quite different ontology of phases (or mixed ontology of phases and substances) in the years thereafter. How this apparently unconscious change of mind on Kotarbiński’s part came about can be seen if we look at the final section of Leśniewski’s work “On the Foundations of Mathematics”. Here Leśniewski begins by expressing his gratification that so many of his own views, especially in connection with the system of Ontology, had found support in Kotarbiński’s work.62 Leśniewski goes on to quote extensively from the Elementy, including the famous passage in which Kotarbiński compares Leśniewski’s work to that of Aristotle, thereby providing a retrospective justification for Leśniewski’s use of the term ‘Ontology’– a justification which Leśniewski himself was only too willing to accept.63 For a long time, Kotarbiński writes, the term ‘ontology’ has come to designate investigations of the ‘general principles of being’ conducted in the spirit of certain parts of the Aristotelian ‘metaphysical’ books. However, it should be noted that if the Aristotelian definition of First Philosophy, perhaps the main concern of these books, is interpreted in the spirit of a ‘general theory of

148

BARRY SMITH objects’, then both the word and its meaning are applicable to the calculus of names as conceived by Leśniewski. (1966, pp. 210f.)64

In the paragraphs which follow this discussion of Kotarbiński, however, Leśniewski goes on to consider a certain difficulty for Ontology posed by the colloquial reading of Ontological sentences of the form ‘a is b’, and it is in this context that he first introduces his notion of phase or ‘temporal segment’. Let us suppose, Leśniewski writes, that someone were to assert: (a) Warsaw is older than the Saxon Gardens (b) Warsaw in 1830 is smaller than Warsaw in 1930 (c) Warsaw in 1930 is Warsaw (d) Warsaw in 1830 is Warsaw. Then, taken together with the axiom of ontology, we can derive from these sentences the following assertion: (e) Warsaw in 1930 is smaller than Warsaw in 1930, which is absurd. In his response to this objection Leśniewski insists, first of all, that the expression ‘Warsaw’ be used consistently throughout. Either, he claims, it should be used to refer to ‘only one object having a definite time span, which at present we do not know’, in which case it has the sense of ‘Warsaw from the beginning to the end of its existence’. Or it should be used in such a way that it refers to indefinitely many different objects, so that it would be possible to assert “of ‘Warsaw from the beginning to the end of its existence’ as well as of ‘Warsaw in 1930’ and of ‘Warsaw in 1830’...that they are Warsaws.” Moreover, “it would be possible to say with complete generality that if some object is Warsaw, and some other object is a temporal segment of the first object, then the second object is also Warsaw.”65 On the first reading, now, which sees ‘Warsaw’ as a singular name, “it is not possible to call by the name ‘Warsaw’ any temporal segment or temporal ‘section’ of the unique Warsaw referred to”. In this case we shall be

ON THE PHASES OF REISM

149

able to assert neither (b) nor (c) nor (d). On the second reading, on the other hand, ‘Warsaw’ is a plural name, which means that we shall find it impossible to assert any sentence of the form (a). Only on the basis of some such sentence, however, can we infer the consequence (e). Whichever alternative is chosen, therefore, the supposed absurd implication can be avoided. We may infer, now, that Kotarbiński, wishing to keep in step with Leśniewski in this, as in other matters of Ontology, simply took over the notion of temporal phase, going so far as to accept arbitrary temporal phases of an object as of fully equal status with that object itself. Closer examination of the relevant passage reveals that Leśniewski himself is here more circumspect. Thus he accepts the greater appropriateness of talking not so much of the temporal segment of the rector of the University of Warsaw in January, 1923 but rather of the man (‘from the beginning to the end of the existence of this man’) who was in January, 1923 the rector of the University of Warsaw. Further, he has “the inclination to use the expression ‘Warsaw’ as a name denoting one object only”. Since, however, he is using ‘man’ and similar expressions to designate simply the relevant maximal phase (‘man, from the beginning to the end of his existence’), it seems that Leśniewski is even here embracing the phase ontology, though in a form which seeks to come to terms with the fact that the ontology in question threatens conflict with our ordinary usage.

§10. What are Phases? Certainly we refer quite naturally to Napoleon in his youth, to Nixon during the period of his presidency, to the later Wittgenstein, and so on. Normally, however, we take such expressions in their sentential contexts, as signifying for example that Napoleon himself was such and such in his youth. That is, we treat expressions like ‘in his youth’ as adverbial modifications of the relevant verb. The phase ontologist, in contrast, takes such forms of speech to sanction the view that there are special objects, Napoleon in his youth, Nixon during the period of his presidency, and so on, which are temporal parts of Napoleon and Nixon respectively (‘from the beginning to the end of their existence’). Objects are therefore seen as having temporal parts in just the way that they have spatial parts like arms and legs. Thus where common

150

BARRY SMITH

sense and Aristotle prefer a view according to which things (for example people) exist in toto in any given moment of their existence, the phase ontologist seems to condone a view according to which only the relevant temporal parts of things would exist in any given moment. He may thereby be driven to the view that temporal parts must be in every case instantaneous, for any temporal part of duration longer than a single instant would have just as little claim to exist in that instant as would the relevant temporal whole. Adoption of the phase ontology may thereby lead to a view of ordinary things as mere entia successiva, the separate ‘momentary slices’ of which would exist in successive instants of time (as, according to some philosophers, the world as a whole has to be recreated anew by God in each successive instant). An enduring thing, on this view, is a mere logical construction upon the various instantaneously existing entities that may be said to do duty for it.66 Alternatively, however, the phase ontologist may seek to understand ‘Johnnow’ as signifying John himself, exactly as understood within the Aristotelian theory, but restricted to some interval of time (t,t’) which includes the present moment.67 If, however, as is required by the Aristotelian theory, John exists in toto in every moment of his existence, then it must surely follow, according to a process of reasoning encountered already above, that John(t,t’) is in fact identical with John himself. To get round this problem the phase ontologist might seek to regard ‘John(t,t’)’ as referring to John as he would have been had the universe (conceived as being in other respects identical to the actual universe) begun at t and ended at t’. (‘Napoleon in his youth’, on a view of this sort, might be understood as referring to Napoleon as he would have been had he ceased to exist on the point of reaching maturity.) This, however, would make of phases merely possible existents. It would leave us in the dark as to the referent of ‘John(t,t’)’ in this, the actual world and it would tell us nothing as to the relation, if any, between John(t,t’) and John himself. Leśniewski’s own motivation in introducing the notion of temporal segment seems to have derived on the one hand from his timeless conception of truth, and on the other hand from those physical theories (of ‘spacetime’, ‘world lines’, and so on) which have grown out of the idea that there is a certain analogy between the spatial and temporal dimensions of the entities treated of by physics.68 We shall assume, then, that on the

ON THE PHASES OF REISM

151

Leśniewskian view we are to regard each object as a four-dimensional whole, capable of being cut up into parts in any of its four dimensions. Phases result when objects are sliced in the temporal dimension. Why, now, must such a view embody a conception of the way in which the spatiotemporal world is parcelled into separate entities that is in conflict with the Aristotelian ontology of things? To answer this question we must understand what it is for an object to change (to admit contrary accidents at different times). Consider, to this end, the following passage from Zemach’s important paper “Four Ontologies” of 1970: An ontology may construe its entities as either bound or continuous in time and in space. An entity that is continuous in a certain dimension is an entity that is not considered to have parts in the dimension in which it is continuous. It can be said to change or not to change in this dimension, but what is to be found further along in this dimension is the whole entity as changed (or unchanged) and not a certain part thereof. The opposite is true of an entity’s being bound. If an entity is bound in a certain dimension, then the various locations along this dimension contain its parts, not the whole entity again. (Zemach 1970, pp. 231f.)

Thus the Aristotelian substance ontology is an ontology which sees substances as, in Zemach’s terms, continuant in time and bound in space. We see the same substance again on successive occasions, not a different slice thereof.69 The four-dimensionalist phase ontology, in contrast, is one which sees entities as bound both in space and in time, i.e. as having both spatial and temporal parts. Entities so conceived are excluded entirely from change. That a four-dimensional whole has red phases and green phases no more signifies a change, than does the fact that my pen is at one end red and at the other end green. Interestingly, now, the term proposed by Zemach for the fourdimensional wholes that are accepted by Leśniewski is the term event: An event is an entity that exists, in its entirety, in the area defined by its spatiotemporal boundaries, and each part of this area contains a part of the whole event. There are obviously indefinitely many ways to carve the world into events, some of which are useful and interesting (e.g. for the physicist) and some of which – the vast majority – seem to us to create hodge-podge collections of no interest whatsoever. Any filled chunk of spacetime is an event... When philosophers and physicists talk about spatiotemporal worms, about point-events, or about world-

152

BARRY SMITH lines, when they describe material things as ‘lazy processes’ and refer to spatial and temporal slices of entities, they are using the language of this [event] ontology. (Zemach 1970, p. 233)

Now it is no mere terminological matter to suggest that the phase ontology brings us close to an ontology of events. Indeed, Leśniewski himself, while critical of the specific formal treatment of the event ontology that is given by Whitehead,70 seems to have been not too negatively disposed to the idea that objects and events may constitute a single category. The whole tone of Kotarbiński’s Elementy, however, is precisely counter to an outcome of this sort, and there is not a little irony in the fact that Kotarbiński (like Lejewski in our own day) sees no incongruity in doing away with events via ‘onomatoids’, and then (apparently) resurrecting them via temporal parts. Certainly Leśniewski does not countenance anything like the dissolution of things into events or processes that was envisioned by, say, Heraclitus, Schopenhauer or Bergson. For not every four-dimensional whole is such as to count as a ‘thing’ from Leśniewski’s point of view. His reasoning seems to have been, rather, that it is possible to restrict the totality of fourdimensional wholes in such a way that the resulting ontology will remain more or less in keeping with our presuppositions concerning things or concreta. Thus, we might say, a four-dimensional whole, before it can be admitted by Leśniewski into his ontology, must satisfy the two-fold condition that it be (1) ‘bulky and lasting’ (i.e. extended in all its four dimensions) and ‘such as to offer resistance’, and also (2) such that all its (bulky and lasting) parts are resistant in the relevant sense.71 It is, however, far from clear that such conditions can of themselves suffice to transform an ontology of four-dimensional wholes into an ontology of ‘things’ in the sense of the reist. Indeed, there is a suspicion that they involve a surreptitious smuggling in of the goal to be achieved (not least because the phrase ‘offers resistance’ seems to belong still to the language of continuants or substances). Not merely does the phase ontology have no room for change. It precludes also any account of what we might refer to as the ‘stability’ of enduring substances. As W. E. Johnson points out, the ontology of fourdimensional wholes springs from that post-Humean doctrine which regards change as fictitious and substitutes for it ‘merely differently characterised

ON THE PHASES OF REISM

153

phenomena referred to non-identical dates.’ It may be that for most scientific purposes ‘no more transcendental conception than that of a whole constituted by the binding relations of time and space is required; and hence the philosophers who reject the conception of a continuant are satisfied to replace it by the notion of such an extensional whole.’ What, however, is to explain, on this account, the stability of that spatiotemporal nexus which connects, for example, the successive ‘phases’ of a living organism? As Johnson argues, a mere succession of processes ‘offers no explanation whatever of what in objective reality determines the stability of any given nexus.’ (Johnson 1924, pp. 100f.)

§11. In Defence of a Bicategorial Ontology There is an assumption running through the thought of Leśniewski, Kotarbiński and their followers, as also through that of Zemach, to the effect that the most worthy aim of the ontologist is that of producing a monocategorial ontology – and more generally of demonstrating that one or other sort of eliminative reduction can be achieved. A more natural resolution of the problems raised in our reflections on time and change, and on tensed predications like ‘John is jumping’, is achieved, however, if we abandon this concern with reduction and embrace instead a shamefacedly bicategorial ontology of things and events, the latter being conceived as changes in things, as dependent particulars (after the fashion of, say, accidents in the category of action as these are conceived in the Aristotelian ontology).72 Events, we can say, occur in or between things, and are no less individual than these things themselves. They may be instantaneous or extended in time, and in the latter case they have temporal parts which are themselves events. On this basis we can now go on to distinguish clearly between a thing, on the one hand, as something that is given in toto from the very first moment of its existence, and the ‘life’ or ‘history’ of this thing on the other hand, as a certain type of complex event, bound up inseparably with the thing whose history it is. The purported temporal parts of things will then turn out to be parts of such complex events, so that the cleavage between John as child and John as adult can be recognised, commonsensically, as a cleavage not in John, but in his life or history. Note, incidentally

154

BARRY SMITH

that there is no comparable move in regard to spatial parts. We cannot say that these are really parts of a substance’s shape, for example, or of the space a substance occupies. Clearly, the ontology of things and events provides a peculiarly simple account of what makes ‘John is jumping’ true. More interesting, perhaps, is the fact that this same ontology can account for certain properties of the linguistic phenomena of verbal aspect,73 properties which are important for us here in that they reflect a parallel on the side of the verb to the opposition among nouns between ‘mass’ and ‘count’ (as for example between ‘sugar’ and ‘snow’ on the one hand, and ‘tiger’ and ‘ox’ on the other). The former correspond to verbs of progressive and continuous aspect (‘John knows how to jump’, ‘John’s been jumping all day’), the latter to verbs of achievement (‘John jumped over the ridge’, ‘John just jumped to victory’). What this tells us in ontological terms is that the opposition between what is ‘unitary’ and what is ‘mass’ or ‘collective’ is to be found not merely in the realm of things but in the realm of events, too, though in an interestingly more complex form. For while events of reddening or exploding or whistling, as well as institutional affairs such as weddings, funerals and runnings of races, are all such that they have or could have temporal parts, they may also – by inheritance from the (moving and extended) things which support them – be extended in space. Hence they may participate in the opposition between ‘unity’ and ‘mass’ in two distinct dimensions: events may be spread out either in time, or in space, or of course in both. Consider, for example, the process of jumping. This is made up, we may suppose, of minimal unitary temporal parts – which we may call ‘jumps’ – comparable to ‘substances’ in the world of things. Each jump may then be analysed in turn as a continuum of bodily movements of a certain sort. We can distinguish, in the world of things, not only what might be called ‘substantial’ parts (lumps of sugar, molecules of water), but also atomic parts, which are marked by the fact that they have no parts of their own. This distinction, too, can be drawn in the realm of events, where we can distinguish on the one hand unitary events which take time but have no homogeneous sub-events as parts, for example judgings, decidings, and so on, and on the other hand events which are strictly punctual, such as beginnings, endings and instantaneous changes.74 Leśniewski’s Ontology

ON THE PHASES OF REISM

155

and Mereology, now, have shown themselves surprisingly adept at coping in a formally rigorous way with relations such as this in so far as these are manifested among things. Truth, however, is a relation which involves not only things and the names of things. It involves also verbs and that in reality to which verbs correspond, which is typically an event of one or other sort. Hence we can begin to understand why it is that the bicategorial ontology may be particularly suited to the task of giving an account of what makes sentences (particularly empirical sentences) true. For it allows us to take account of just those differences in reality which are reflected in language in the differences of verbal aspect (differences which are preserved, incidentally, even if we move over to a language shorn of tenses, of the sort that was favoured by Twardowski, Leśniewski, and other proponents of the ‘absolute’ theory of truth).

§12. Qualitative Extensions Return, however, to the properly reistic ontology and to our original pair of questions: what is the referent of ‘a jumper’ in ‘John is a jumper’, and what is the relation between John and a jumper that is expressed by the copula ‘is’? Our goal was to provide answers to these questions in monocategorial terms in a way that would tell us what things make ‘John is a jumper’ true. Kotarbiński’s Elementy, at a number of points, suggests an answer to this question that comes interestingly close to simulating the effects of the bicategorial ontology of things and events discussed above. This answer, which is nowhere to be found in Leśniewski, rests on the idea that ‘is’ in the given sentence expresses a special kind of relation of part to whole. ‘A jumper’ – or what might now be called ‘jumping John’ – is, on this reading, the name of a special sort of transiently existing thing, in which John himself is included as part. The idea here is that things may at certain times exist as it were in a raw state, but that they may on occasions extend themselves qualitatively, or become modified in certain ways (by what the tradition called ‘accidents’ or ‘modes’), so that John, for example, may on occasions become jumping John or cursing John or sleeping John, and so on. John himself will survive in each of the latter – though of course, because of the semantic restrictions imposed by Kotarbiński, we cannot

156

BARRY SMITH

isolate that which gets added to John to yield the various ‘qualitative extensions’ in which he may partake. It is more than anything else Kotarbiński’s examples that suggest this qualitative extension view. Thus he tells us that ‘it is obvious that only things are stimuli: burning flames, sounding strings, pressing solids, etc.’ (1966, pp. 434f.), and these are examples which seem to imply not only that Kotarbiński is intending to refer to things that can survive and acquire and lose accidental determinations in something like the Aristotelian sense, but also that the result of a thing’s acquiring an accidental determination may be a new thing, qualitatively extending the thing with which we began: a string becomes a sounding string, solids become pressing solids, a match becomes a lighted match, and so on. The qualitative extension view allows, moreover, a particularly simple interpretation of Kotarbiński’s views on psychology: a thinker is a body that is qualitatively extended in a special (deliberating, worrying, deciding) sort of way. There is not only jumping John, but also thinking John, hoping John, dreaming John, and so on. Even after the Elementy there are hints of this qualitative extension view in Kotarbiński’s treatment of words and sentences as things. Thus consider the following passage from a piece first published in 1954: a linguistic sign is for us a physical body, whether it is a graphic sign or an acoustic sign (in the latter case it is a certain amount of air vibrating in a specified way); thus it is a thing, and not a process in the sense of a specified changing of something. (1966, p. 399, emphasis added)

The same amount of air, considered as enduring thing, is vibrating now in this way, now in that, and it is a different ‘acoustic sign’ in the two successive cases. As such passages reveal, however, there are certain consequences of the qualitative extension view that run counter to our commonsense understanding of ‘thing’. For things, now, include not only Tom and Dick, but also Tom-the-jumper, Dick-the-thinker; not only quantities-of-air but also quantities-of-air-vibrating-in-a-certain-way. Note, however, that while qualitatively extended things may exist for a very short time, so that they typically do not acquire special (proper) names of their own for purposes of re-identification, it seems that all the qualitatively extended things seemingly admitted en passant by Kotarbiński do have

ON THE PHASES OF REISM

157

some duration, however short (so that their existence is never punctual, though it may be intermittent). This reflects, perhaps, the acceptability of names like ‘dying Jim’, ‘racing Tom’, etc., where ‘ending-his-process-ofdying Jim’ or ‘beginning-to-run-a-race Tom’ are ungrammatical. Of course not all complex names of the given sorts need be given the qualitative extension interpretation. Thus it may be that Leśniewski can entirely avoid this interpretation by means of his ‘phases’ (though again, it is difficult to see how this ploy will allow us to distinguish for example ‘jumping John’ from ‘cursing John’ when jump and curse are simultaneous). Moreover, there are cases of expressions of the form ‘—ing N’ or ‘—ed N’ where ‘N’ is a bona fide name but the ‘—ing’ or ‘—ed’ a merely modifying adjective which brings about a diminution or cancelling of the content expressed by ‘N’. Consider expressions like ‘missing arm’, ‘assassinated president’, ‘shattered vase’, ‘annihilated electron’, and so on.75 The qualitative extension view does however give us an elegant account of what makes sentences like ‘John is a jumper’ true, in a way that involves reference exclusively to things. John himself and his qualitative extension are not identical; the former is a part of the latter. Note, however, that there is no third thing which, when added to the former would yield the latter. Hence the usual mereological remainder principle fails to hold. The idea of a mereological theory in which the remainder principle is weakened, or even suspended, is far from incoherent, as a number of algebraic parallels testify.76 This failure of the remainder principle will however suffice to render the qualitative extension view unacceptable in the eyes of Leśniewski’s more devoted disciples.

§13. Kotarbiński and Brentano The interest of the view in question is for us largely historical. There are traces of the idea of qualitative extension in Aristotle, for example in the passages in the Metaphysics where Aristotle raises the question whether Coriscus and musical Coriscus are one and the same.77 Its first coherent defence is however given by Brentano:

158

BARRY SMITH Among the entities that have parts, there are some whose whole is not composed of a multiplicity of parts; it appears much rather as an enrichment of a part, though not as a result of the addition of a second part. One example of such an entity is a thinking soul. It ceases to think and yet remains the same soul. But when it starts to think again no second thing is added to that entity which is the soul. What we have here, then, is not like what we have when one stone is laid alongside another or when we double the size of a body... The substance is a thing and the accidentally extended substance is again a thing, though but a thing not wholly other in relation to the substance; hence we do not have that kind of addition of one and one that leads to a plurality. (Brentano 1933, pp. 53f., Eng. pp. 47f.)

Return, for the moment, to our list of ‘stages’ in the development of reism. Stage 5 in this development was provoked by a letter Kotarbiński received from Twardowski in 1929 on the publication of the first edition of the Elementy, in which Twardowski pointed out that the doctrine of reism had been propounded already some years earlier in a series of pieces dictated by Brentano towards the end of his life and appended to the second edition of his Psychologie.78 Kotarbiński, in response, came to see the need to add to his reflections on reism a certain historical dimension. Above all he began to stress the difference between his own pansomatist views and the reistic views he attributed to Brentano. In his paper on “Brentano as Reist” of 1966, Kotarbiński points out further that Leibniz, too, could be viewed as a precursor of reism, not only in the light of his monadology but also in reflection of his principle that all formulations containing names of abstract objects should be avoided. Kotarbiński sees himself, however, as the only ‘consistent and conscious’ somatist reist. He describes Leibniz as a ‘spiritualist reist’, i.e. as one who accepts souls or spirits as the only type of things. Brentano he describes as a dualist reist who accepted into his ontology both bodies (res extensa) and souls (res cogitans): ‘As a former priest, [Brentano] stopped at the threshold of somatism and never crossed it.’ (1966, p. 428.) We shall investigate below the extent to which this is an adequate account of Brentano’s reism and of its relation to that of Kotarbiński. Brentano, familiarly, distinguishes three sorts of ways in which a subject may be conscious of an object in his mental acts: in presentation, in judgments and in what he calls ‘phenomena of love and hate’, a category embracing feelings, emotions and all other value or interest phenomena. In presentation the subject is conscious of the object, has it before his mind,

ON THE PHASES OF REISM

159

without taking up any position with regard to it. The object is neither accepted as existing nor rejected as non-existing, neither loved as having value nor hated as having disvalue. A judgment arises when, to this simple manner of being related to an object in presentation, there is added one of two diametrically opposed modes of relating to this object, which we might call acceptance and rejection or ‘belief’ and ‘disbelief’. A judgment is in effect either the affirmation or the denial of existence of an object given in presentation, so that Brentano embraces a view according to which all judgments are reducible to judgments of existential form.79 A positive judgment is true if the object of the underlying presentation exists; a negative judgment is true if this object fails to exist. The early Brentanian ontology of ‘things’ or ‘objects’ arises, now, when one turns from the psychology of presentation to an investigation of the non-psychological correlates of presenting acts. ‘Object’ is accordingly to be understood as: ‘possible correlate of presentation’; it is a term whose meaning we can understand only by reflecting on the meaning of the term ‘presentation’, the latter term itself being such that we grasp its meaning directly on the basis of our own intuitive experience of our acts of presentation.80 According to Brentano’s earlier view, there exist non-real objects (‘entia rationis’, later called ‘entia irrealia’) of various kinds – mental contents or ‘objects of thought’, universals, states of affairs, possibilities, lacks and so on – all of which can be given in presentation and affirmed or denied in judgment. This points to a distinction, accepted by the early Brentano and his followers, between the existence or non-existence of an object of presentation on the one hand, and its reality or non-reality on the other. Thus what exists (for example values or universals) need not be real, and what is real (for example centaurs or chunks of wooden metal, and even the objects of simple acts of sensation) need not exist. These two oppositions, Brentano held, are independent of each other, and only the former is involved directly in the correctness or incorrectness of a judgment.81 Later, however, Brentano moved to a view according to which ‘reality’ and ‘existence’ would be equivalent, so that everything which exists is an ens reale. Brentano’s change of mind occurred, in fact, during the period when Kotarbiński was studying in Lvov. However, it was initially made known only to Brentano’s closest associates, so that we can rule out any

160

BARRY SMITH

influence of Brentano’s later view on Kotarbiński via his teacher Twardowski. Kotarbiński’s and Brentano’s thing-ontologies may however have a common source. Thus it is noteworthy that both views arose, in part at least, in reaction to certain apparent ontological excesses of Brentano’s students, not least of Twardowski himself, a reaction which led in both cases to a reversion to an ontology rooted effectively in the Aristotelian conception of thing or ‘first substance’. Moreover, as Ingarden pointed out already in the early thirties, there is a sense in which the roots of both Brentano’s and Kotarbiński’s reism are present already in Brentano’s own earlier existential theory of judgment, since – when once ‘Sachverhalte’ or ‘facts’ have come to appear suspicious – this has the effect of reducing each judgment to a form which asserts either the existence or the non-existence of some object:82 It is not that the being of A must come into being in order for the judgment ‘A is’ to be transformed from one that is incorrect to one that is correct; all that is needed is A. And the non-being of A need not come into being in order for the judgment ‘A is not’ to be transformed from one that is incorrect to one that is correct; all that is required is that A cease to be. And if only this happened and nothing else... would there not be in this fact alone, which relates to what is real, everything that is needed for the correctness of my judgment? Without doubt... And thus the doctrine of the existence of such non-things has nothing whatever in its favour. (Brentano 1930, p. 95, Eng. p. 85)

It is important, however, to be clear as to the precise nature of the respective views of Brentano and Kotarbiński. Brentano came to believe that all objects belong to a single category, the category of ens reale or Realitäten.83 The fact that he sometimes uses the word ‘thing’ [Ding] to refer to the entities in this category of itself tells us little as to the extent to which he shared with Kotarbiński tenets of the latter’s reism. Certainly there are a number of sometimes striking similarities between their respective philosophies: They agree, first of all, on negatives: for both philosophers, only concrete individuals exist. There are no abstracta, no universals or general objects, no properties, classes, meanings or concepts,84 Brentano’s motive for rejecting all such entities being rooted in his conviction that all that exists is completely determined, down to lowest differences.

ON THE PHASES OF REISM

161

Secondly, they agree as to the importance of the dimension of semantic or linguistic analysis as a complement to ontology: the apparent grammatical form of an expression is not always its actual or ultimate form. They agree also in the view that this actual form is to be achieved by translation into the language of things.85 Thirdly, and most importantly, Brentano agrees with Kotarbiński (and against Aristotle) in allowing collectives of things to count as things. Organisms, for Brentano, are collectives in this sense. Further, neither philosopher takes seriously the requirement that things should form a unity: By that which is when the expression is used in the strict sense, we understand a thing...; a number of things taken together may certainly also be called a thing, though one must not suppose that the two parts of a thing taken together constitute an additional third thing. For where we have an addition, the things that are added must have no parts in common. (1933, p. 4, Eng. p. 16)

Similarly they agree in allowing parts of things to count as things. Hence both are ‘actualists’, in the sense that they believe that a part of something actually real is itself actually real even when it is a part. Aristotle, in contrast, may be referred to as a ‘potentialist’, in the sense that he holds that parts of things are as such only potentially real.86 Neither in Brentano nor in Kotarbiński, we can say, do we find any trace of the Aristotelian theory of act and potency and of the hylomorphic conception of substances to which it led.

§14. Formal vs. Material Ontology The points of disagreement between the two philosophers derive especially from the fact that Kotarbiński starts out with the idea that physical bodies are the prime examples of things, and sees ‘resistance’ or the ability to stand in causal relations as a distinguishing mark of the concept thing. For Brentano, on the other hand, the concept thing, a concept which grew out of his earlier psychological doctrine of ‘objects of presentation’ does not essentially have to do with the concept of causality (with the opposition between what is inert or energetic and what is non-inert or anergetic). And

162

BARRY SMITH

even though things as Brentano conceives them are not, except in special cases, psychological entities, there is nevertheless a sense in which even on this latter doctrine the term ‘thing’ is a psychological term. This is shown most clearly in the different arguments the two philosophers bring forward to defend their respective views. Brentano’s argument for the ontology of ens reale rests on the fact that the univocity of ‘presentation’ implies the univocity of ‘thing’.87 Kotarbiński’s argument for his own reistic ontology, in contrast, is negative in form, resting on the unacceptability (for a variety of reasons) of theses to the effect that there exist universals, facts, classes and the like.88 The Brentanian concept of thing is, we might say, a formal concept. It is, in other words, a concept capable of applying without reservation to objects in all material categories – since, for Brentano, objects in all material categories may serve as objects of presentation.89 And then: It doesn’t matter at all what word we use to refer to the concept which is common to all that is to be presented. Whether we speak of ‘thing’ or ‘entity’, it is enough that it represents a highest universal to which we attain by means of the highest degree of abstraction no matter where we look (1930, p. 108, Eng. p.96).

Aquinas, too, sees the concept of a thing or of what is real as the most general concept to which reason can attain,90 and a broadly similar view is present for example in Husserl’s Philosophie der Arithmetik, where the purely formal concept of an Etwas or ‘something’ lies at the basis of Husserl’s theory of number and counting. For Kotarbiński, on the other hand, ‘thing’ is a term of material ontology, to be understood by reflecting on specific sorts of examples of thing and on the meanings of terms like ‘bulky’, ‘extended’, ‘inert’, and so on, whose significance is confined to the region of physical bodies. Marty is in this respect closer to Kotarbiński than he is to Brentano, since he pursued energetically the idea that the concept of thing or ens reale should be confined to those entities which participate in causal relations (are ‘energetic’ in Marty’s terms).91 However, Brentano argues, in transforming the concept of thing into the concept of what is capable of standing in causal relations, Marty has “permitted himself to deviate from long-established usage”:

ON THE PHASES OF REISM

163

a term which has traditionally been the most simple and the most general of all our terms has hereby been transformed into a sophisticated thought-combination which has been a matter of controversy since the time of Hume. Given Marty’s sense of the term ‘thing’, we would have to say that according to Hume and Mill and many others, there are no things at all! (1930, p. 108, Eng. p. 96)

The opposition between physical and psychical things, too, is an opposition formulated in material-ontological terms, so that to describe Brentano as a ‘dualist’ is to misunderstand the formal nature of his views. Certainly Brentano accepts spiritual substances (souls) as possible objects of presentation. And he accepts three-dimensional bodies also. However, the psychological origins of Brentano’s views imply that it is not at all clear that he accepted as objects of presentation physical bodies in Kotarbiński’s sense. Thus material things, for Brentano, are not restricted to three-dimensional physical bodies. They embrace first of all ‘topoids’ of higher numbers of dimensions, which conceivably exist as it were alongside the more familiar three-dimensional bodies given in perception. Brentano’s concept of thing embraces further things of lower numbers of dimensions, above all boundaries (points, lines and surfaces). Note, however, that while Brentano does not rule out topoids of higher numbers of dimensions, he rejects the idea that three-dimensional bodies might turn out to be boundaries of fouror more-dimensional topoids. This is because a boundary can exist only as the boundary of the thing which it bound. Thus spatial and temporal points, on Brentano’s conception, never exist in isolation from the things, extended in time and space, of which they are the boundaries. A body, on the other hand, is a thing in its own right, which requires no other thing (except possibly God) in order to exist.92 Kotarbiński, in contrast, reflects not at all on the nature of boundaries. More generally his work, like that of Leśniewski, lacks any topological dimension; things are seen as being arbitrarily divisible and as arbitrarily conjoinable (without even the topological requirement of connectedness in the latter case). This is despite the strong tradition of topology in Polish mathematics, and despite the fact that – as is shown for example by Tarski and by Grzegorczyk (1977) – topological axioms can be added very easily to the axioms of Leśniewski’s Mereology.

164

BARRY SMITH

Brentano accepts also certain sui generis zero-dimensional things, which he calls souls. These have the capacity to comprehend intentionally things of all higher dimensions,93 a notion which recalls Aristotle’s dictum in De Anima (429 b 25ff., 430 a 14f.), to the effect that the soul is somehow everything, for its nature is to be able to know everything and therefore in a certain sense to include everything within itself. Like the Leibnizian monad, so also the Brentanian soul is unextended, and therefore not continuously many; yet it is for all that continuously manifold, comparable in this respect to the midpoint of a disc divided radially into segments of continuously varying colours.94 The two philosophers differ further in virtue of the fact that, for Brentano, not all things need be perceivable. Thus souls are not perceivable, or at least not directly: we can apprehend intuitively at most the activities of the soul (the soul as modally extended in certain ways).95 Further, topoids of greater numbers of dimensions would not be perceivable; and nor, either, would the empty spaces which Brentano came to accept at the very end of his life. The most radical difference between the two philosophers turns, however, on the fact that Brentano takes tense seriously in the sense that for him ‘exists’ is in every case synonymous with ‘exists now, in the present moment’, so that everything exists for Brentano only according to a boundary (einer Grenze nach). This means that every existing object is as it were punctual from the point of view of its temporal extension, though always in such a way as to depend for its existence on that which has just existed or on that which will exist as supplying the continuum which it bounds.96 For Brentano, then, it is as if the world of things is continuously annihilated and recreated anew with each successive passing instant. ‘Past thing’ and ‘future thing’ do not, therefore, refer to special kinds of things, but are modifying expressions, to be compared with ‘hoped for thing’, ‘imagined thing’ and so on. Our apparent references to the past and future are in fact in every case references to the present as set apart temporally either as future or as past from something else. This ‘something else’ refers, however, merely modo obliquo and embodies no ontological commitment (in much the same way that our reference to a believer in demons or in fates involves no ontological commitment to demons or fates).97

ON THE PHASES OF REISM

165

Brentano’s world is, therefore, in this respect, too, entirely different from that of Kotarbiński, for it is a world in which there exists only one instant of time (even if this time is continuously changing). Brentano in fact identifies what is real with what is subject to a certain continuous temporal transformation which is simultaneously a matter of existing in the present, ceasing to be future and becoming past.98 Kotarbiński certainly holds that all things exist in time. Yet he reflects very little on the peculiar ontological features of things not yet and no longer existing – though we may assume e.g. from his discussion of the ‘things of history’ that he accepts into his ontology also past and future things.99 The ontological views adopted by Brentano at the very end of his life take him even further from reism in the Kotarbińskian sense. Things are now conceived by Brentano as falling into two groups, which we might refer to as places and souls. Places may be ‘empty’ (lacking in all qualitative determination), or they may be qualitatively extended or enriched (filled by qualities) in different ways. A three-dimensional qualitatively extended place is called a body.100 Brentano had earlier adopted a Cartesian view of space according to which space and quality are mutually dependent. On his later view, quality is one-sidedly dependent on space: space can exist without quality but quality cannot exist except as the quality of some particular place. A somewhat counterintuitive consequence of this view is that any change of place or shape brings about the annihilation of the ‘body’ in question, so that bodies cannot move. Movement is, rather, the becoming modally extended in appropriate ways of a continuum of different places in continuous temporal succession, rather like a ripple ‘moving’ across the surface of a pond, no molecule of which is displaced in the horizontal. There is no Chisholm (qua material object), but only a continuous sequence of Chisholmy places, so that Brentano’s later ontology implies just that view which is at the basis of Leśniewski’s logical grammar – that there is a sense in which we need not distinguish between proper names and predicates. We can come to some understanding of the reasons why Brentano came to choose places as the ultimate non-psychic substances if we examine again the list of the marks of substance set out in §6 above. Places are (i) individual and (ii) neither ‘present in a subject’ nor such as to ‘require a support from things or substances in order to exist’. (iii) They can exist on

166

BARRY SMITH

their own, i.e. without being filled or qualified in any way. If, further, substance is identified with place, then it becomes clear why substances underlie accidents and do not themselves need accidents in order to exist (where it would seem that the organic substances canvassed by Aristotle would depend for their existence on processes of breathing, of metabolising, etc.). Places, from this point of view, come to appear similar in this respect to the later Aristotelian concept of prima materia – both are infinitely plastic in the sense that they can take on qualities ad libitum. They come close also to the undifferentiated Lockean I-know-not-what which would serve as the ultimate support for the qualities given in experience. Most importantly, as Brentano insists, places are (iv) the best possible candidate for the role of that which individuates the accidents by which they are filled. We can say that for the later Brentano a body is the accident (qualitative extension) of a place, and that a place is that which individuates one body from another. Two qualitatively identical things at different places are distinct, as Brentano sees it, only because their location is distinct.101 Further, (v) places can admit contrary accidents, being now filled by something red, now by something black, and they are also (viii) ‘independent of thinking’, and (ix) such as to endure through time. Only (vi), (vii) and (x) are less easily applied to the concept of place. As we have seen, however, the mismatch in regard to (vi) can be explained by pointing to the psychological origins of Brentano’s views, and in regard to (x) there is in fact no essential disagreement, since while Brentano denies that things/places exist as a whole at all times at which they exist, his view that they exist merely according to a boundary, a view quite different from the phase ontologist’s view that they exist only in one or other temporal part, is in fact in keeping with the views of Aristotle. Which leaves only (vii), which is indeed rejected with great force by Brentano – as also by Kotarbiński – since neither philosopher imposes on things a requirement of unity. From this point of view, however, the concept place recommends itself still more strongly as that in terms of which an account of material (non-mental) substance is to be provided, since to regard things as merely (differently qualified) places is precisely to guarantee that arbitrary divisibility and conjoinability of things on which both philosophers insist. So strongly, indeed, does Kotarbiński identify the divisibility of things into parts with their extension in space and time, that we may argue that

ON THE PHASES OF REISM

167

Kotarbiński, too, ought properly to have accepted the idea that things are ultimately four-dimensional volumes of spacetime – a view that has been found attractive by not a few contemporary philosophers. Note that in order properly to understand Brentano’s view of (nonmental) substances as places, and in order to make sense of its superficially absurd implication that substances cannot move, nor change their shape, it is important once again to recall the psychological origins of Brentano’s views. For Brentano had tended from the very start to view the world of transcendent objects as something like a sensory surface (as, say, the surface of the visual field). Objects come thereby to be seen as similar in many respects to the images reflected on a screen. They are capable of being demarcated as things and as parts of things, and they are capable of being presented as moving, yet in both cases we have to do not with autonomous properties on the side of the objects themselves, but with mere ascriptions of properties to the images we experience. It would take us too far afield to give a precise account of Brentano’s views of autonomous reality. Clearly, however, his thinking is at least in some respects comparable to that of Mach and Einstein, both of whom sought to cast off ‘metaphysical’ assumptions such as that of independent substance.102

§15. The Varieties of Reism For all the divergences between Kotarbiński’s pansomatist reism and the later formal ontology of Brentano, there is a clear sense in which they are proponents of a common approach to ontology. This approach is shared also by Leśniewski and his followers, as also by Quine, Goodman and other modern nominalists. It can perhaps be characterised as an approach which takes as its starting point in the construction of its ontology a view of things drawing equally on examples of quantities, masses or homogeneous collectives as on the unitary substances of the tradition. Thus it is contrasted with the approach to ontology of Aristotle,103 Leibniz, Twardowski and Ingarden, which takes its cue primarily from the unitary substance and from the individual accidents which may inhere therein. There are, interestingly, a number of different routes taken by the philosophers mentioned to the homogeneous collective view of things (‘homo-

168

BARRY SMITH

geneous’ because the distinction between thing and mass is held to reflect no fundamental ontological division). Thus Quine, for example, seems to have been inspired particularly by those physical examples (energy fields, liquids, gases) where arbitrary delineability does indeed seem to hold, as also by related considerations deriving from the semantic treatment of mass terms in natural language. Quine, like Brentano104 and Kotarbiński, sees masses as full-fledged even though possibly scattered individuals. Thus he regards as of no importance the difference between what is spatially continuous and what is spatially scattered,105 and indeed his general approach is to view every object as a four-dimensional section of the world, after the fashion of what Zemach calls ‘events’. Field-theoretic physics played a role also in inspiring the later Brentano’s view of things as accidents of places, as also in securing Brentano’s acceptance of topoids of higher numbers of dimensions.106 Brentano’s acceptance of the homogeneous collective view was however motivated principally by his early work on the psychology of sensation, and for this reason he may have resisted the idea that ‘thing’ involves as one of its marks the concept of resistance or inertia. Goodman, too, was provoked by considerations deriving from the psychology of sensation in developing his ontology of ‘individuals’ in The Structure of Appearance, and Quine was to some extent led to the homogeneous collective view by psychological considerations concerning ostension.107 Leśniewski, on the other hand, was brought to his version of the homogeneous collective view of things by formal considerations deriving from the general theory of part and whole and from his critique of the settheoretic paradoxes, though such formal considerations played a role of course also in the work of Goodman and Quine (as indeed in the work of Whitehead). Both the homogeneous collective ontology and the Aristotelian substance ontology are contrasted, now, with ontologies allowing general, abstract and non-temporal entities of various kinds. Thus they may be contrasted with the positions of, for example Bolzano and Frege, or with Platonist ontologies of sets or classes. Bolzano, Frege and the set-theoretical Platonists are, we might say, maximally liberal in the sense that they impose on the entities admitted into their ontologies none of the conditions of temporality, inertness, perceivability and so on that have concerned us in the

ON THE PHASES OF REISM

169

foregoing. Marty and the early Brentano are one degree less liberal than this, in that, while they admit into their ontologies entia rationis of various kinds, they insist that all such entities enjoy a strictly temporal existence. They thereby recognise a division among temporal entities between the real or energetic on the one hand and the non-real or anergetic on the other. The bicategorial nominalist ontology sketched in §11 above imposes the further restriction that all entities be not only temporal but also energetic. The bicategorialist does not, however, insist that all energetic objects must also count as things, since he holds that events in his sense may enter into causal relations. The later Brentano did however impose this further restriction, though at the same time he abandoned the requirement that all things must be ‘energetic’ in the sense of being such as to enter into causal relations. Kotarbiński can be said to have gone one step further than Brentano in insisting that all things are physical bodies. Neither Brentano nor Kotarbiński however lays any requirement of unity on the objects in their respective ontologies, as contrasted with Aristotle – qua ontologist of first substance – who does impose a requirement of this sort. All of which implies that there is a spectrum of gradually more restrictive positions, from the Platonism of abstract objects at the one extreme to the Aristotelianism of unitary substances at the other, a spectrum which may be represented as follows:108 All entities are Temporal Energetic Things Physical bodies Unitary

Platonism (Frege etc.) No No No No

Marty

Brentano

Kotarbiński

Aristotle

Yes No No No

Bicategorial Nominalism Yes Yes No No

Yes No Yes No

Yes Yes Yes Yes

Yes Yes Yes No

No

No

No

No

No

Yes

We could go further, and extend the chart by taking into account oppositions of other sorts, relating for example to the issue as to whether entities are or are not general,109 independent of mind,110 atomic,111 or such as to have temporal parts.112 In addition we could think more carefully about the different meanings of ‘unitary’, distinguishing for example the requirement of connectedness of parts, the requirement of spatial separateness from other

170

BARRY SMITH

entities, the requirement of functional interdependence of parts, and so on. We could investigate further the extent to which things may have parts which are themselves things – as an organism may include cells, chromosomes, genes, etc. as parts.113 Already as it stands, however, the chart will enable us to see the inadequacy of any simple-minded opposition between ‘reism’ on the one hand and ‘Platonism’ on the other. Thus it would be wrong to go along with Lejewski in his view that ‘Ontologists who oppose reism are believers in so called abstract entities’ (1979, p. 210114), a view dictated no doubt by the fact that the principal enemies of the homogeneous collective view in recent philosophy have been advocates of ontologies based on the theory of sets or of a more or less Platonistically oriented semantics.115 More recently, however, and especially with the bringing to light of hitherto neglected aspects of the Brentanian and Husserlian ontologies, it has become clear that reism has other, non-Platonistic opponents. Thus for example it would not be to move into the realm of abstract entities were one to embrace in one’s ontology events as well as things. For events may be accepted – as for example on Davidson’s account, and indeed on that of Whitehead – as bona fide individual entities existing in time and space and entering into causal relations with other events. More controversially, we may say that Brentano (and Husserl) have shown how we may cope in a non-Platonistic framework with those kinds of dependent, inseparable, divisive, interpenetrative parts which fall outside the purview of mereology as standardly conceived but which yet seem indispensable to an understanding of the structures of minds, and of the different sorts of relations between minds and other concrete objects.

Notes 1

First published in: Woleński 1990, 137–184, with kind permission of Springer Verlag. I am grateful to Audoenus Leblanc, Czesław Lejewski, Dieter Münch, Karl Schuhmann, Peter Simons and Jan Woleński for helpful comments on an earlier version of this paper. 2 See his Logic and Philosophy in the Lvov-Warsaw School, and compare also Schnelle 1982 and the papers by Schnelle in Cohen and Schnelle, eds. 1986. 3 Kotarbiński distinguished also an eighth stage, in which reism is no longer a theory but rather a ‘semantic problem’.

ON THE PHASES OF REISM 4

171

Twardowski 1894, p. 105, Eng. p. 100. 1900/01, 2nd Investigation, §11. 6 See Leśniewski 1913, p. 319, and also the summary of Leśniewski’s argument in Kotarbiński 1920, and Lejewski 1979, pp. 200f. As Woleński (1988) shows, Leśniewski was influenced here by Marty’s criticism of Husserl’s Platonism in his Investigations of General Grammar (1908), §71. 7 Smith 1989, 1988. 8 See Leśniewski 1914, 1927/31. It may be also that Leśniewski’s criticism of properties in his 1913 helped to provoke Kotarbiński’s initial nominalism. 9 Leśniewski 1927/31, p. 17, citing Cantor 1887/88, pp. 421f. 10 Leśniewski 1927/31, p. 18, citing Frege 1893, pp. 2f., Eng. p. 31. 11 Leśniewski 1927/31, p. 22. 12 1927/31, p. 24. 13 Leśniewski 1929, pp. 6, 78, Lejewski 1958, pp. 123f. 14 Interestingly, Tarski, at least in his early years, up to and including his paper on “The Semantic Conception of Truth”, agreed with Leśniewski in this (see esp. pp. 342f. of Tarski 1944). Tarski, be it noted, was never a formalist: Tarski and Leśniewski parted company rather because Tarski came gradually to accept the use of set theory and infinitistic methods in his work. 15 See Twardowski 1894, §§1–2. Leśniewski did not himself develop a theory of contents, since he held that the problems involved would be too difficult to allow him to achieve the appropriate degree of theoretical rigour. 16 See the detailed account in Lejewski 1979. Kotarbiński’s initial reaction to Ajdukiewicz’s criticism was in part inspired by Carnap. 17 Cf. 1966, p. 433. 18 Something similar would be required to exclude from the realm of things also certain sorts of events. Consider, for example, a rotation of a metal sphere. This rotation is extended in space and time and, again, it is such as to offer resistance. Note that what is established by the argument in the text is merely that the canonical reistic notion of thing is indeterminate in its application. The argument is not designed to show that one could not deal satisfactorily with colours (or three-dimensional shapes or masses of sound or heat) within the Leśniewskian framework. As Lejewski has suggested, just as Chronology and Stereology (theories of time and space) can in principle be obtained from Mereology by the addition of certain extra-logical constants, so it would be possible to conceive a discipline of Colourology, obtained by adding constant terms such red, blue, etc., and a relational predicate such as is the same colour as. 19 See my 1989 for an account of Twardowski’s theory of states of affairs. Later, however, Twardowski criticised theories of truth, such as that of Russell, which employed categories such as state of affairs or fact. 20 Met., 1011 b 25ff. Less neutral is Aristotle’s remark, somewhat later in the Metaphysics, to the effect that truth and falsehood depend ‘on the side of the objects on 5

172

BARRY SMITH

their being combined or separated, so that he who thinks the separated to be separated and the combined to be combined has the truth, while he whose thought is in a state contrary to that of the objects is in error.’ (1051 b 3, emphasis supplied.) 21 On this terminology of ‘making true’ see Mulligan, Simons and Smith 1984. The terminology has a number of advantages over the more usual talk of correspondence. It is disembarrassed, first of all, of all connotations of ‘copying’. It does not suggest that the relation between a sentence and that in virtue of which it is true would be a symmetrical relation. And it can cope with the fact that there may be more than one entity which makes or helps to make a given sentence true. Thus, in the simplest possible case, ‘I have a headache’, may be made true by my present headache (‘from the beginning to the end of its existence’), or by any phase of this headache overlapping with my present utterance, or by relevant states of nervous tissue upon which my headache supervenes. 22 Similarly when a person states the fact of London’s lying somewhere on the Thames, ‘he merely states, in a devious way, that London lies on the Thames’ and here – as Kotarbiński conceives it – no reference is made to facts or states of affairs or anything other than things. (1966, p. 429) 23 Kotarbiński 1966, pp. 52, 401, n. 4. 24 1966, p. 426. Compare Szaniawski 1977. 25 1966, pp. 330f., 426. 26 Similar difficulties arise for the proposed reistic translation of ‘Justice is a virtue of honest people’ by ‘Any honest man is just’ (Woleński 1987, p. 168). For it may be that all honest men are as a matter of fact just, though not in virtue of being honest. 27 See §3 of Mulligan, Simons and Smith 1984, for a discussion of this example. There is, notoriously, a parallel selectivity of ‘cause’. Consider for example a sentence such as: The fact that agreement was reached caused universal joy, which Kotarbiński (somewhat counterintuitively) wants to render as: All were overjoyed when they agreed. (1935, p. 491) 28 See Tegtmeier 1981 for an extended discussion of such cases and of the reasons why they seem to dictate an ontology richer than that of the reist. 29 Lejewski 1979, p. 206. A compromise position is put forward by Wolniewicz in his paper in Woleński 1990. Wolniewicz maintains that sufficient support for reism is provided by a demonstration that it is possible to reduce apparently non-reistic theories to theories having a reistic axiomatisation. Reism, on this basis, would thereby concern whole theories and not separate sentences. 30 See e.g. Brentano 1874, pp. 124–32, Eng. pp. 88–94; Twardowski 1894, p. 3, Eng. p. 1; Höfler 1890, §6. 31 See e.g. Marty 1908, p. 401. 32 Cf. Kotarbiński 1966, p. 342. From this it follows, too, that there cannot be psychic subjects if the latter are conceived as systems or sequences of contents or images: see Kotarbiński 1935, p. 493.

ON THE PHASES OF REISM 33

173

1966, p. 344. Cf. Kotarbiński 1966, p. 346; 1935, pp. 495f. 35 Kotarbiński 1966, p. 347; 1935, p. 499. 36 See Twardowski 1912 and the discussion in Smith 1989. 37 See Köhler 1947, pp. 128ff., Koffka 1935, pp. 655ff. 38 1966, p. 347. Similar notions are present also in the writings of Theodor Lipps on the notion of empathy: see e.g. his 1905. 39 A view employed especially by Marty as the basis of a theory of linguistic communication. 40 Davidson 1968, p. 108. 41 1927/31, p. 10. 42 Cf. Küng and Canty 1970. 43 Compare, for what follows, Novak 1963/64. 44 Cf. Brentano 1933, pp. 37, 108, 112, 131, Eng. pp. 37, 86, 88, 109. 45 See Met., 1042 b 15–32, for other kinds of unity. 46 See Ingarden 1964/65, vol. I, §§28f. 47 As is clear from Aristotle’s treatment of (ii), it is possible that the marks of the concept of substance may be established in part through considerations of the language we use to refer to substances themselves. See Met., 1029 b 13. 48 See e.g. 1966, pp. 326ff. 49 1966, pp. 327, 435, 342. Note that, like inertness, these marks are held by Kotarbiński to be incidental; that is, they do not affect the extension of the concept thing or body. 50 The latter consist of human beings standing in certain relations to each other – which is not, of course, to say that there exist entities called relations (of dependence, leadership, authority etc.), in addition to and as it were alongside the human beings themselves. The various elements of a given institution are somehow related one to another for example in the sense that some of them behave in such a way because the others behave in such a way. A similar treatment is offered by Kotarbiński for terms like ‘function’, ‘disposition’, and so on. As Kotarbiński would have it, “‘X has the function of typist in a bank’ means the same as: X systematically types letters according to the instructions of her superior” (1966, p. 490). 51 1894, pp. 88, 105, Eng. pp. 86, 100. Cf. Ingarden 1964/65, vol. I, p. 219. 52 Irrespective, even, of whether a and b exist at the same time. See Simons 1987 for criticisms of this and related aspects of Leśniewski’s Mereology. 53 Chisholm 1978, p. 199. 54 More generally we arrive at a sentence containing exclusively logical constants and ‘genuine names’. See Woleński 1987, p. 168. 55 We here leave out of account peculiarities arising from Leśniewski’s special reading of the quantifiers. 34

174 56

BARRY SMITH

Leśniewski’s concern to allow fictitious or empty names was almost certainly inspired in part by Twardowski, whose treatment of non-existence also influenced Meinong. See Twardowski 1894, pp. 30, 33, Eng. pp. 28, 30f. 57 Küng 1967, p. 111. We may conjecture that such apparent logico-grammatical distinctions as that between common and proper names, marked in English by the presence or absence of articles, were overlooked by Leśniewski, again, because of the lack of articles in the Polish language. 58 Here, again, he was almost certainly influenced in part by Twardowski. See the latter’s defence of absolute truth in his 1902. 59 Bolzano, too, in §45 of the Wissenschaftslehre, sees time-determinations as part of the subject, so that, as he puts it, ‘a pair of propositions such as “Caius is now learned” and “Caius was not learned ten years ago” turn out to have different subjects.’ 60 See Brentano 1976, Part II, Ch. VIII. 61 A similar suggestion is implicit also in Leśniewski’s work. See e.g. Sinisi 1983, pp. 57ff., quoting from the final section of Leśniewski 1927/31. Cf. also Lejewski 1982. 62 Sinisi 1983, p. 55. 63 See Woleński 1987, pp. 170f. 64 It should go without saying, in light of our discussion of Ontology in §7 above, that we do not share this estimation of the nature of Leśniewski’s achievement. 65 Sinisi 1983, p. 58, quoting Leśniewski 1927/31. 66 Chisholm 1976, pp. 98f. Woodger, who independently developed a phase ontology similar to that of Leśniewski, comes close to a view of this sort in his 1939. 67 A natural language reading of ‘Johnnow’ may involve a necessary indeterminacy in the precise extent of the relevant interval; since this indeterminacy can in principle be eliminated, however (for example by utilising the resources of a formal theory such as the ‘Chronology’ proposed by Lejewski in his 1982), we can ignore the matter here. 68 This analogy is of course very restricted (see e.g. Mellor 1981, pp. 66f., 128ff.). The acceptance of the concept of a world line in a four-dimensional continuum is moreover fully consistent with a continued belief in the ontology of things (Sellars 1962, p. 578). One might indeed go further and argue that the concepts of the four-dimensional ontology themselves presuppose the thing ontology for their coherent formulation, that, for example, the idea of a world line makes sense only if there is some identical thing that is tracked from one time-point on the line to another (see Simons 1987, pp. 126f.). See also Brentano 1976, pp. 296ff. for further criticisms of the view of time as a ‘fourth dimension’ of space. 69 See also the discussion of ‘existence in the present’ in Ingarden 1964/65, §30. 70 Sinisi 1966, summarising part of ch. IV of Leśniewski 1927/31. 71 This is to exclude, for example, the case where a bona fide bulky and resistant whole is united with, say, an empty volume of spacetime. 72 See Simons 1983 for a formal treatment of a view along these lines within a Leśniewskian framework.

ON THE PHASES OF REISM 73

175

See Mourelatos 1981; Galton 1984, Appendix II; Hoeksema 1985, ch. 6. It is above all Ingarden who has invested effort in the ontological analysis of these distinctions. See his 1964/65, ch. V. 75 See Twardowski 1894, pp. 12f., Eng. pp. 11f., and also his 1923. Twardowski’s theory was derived in turn from Brentano 1924/25, vol. II, p. 62, Eng. p. 220. See also Husserl 1979, p. 309, Marty 1884, pp. 179f., 1895, p. 34, 1908, pp. 518f., and Leśniewski 1927/31, p. 48, n. 78. There may be a remnant of this doctrine of modifying expressions also in Kotarbiński’s notion of ‘substitutive renderings’ or ‘onomatoids’ discussed above. 76 Thus, for example, just as we can have a pseudo-Boolean algebra with pseudocomplements (see e.g. Rasiowa and Sikorski 1963, pp. 52f.), so also we might distinguish a family of pseudo-mereologies. 77 Met. 1206 b 16; see also 1024 b 30, 1018 a 2, 1030 b 13. See also Suarez’s discussion of the ‘accidens concretum’ in his Disputationes Metaphysicae, XXXIX, s.1, n. 10–12 and cf. Cajetan, In De Ente et Essentia d. Thomas Aquinatis, §§153ff. 78 Cf. Kotarbiński 1966a, p. 461, Eng. p. 195; Brentano 1924/25, pp. 228–77, Eng. pp. 330–68. 79 1924/25, vol. II, pp. 56ff., Eng. pp. 213ff. (Brentano later adopted an amended version of this view: see his 1924/25, vol. II, pp. 164–72, Eng. pp. 295–301.) The topic of existential judgments was taken up by a number of philosophers in the years around the turn of the century and seems to have played a role also in inspiring the subject of Leśniewski’s dissertation under Twardowski in 1911. This includes a discussion not only of Brentano’s Psychologie but also of Twardowski’s Content and Object and of Husserl’s Logical Investigations. Leśniewski may also have been influenced by the Sketch of a Theory of Existential Judgments of Hans Cornelius, with whom he studied in Munich. 80 Cf. Smith 1990 for a more detailed discussion of this point. 81 1889, §§50, 55; Marty 1884, pp. 171ff.; Twardowski 1894, p. 36, Eng. pp. 33f. 82 Ingarden put this view to Kotarbiński around 1935 when the latter came to Lemberg to give a talk on Leibniz as a precursor of reism. (Personal communication of W. Bednarowski.) 83 He was followed in this by his later disciples, above all Oskar Kraus, Alfred Kastil and Georg Katkov, who took over from Kotarbiński the word ‘reism’ to describe the later Brentanian view. See Kraus 1937, pp. 268ff. 84 Brentano 1924/25, vol. II, p. 162, Eng. p. 294. 85 Cf. Brentano 1930, pp. 87–97, Eng. pp. 77–87; 1956, pp. 38–48; Srzednicki 1965, p. 48; Chisholm 1978. 86 See Met. Z, and the discussion in Smith 1987. 87 1924/25, vol. II, pp. 162, 213f., Eng. pp. 294, 321f.; 1933, p. 18, Eng. p. 24; 1956, p. 38. 74

176 88

BARRY SMITH

These negative arguments are supplemented by an appeal to the fact that the language of things is psychologically more natural. Thus one natural way of explaining the meanings of words is to eliminate substitutive terms: ‘Should we wish to explain to a child what the word “similarity” means, should we not show him in turn several pairs of objects which look alike?’ (1966, p. 423) 89 On the opposition between formal and material concepts see Smith 1981; on Brentano and formal concepts see Münch 1986; on formal vs. material ontology see Ingarden 1964/65, esp. §9. 90 Albert the Great defends a similar view of being as the simplest concept from which no further abstraction can be made, and according to Brentano (1930, p. 108, Eng. p. 97) a parallel notion is present also in Aquinas. 91 1908, §66. 92 See Brentano 1933, p. 108, Eng. p. 96; 1976, pp. 38, 95. 93 See Brentano 1933, pp. 158f., Eng. pp. 119f.; 1976, pp. 20, 120f. 94 See Brentano 1976, pp. 41ff. 95 Aristotle, too, seems to exclude immaterial substances from the realm of sensible substances. Cf. Met., 988 b 26, 1040 b 31ff. 96 See Brentano 1976, e.g. p. 37. 97 Brentano 1976, Part II, Section III. 98 Brentano 1976, Part Two, Section V. 99 See 1966, pp. 369f. Cf. also op. cit., p. 191, where Kotarbiński denies the suggestion that ‘is’ would be an abbreviation of ‘is now’. 100 Brentano 1933, pp. 35f., Eng. p. 36. 101 Note that times could not be candidates for the role of that which individuates one body from another (as was held for example by Lotze); for as we have seen, everything that exists is for Brentano such as to exist in the same time. 102 Some clues as to Brentano’s views are provided by the argument for the a priori impenetrability of bodies in his 1976, pp. 180ff. Brentano’s views in this respect seem to have been stimulated by the positivism of Auguste Comte: see Münch 1989. 103 Here and in what follows we consider Aristotle exclusively as an ontologist of first substance. 104 See Brentano’s treatment of continua in Part I of his 1976. 105 Quine 1960, pp. 90–110. 106 See the “Appendix” to Brentano 1933: “The Nature of the Corporeal World in the Light of the Theory of Categories”, and also Brentano 1976, p. 184. 107 Cf. e.g. Quine 1960, p. 52. 108 Again, the reader should bear in mind that it is by no means clear where Aristotle would have to be posted on this spectrum if the whole of his ontology were taken into account. 109 There is a sense in which Twardowski admitted general things. See 1894, §15, and also Ingarden 1964/65, vol. I, p. 219.

ON THE PHASES OF REISM

177

110

Ingarden admitted both autonomous and dependent things, examples of the latter being creatures of fiction. See 1964/65, §12 and ch. IX. 111 Is the ultimate furniture of the universe itself atomic? Do all entities have atomic parts? See Sobociński 1971, and also Bunt 1985, for an interesting treatment of these issues from the point of view of the semantics of mass and count expressions. 112 See, again, Zemach 1970. 113 See Woodger 1937, for an experiment in this direction. 114 Compare Lejewski’s assertion in his 1976 to the effect that events are non-material objects, so that anyone who admits events into his ontology would be a ‘Platonist’ on Lejewski’s reading of this term. This reading is no doubt derived from the fact that events, for Lejewski, are the referents of abstract nouns like swimming, falling, talking. 115 More generally, they have been advocates of what Woleński calls ‘standard predicate logic semantics’ in his 1987, p. 170.

References Ajdukiewicz, Kazimierz 1930. “Review of Kotarbiński” 1929, trans. in Kotarbiński 1966, 607–31. Brentano, Franz 1874. Psychologie vom empirischen Standpunkt, vol. I, Leipzig: Duncker & Humblot, reproduced with additional material as vol. I of Brentano 1924/25. Brentano, Franz 1889. “Über den Begriff der Wahrheit”, first publ. in Brentano 1930, 3– 29. Brentano, Franz 1924/25. Psychologie vom empirischen Standpunkt, 2nd ed., in 2 vols. ed. by O. Kraus, Leipzig: Felix Meiner. Eng. trans. ed. by L. L. McAlister, Psychology from an Empirical Standpoint, London: Routledge and Kegan Paul, 1973. Brentano, Franz 1930. Wahrheit und Evidenz, ed. by O. Kraus, Leipzig: Felix Meiner. Eng. trans. ed. by R. M. Chisholm, The True and the Evident, London: Routledge and Kegan Paul, 1961. Brentano, Franz 1933. Kategorienlehre, Leipzig: Felix Meiner. Eng. trans. The Theory of Categories, by R. M. Chisholm and N. Guterman, The Hague/Boston/London: Martinus Nijhoff, 1981. Brentano, Franz 1976. Philosophische Untersuchungen zu Raum, Zeit und Kontinuum, ed. by S. Körner and R. M. Chisholm, Hamburg: Felix Meiner. Eng. trans. Philosophical Investigations on Space, Time and the Continuum, by B. Smith, London and Sydney: Croom Helm, 1987. Bunt, Harry C. 1985. Mass Terms and Model-Theoretic Semantics, Cambridge: Cambridge University Press.

178

BARRY SMITH

Cantor, Georg 1887/88. “Mitteilungen zur Lehre vom Transfiniten”, as repr. in Cantor’s Abhandlungen mathematischen und philosophischen Inhalts, E. Zermelo, ed., Hildesheim: Olms, 1966, 378–439. Chisholm, Roderick M. 1978. “Brentano’s Conception of Substance and Accident”, in Chisholm and Haller, eds., 197–210. Chisholm, Roderick M. and Haller, Rudolph eds. 1978. Die Philosophie Franz Brentanos, Amsterdam: Rodopi (also as Grazer Philosophische Studien, 5). Cohen, Robert. S. and Schnelle, Thomas eds. 1986. Cognition and Fact. Materials on Ludwik Fleck, Dordrecht/Boston/Lancaster/Tokyo: D. Reidel. Cornelius, Hans 1894. Versuch einer Theorie der Existentialurteile, Munich: Rieger. Davidson, Donald 1968. “On Saying That”, as repr. in his Inquiries into Truth and Interpretation, Oxford: Clarendon Press, 1984, 93–108. Frege, Gotlob 1893. Grundgesetze der Arithmetik, begriffsschriftlich abgeleitet, Jena (repr. Hildesheim: Olms, 1966). Partial Eng. trans. by M. Furth, The Basic Laws of Arithmetic, Berlekey and Los Angeles: University of California Press, 1964. Galton, Antony. 1984. The Logic of Aspect. An Axiomatic Approach, Oxford: Clarendon Press. Goodman, Nelson 1951. Structure of Appearance, 3rd ed., Dordrecht/Boston: D. Reidel, 1977. Grzegorczyk, Andrzej 1977. “On Certain Formal Consequences of Reism”, Dialectics and Humanism, 1, 75–80. Hoeksema, Jan 1985. Categorial Morphology, New York/London: Garland Publishing. Höfler, Alois 1890. Logik. Philosophische Propädeutik, written in collaboration with A. Meinong, Vienna: Tempsky, 2nd enlarged ed. 1922. Husserl, Edmund 1891. Philosophie der Arithmetik. Psychologische und logische Untersuchungen, vol. I, Halle a. d. S., critical edition as Husserliana XII. Husserl, Edmund 1900/01. Logische Untersuchungen, Halle a.d.S.: Max Niemeyer, critical edition as Husserliana XVII, XIX/1, XIX/2. Eng. trans. of 2nd ed. (1913/21) by J. N. Findlay, Logical Investigations, London: Routledge and Kegan Paul, 1970. Husserl, Edmund 1979. Aufsätze und Rezensionen (1890–1910), ed. by B. Rang, The Hague/Boston/London: Martinus Nijhoff (Husserliana XXII). Ingarden, Roman 1964/65. Der Streit um die Existenz der Welt, 2 vols., the 2nd in 2 parts, Tübingen: Max Niemeyer. (Original Polish edition, 1947/48.) Ingarden, Roman 1969. “Le Concept de Philosophie chez Franz Brentano”, Archives de Philosophie, 32, 458–75 and 609–38. Johnson, William, Ernst 1924. Logic, Part III, The Logical Foundations of Science, London: Constable, repr. New York: Dover, 1964. Koffka, Kurt 1935. Principles of Gestalt Psychology, London: Kegan Paul, Trench, Trubner. Köhler, Wolfgang 1947. Gestalt Psychology, New York: Liveright.

ON THE PHASES OF REISM

179

Kotarbiński, Tadeusz 1920. “Sprawa istnienia przedmiotów idealnych” (The problem of the existence of ideal objects), Przegląd Filozoficzny, 23, 149–70. Kotarbiński, Tadeusz 1929. Elementy teorii poznania, logiki formalnej i metodologii nauk, Lwów. Kotarbiński, Tadeusz 1935. “The Fundamental Ideas of Pansomatism”, as trans. in Mind, 64, 1955, 488–500. Kotarbiński, Tadeusz 1958. “The Development Stages of Concretism”, Studia Filozoficzne, 4(7), cited as trans. in Kotarbiński 1966, 429–37. Kotarbiński, Tadeusz 1966. Gnosiology. The Scientific Approach to the Theory of Knowledge, Oxford: Pergamon Press, Eng. trans. by O. Wojtasiewicz, of 2nd ed. of Kotarbiński 1929, with appendices. Kotarbiński, Tadeusz 1966a. “France Brentano comme réiste”, Revue internationale de Philosophie, 20, 459–76. Eng. trans. in McAlister, ed., 194–203. Kraus, Oscar. 1937. Die Werttheorien. Geschichte und Kritik, Brünn/Vienna/Leipzig: Rohrer. Küng, Guido 1967. Ontology and the Logistic Analysis of Language, Dordrecht: Reidel. Küng, Guido 1974. “Prologue-Functors”, Journal of Philosophical Logic, 3, 241–54. Küng, Guido and Canty, John Thomas 1970. “Substitutional Quantification and Leśniewskian Quantifiers”, Theoria, 36, 165–82. Lejewski, Czesław. 1958. “On Leśniewski’s Ontology”, Ratio, 1, 150–76, as repr. in J. J. T. Srzednicki, et al., eds. 1984, 123–48. Lejewski, Czesław 1976. “Ontology and Logic”, in S. Körner, ed., Philosophy of Logic, Oxford: Blackwell, 1–27. Lejewski, Czesław 1979. “On The Dramatic Stage in the Development of Kotarbiński’s Pansomatism”, in P. Weingartner and E. Morscher, eds., Ontology and Logic, Berlin: Duncker & Humblot, 197–218. Lejewski, Czesław 1982. “Ontology: What Next”, in W. Leinfellner, et al., eds., Language and Ontology, Vienna: Hölder-Pichler-Tempsky, 173–86. Leśniewski, Stanisław 1911. “Przyczynek do analizy zdań egzystencjalnych” (Contributions to the analysis of existential propositions), Przegląd Filozoficzny, 14, 329–45. Leśniewski, Stanisław 1913. “Krytyka logicznej zasady wyłączonego środka” (Critique of the principle of excluded middle), Przegląd Filozoficzny, 16, 315–52. Leśniewski, Stanisław 1914. “Czy klasa klas, nie podporządkowanych sobie, jest podporządkowana sobie?” (Is the class of classes not subordinate to themselves subordinate to itself?), Przegląd Filozoficzny, 17, 63–75. Leśniewski, Stanisław 1927/31. “O podstawach matematyki” (On the foundations of mathematics), Przegląd Filozoficzny, vols. 30–34. Cited according to the abridged Eng. trans. by V. F. Sinisi, “On the Foundations of Mathematics”, Topoi, 2, 1983, 7–52. See also Sinisi 1966 and 1983.

180

BARRY SMITH

Leśniewski, Stanisław 1929. “Grundzüge eines neuen Systems der Grundlagen der Mathematik”, Fundamenta Mathematicae, 14, 1–81. Lipps, Theodor 1905. “Das Wissen von fremden Ichen”, in Lipps, ed., Psychologische Untersuchungen, Leipzig: Engelmann, I/4. McAlister, Linda L. ed. 1976. The Philosophy of Brentano, London: Duckworth. McCall, Storrs ed. 1967. Polish Logic 1920–1939, Oxford: Clarendon Press. Marty, Anton 1895. “Über subjektlose Sätze und das Verhältnis der Grammatik zu Logik und Psychologie”, 7th article, Vierteljahrsschrift für wissenschaftliche Philosophie, 19, 19–87. Marty, Anton 1908. Untersuchungen zur Grundlegung der allgemeinen Grammatik und Sprachphilosophie, vol. I (only volume published), Halle a. d. S.: Niemeyer. Mellor, David Hugh 1981. Real Time, Cambridge: Cambridge University Press. Mourelatos, Alexander P. D. 1981. “Events, Processes and States”, in P. J. Tedeschi and A. Zaenen, eds., Tense and Aspect (Syntax and Semantics, vol. 14), New York: Academic Press, 191–211. Mulligan, Kevin, Simons, Peter M. and Smith, Barry. 1984 “Truth-Makers”, Philosophy and Phenomenological Research, 44, 287–321. Münch, Dieter 1986. “Brentanos Lehre von der intentionalen Inexistenz”, in J. C. Nyíri, ed., From Bolzano to Wittgenstein. The Tradition of Austrian Philosophy, Vienna: Hölder-Pichler-Tempsky, 119–27. Münch, Dieter 1989. “Brentano and Comte”, Grazer Philosophische Studien 35, 33– 54. Novak, Michael 1963/64. “A Key to Aristotle’s ‘Substance”. Philosophy and Phenomenological Research, 24, 1–19. Pelc, Jerzy ed. 1979. Semiotics in Poland 1894–1969, Dordrecht/Boston: Reidel. Quine, Willard V. O. 1960. Word and Object, Cambridge, Mass.: M. I. T. Press. Rasiowa, Helena and Sikorski, Roman 1963. The Mathematics of Metamathematics, Warsaw: P. W. N. Schnelle, Thomas 1982. Ludwik Fleck – Leben und Denken. Zur Entstehung und Entwicklung des soziologischen Denkstiks in der Wissenschaftsphilosophie, Freiburg i. Br.: Hochschulverlag. Sellars, Wilfrid 1962. “Time and the World Order”, in H. Feigl and G. Maxwell, eds., Scientific Explanation, Space, and Time, (Minnesota Studies in the Philosophy of Science, vol. III), Minneapolis: University of Minnesota Press, 527–616. Simons, Peter M. 1983. “A Leśniewskian Language for the Nominalistic Theory of Substance and Accident”, in Topoi, 2, 99–109. Simons, Peter M. 1987. Parts. An Essay in Ontology, Oxford: Oxford University Press. Sinisi, Vito 1966. “Leśniewski’s Analysis of Whitehead’s Theory of Events”, Notre Dame Journal of Formal Logic, 7, 323–27. Sinisi, Vito 1983. “The Development of Ontology”, Topoi, 2, 53–61.

ON THE PHASES OF REISM

181

Smith, Barry 1981. “Logic, Form and Matter”, Proceedings of the Aristotelian Society, Supplementary Volume 55, 47–63. Smith, Barry 1987. “The Substance of Brentano’s Ontology”, Topoi, 6/1, 37–47. Smith, Barry 1988. “Sachverhalt”, in J. Ritter and K. Gründer, eds., Historisches Wörterbuch der Philosophie, Basel/Stuttgart: Schwabe, vol. VIII. Smith, Barry 1989. “Kasimir Twardowski: An Essay on the Borderlines of Psychology, Ontology and Logic”, in K. Szaniawski, ed. Smith, Barry 1990. “Brentano and Marty: An Inquiry into Being and Truth”, in K. Mulligan, ed., Mind, Meaning and Metaphysics. The Philosophy and Theory of Language of Anton Marty, Bordrecht/Boston/Lancaster: Martinus Nijhoff. Sobociński, Bolesław 1971. “Atomistic Mereology”, Notre Dame Journal of Formal Logic, 12, 89–103, 203–13. Srzednicki, Jan J. T. 1965. Franz Brentano’s Analysis of Truth, The Hague: Martinus Nijhoff. Srzednicki, Jan J. T. et al., eds. 1984. Leśniewski’s Systems: Ontology and Mereology, The Hague/Boston/Lancaster: Martinus Nijhoff. Szaniawski, Klemens 1977. “Philosophy of the Concrete”, Dialectics and Humanism, 1, 67–72. Szaniawski, Klemens 1989. The Vienna Circle and the Lvov-Warsaw School, Dordrecht: Nijhoff. Tarski, Alfred 1944. “The Semantic Conception of Truth and the Foundations of Semantics”, Philosophy and Phenomenological Research, 4, 341–76. Tarski, Alfred 1956. Logic, Semantics, Metamathematics, Oxford: Clarendon Press. Tegtmeier, Erwin 1981. Komparative Begriffe. Eine Kritik der Lehre von Carnap und Hempel, Berlin: Dunker & Humblot. Twardowski, Kasimir 1894. Zur Lehre vom Inhalt und Gegenstand der Vorstellungen. Eine psychologische Untersuchung, Vienna: Hölder, repr. Munich/Vienna: Philosophia, 1982. Eng. trans. by R. Grossmann, On the Content and Object of Presentations, The Hague: Martinus Nijhoff, 1977. Twardowski, Kasimir 1902. “Über relative Wahrheit”, Archiv für systematische Philosophie, 8, 415–447. Twardowski, Kasimir 1912. “Actions and Products. Comments on the Border Area of Psychology, Grammar and Logic”, as extracted in Pelc, ed., 13–27. Twardowski, Kasimir 1923. “Issues in the Logic of Adjectives”, as extracted in Pelc ed., 28–30. Whitehead, Alfred N. and Russell, Bertrand. A. W. 1910/13 Principia Mathematica, Cambridge: Cambridge University Press. Woleński, Jan 1987. “Reism and Leśniewski’s Ontology”, History and Philosophy of Logic, 7, 167–76.

182

BARRY SMITH

Woleński, Jan 1988. “Marty and the Lvov-Warsaw School” in K. Mulligan, ed., Mind, Meaning and Metaphysics: Philosophy and Theory of Language of Anton Marty, Dordrecht: Nijhoff. Woleński, Jan 1989. Logic and Philosophy in the Lvov-Warsaw School, Dordrecht/ Boston/Lancaster: Reidel. Woleński, Jan ed. 1990. Kotarbiński: Logic, Semantics and Ontology, Dordrecht/ Boston/London: Kluwer. Woodger, Joseph H. 1937. The Axiomatic Method in Biology, Cambridge: Cambridge University Press. Woodger, Joseph H. 1939. The Technique of Theory Construction (International Encyclopedia of Unified Science, vol. II, no. 5), Chicago: University of Chicago Press. Zemach, Eddy 1970. “Four Ontologies”, Journal of Philosophy, 67, 213–47, repr. in F. J. Pelletier, ed., Mass Terms. Some Philosophical Problems, Dordrecht/Boston/ London: Reidel, 1979, 63–80.

Brentanian Philosophy and Czeżowski’s Conception of Existence∗ DARIUSZ ŁUKASIEWICZ

Introduction Tadeusz Czeżowski was one of the closest disciples of Kazimierz Twardowski, and because of this relation Czeżowski’s philosophy is deeply rooted in the Brentanian tradition. In what follows I would like to show the Brentanian character of Czeżowski’s metaphysics by considering his conception of existence. To understand better and evaluate Czeżowski’s views it is advisable to consider first Brentano’s and Twardowski’s relevant views. The analysis of the Brentanian approach to the problem of existence is, in turn, an occasion to rethink the classical philosophical question of the nature of existence. Since Anselm’s ontological argument, the concept of existence has very often reappeared, particularly, in the context of the question whether existence is a predicate, or, to put the question in a more precise way, is existence any property? It was in Aristotle’s metaphysics that a distinction was made between the first and the second substances. The first substance (ens per se) is regarded as an autonomous being, and the second substance is not an autonomous but a dependent being (ens ab alio). A property is, in Aristotle’s terms, a second substance (ens ab alio). It is also said that properties are ascribed to substances, and that they characterize substances, or, that they determine substances, or, that they are embedded in substances as their constituents or parts. There is a controversy between the proponents of properties (not all philosophers would like to accept properties as a kind of being) concerning whether a given property is always a property of one substance, or, many substances? Is there only one redness common to all red things, or is it so that each thing called a ‘red thing’ has its own individual redness? Aristotle maintained that one property is common to more than one substance (in modern philosophical language we say that different individuals exemplify one identical property). The view opposite to the Aristotelian position says that no property is common to many substances; Actions, Products, and Things. Brentano and Polish Philosophy. Arkadiusz Chrudzimski and Dariusz Łukasiewicz (eds.), Frankfurt: ontos, 2006, 183–215.

184

DARIUSZ ŁUKASIEWICZ

each substance has its own individual and unique properties. Among the proponents of properties Plato and his numerous followers defended quite special views. They claimed that it is possible that properties exist independently of the existence of any substance. Therefore, it is possible that there exist properties which are not constituents of any substances. Plato called such peculiar properties ‘ideas’. Plato’s concept of property is essentially different from Aristotle’s view because properties understood as ‘ideas’ exist autonomously; they are ens per se and not ens ab alio. Aristotle taught that we come to know first substances through the knowledge of second substances (properties) which are intelligible, or, as it was later expressed, ‘presentable’, i.e., properties can be given in an act of presentation. Intelligibility of properties was represented in the traditional logic by the proposition ‘S is P’ where ‘S’ is a symbol for the subject (the name of the object possessing a given property), and ‘P’ is a symbol of the predicate (the name of what is predicated of an object designated by ‘S’). So the first substance is the referent of the subject and the second substance (property) is the referent of the predicate. The subject-predicate judgment is the most primary and essential form of judgment, i.e., all judgments may be reduced to this form, and judging consists in combination or separation of subject and predicate1. This kind of judgment is an answer to the question: what is the object designated by ‘S’ like? Now let us turn to ‘existence’. Aristotle did not regard existence as a property of any object. Existence, as the followers of Thomas Aquinas say, was not perceived by ancient Greeks. This may have been so because Greeks believed in the eternity of the world; the world exists without a beginning, and hence its existence was never a problem for them. They did not pose the question of why there is something and not nothing. It was Thomas Aquinas who made existence the main issue of philosophical considerations, and he regarded existence as an act and not as a property. The act of existence was the necessary and sufficient condition for an object to possess any properties. These historical issues are not of central importance here; however, we may note that Anselm, in turn, regarded existence as a unique property which gives an object perfection: the existing God is more perfect than a non-existing God. It is not our aim to provide any sound definition of ‘property’ grounded in classical metaphysics, let it suffice to say that a property is an entity

CZEŻOWSKI’S CONCEPTION OF EXISTENCE

185

which is embedded in, made intelligible by and predicated of an object in subject-predicate judgments. Neither ancient Greeks (Aristotle and his followers included) nor Thomas Aquinas’s disciples regarded the existence of an object as a kind of property.

Brentano Brentano’s philosophical activity is customarily divided into three distinct phases: the early period (the time spent in Würzburg ), the middle period (1874-1904) and the later period (after 1904). One of the distinctive features of his metaphysics is that he never treated existence as a property. So the existence of an object is not an entity embedded in a given object, and it is not predicated of a given object in a subject-predicate judgment. What kind of justification does Brentano provide for, let us call it, the ‘nonpredicative’ concept of existence? The justification was different in each phase of his philosophical activity. In the first phase, existence is not a property because there are no properties at all. We build concepts, by means of which we think of things, but our concepts do not correspond with reality. We also use the concept of property, and by a property we mean a part of a thing. However, a thing is a unity without any parts (Chrudzimski 2004, 64). Therefore, the existence of a thing cannot be any property of it. Brentano accepted properties in the second phase of his philosophical life, but he rejected the view that existence is a property. He did so because he defended the reliability of inner perception. Brentano in this context criticized the subject-predicate conception of judgment, and one of the results following from his criticism was a new theory of judgment, according to which, judging is an act sui generis: judgment is not just a composition of concepts but a unique mental act (idiogenetic theory of judgment). In the later phase Brentano adopted more radical views and claimed that only concretes (souls and bodies) exist. There are no concepts and no properties. There is no sense and possibility in nominalistic language to ask whether existence is a property. Now I would like to focus on the middle phase of Brentano’s philosophy because this philosophy was known and analyzed in Poland by K. Twardowski and T. Czeżowski.

186

DARIUSZ ŁUKASIEWICZ

In Psychology from an Empirical Standpoint Brentano provided reasons why judgment does not consist in combination or separation of subject and predicate, and, in consequence, the basic form of judgment is not a subject-predicate form. A by-product of his argumentation is, according to Brentano himself, the thesis that existence is not a property. Let us briefly examine Brentano’s reasons for refusing the subject-predicate conception of judgment, which, in turn, brought about his refusing of the non-predicative conception of existence. Firstly, Brentano claims that judging consists in affirmation or rejection of something (the same view was defended by J.S. Mill), and not in combination of concepts. Thus, when someone judges that a exists, and assumes that existence is a property, he affirms a whole composed of the existence and the object designated by ‘a’.2 When someone affirms a whole, he affirms implicitly each part of a given whole. In the case of judgment ‘a exists’ one affirms a itself (Brentano 1973, 208). However, there is no difference between the affirmation of a and the affirmation of the whole: the existence and a. Thus, someone who judges that a exists, affirms a itself and nothing more. From this would follow, contrary to the assumption made above, and provided that Brentano’s reasoning is correct, that existence is no property because, if existence were a property, then there would be a difference between a (part) and existing a (whole). Brentano’s reasoning is not clear; it contains gaps and it needs completion. Therefore, it is possible to evaluate its correctness only after some further clarification. First of all, it is not clear how one should understand the principle saying that affirmation of a whole entails affirmation of its parts. Such a principle is false in logic because ‘affirmation’ means only ‘regarding a proposition as true’. Someone who judges (takes as true) that ‘If p, then q’ does not have to regard as true any part of a given proposition. The word ‘affirmation’ (‘to affirm’) may also mean ‘to believe in something’ or ‘to believe in the existence of something’. If Brentano understood ‘affirmation’ as ‘belief in something’, then his reasoning might be as follows. Someone who judges that a exists (let us suppose that ‘a’ stands for a table), believes in the table, and believes in the existence of the table (in the existing table). Since the belief in the table does not differ from the belief in the existence of the table, then the affirmation of a whole is the affirmation of its parts as well. This reasoning is not possible, if we reject the understanding of af-

CZEŻOWSKI’S CONCEPTION OF EXISTENCE

187

firmation according to which ‘to affirm’ means ‘to believe in something’. It seems, however, that one should reject the understanding in question because such an expression, for example, as ‘John believes in the table’ is semantically (and grammatically) deviant, and hence cannot be true. Let us take into consideration the second sense of ‘affirmation’, i.e. ‘belief in the existence of something’. Someone who judges that a exists, affirms part a; he believes in the a’s existence, and he believes in the existence of the a’s existence. Thus, judging does not consist in a combination of concepts, but it consists in the affirmation of an object, i.e. in the belief in its existence3. It would be correct reasoning if it were demonstrated that judging consists in the believing in the existence of something. This, however, was initially assumed by saying that ‘affirmation’ means ‘believing in the existence of something’. If Brentano had understood the word ‘affirmation’ in this manner, then he would have understood it in this way not only in the principle used to prove the thesis that judging consists in the affirmation of something, but also in the proved thesis itself. Thus, his reasoning would be logically invalid because it leads to circulus vitiosus in demonstrando. In other words, the proved proposition says that judging consists in affirmation of object a. In order to demonstrate that it is really the case, Brentano uses the principle saying that someone who affirms a whole also affirms its parts, and Brentano understands here ‘affirmation’ as ‘believing in the existence of something’. However, he has to understand ‘affirmation’ in the same way in the proposition which he tries to prove. If it were not the case, then the principle which he employs would be useless in his reasoning (See also Gumański 1961, 29) The second reason provided by Brentano why judging is not a combination or separation of concepts is that each judgment can be reduced to the existential form ‘a exists’. From what has been said above it follows that an existential judgment, according to Brentano, is neither a judgment which asserts the existence of an object nor a judgment about existing objects (as B. Russell claimed for a time)4. An existential proposition is a proposition expressed in language by a sentence containing the verb ‘exists’. This word, however, is meaningless, and, therefore, does not denote anything. The verb ‘exists’ is only a linguistic sign of an affirmation act, just as ‘oh’ is a sign of pain or joy, etc. As far as categorical propositions symbolized in logic as A, E, I, O are concerned, Brentano’s thesis can be

188

DARIUSZ ŁUKASIEWICZ

accepted (each of them can be formulated in standard predicate logic). Complications arise in cases of such predicative propositions like: (1) Roman gods are guardians of merchants and thieves or: (2) A centaur is a poetic fiction. Such propositions compelled Brentano to revise the rule of reducibility with the help of which he tried to replace each non-existential proposition by an existential one. For any categorical proposition the rule of reducibility said that a categorical proposition may be replaced by an existential proposition which is synonymous, and, in consequence, logically equivalent to a previous one. It is easy to demonstrate that this rule is not satisfied by propositions (1) and (2). It is so because according to the principle of generalization, which is valid in standard predicate logic, (3) Fa → ∑x Fx if an object a has a property F, then there exists such an x which has a property F. In particular, if (1) is true, then (4) There exists a Roman god which is a guardian of merchants and thieves has to be true as well. However, (4) is of course false. Thus, propositions (1) and (4) do not obey the principle of generalization, and, in consequence, they do not satisfy Brentano’s rule of reducibility because they are not logically equivalent, and, hence synonymous. The same problem arises in case of (2). Proposition (2): ‘A centaur is a poetic fiction’ is true. Thus, according to the principle of generalization, (5) There exists a centaur which is a poetic fiction

CZEŻOWSKI’S CONCEPTION OF EXISTENCE

189

should be true too, but (5) is of course false. Brentano’s principle of reducibility fails in this case as well; propositions (2) and (5) are not logically equivalent, and, therefore, synonymous. Propositions (2) and (5) would be logically equivalent provided that there exists such an object as a centaur imaginatively created by the poets. After 1874 Brentano clearly accepted a rich ontology embracing immanent objects; and a centaur imaginatively created by the poets would be one of the members of this class. However, propositions (2) and (5) are not synonymous, and Brentano was aware of that. The expression ‘being a poetic fiction’ is not an attribute of a centaur (in Twardowski’s language: a ‘determiner’) but it modifies the meaning of the concept ‘centaur’ (Brentano 1973, 219). ‘A centaur which is a poetic fiction’ does not denote a living creature which is a man and a horse, but an immanent object existing in a poet’s thoughts. Therefore, the truth condition of (5) is not the existence of an actual centaur but the existence of an immanent centaur. This is why propositions (2) and (5) are not synonymous. From this it follows that the rule of reducibility must be changed, in particular one should give up the synonymy claim, and preserve only the claim of equivalence. The price of such a change is, among others, the rich pluralistic ontology. It was J.S. Mill who pointed out the problem of reducibility of all nonexistential propositions to existential form. Brentano responded to Mill’s observation in one of his letters quoted by him in the Psychology that the application of the rule of reducibility to proposition (2) may in fact result in: (6) There exists a presentation of a centaur. (Brentano 1973, 219) But (2) and (6) are not synonymous either because (2) is about a centaur and (6) is about the presentation of a centaur. There is, however, a difference between the presentation (idea) of an object and the object itself. Thus, this Brentano’s argument would be logically valid provided that one accepts his rich ontology embracing immanent objects, and the principle of reducibility in the weak version: with the claim of equivalence but without the claim of synonymy. The principle of reducibility in the weak form, however, may lead to arbitrary and paradoxical consequences. It would be possible, for example, to reduce (5) to:

190

DARIUSZ ŁUKASIEWICZ

(7) There exist the poets who created a centaur or even to: (8) There exists an even number greater than 2. From all these it follows, contrary to Brentano’s thesis, that it is not possible to reduce without difficulty each non-existential judgment to an existential judgment. However, let us tentatively assume that Brentano is right and his position is fully tenable: i.e. it is possible to reduce all non-existential judgments to the existential form. The thesis of reducibility itself does not suffice to refute the thesis that existence is not a property. One could hold the view that an existential judgment is a kind of subject-predicate judgment (especially after having evaluated negatively Brentano’s first reason for the non-predicative nature of existence); a judgment which asserts that a given object has the property of existence (lacks it), or, that it is an existing (a non-existing) object. The thesis of reducibility would contribute to the view that existence is not a property if one could satisfactorily demonstrate that judging consists in affirmation (rejection) of something, and the word ‘exists’ had no existential meaning, or, if one provided another, more decisive argument against the predicative nature of existence. Brentano, in fact, tried to find such an argument. It was the argument from the inner perception. The inner perception is a perception of mental acts and not of external objects of these acts. Brentano held the view that each mental experience (seeing, hearing of something) is accompanied by the perception of this experience, and, by means of such a perception, we are conscious of our mental life. Each perception, in turn, is a cognitive act, and cognition comes down to judgment. The concept of inner perception plays an important role in Brentano’s epistemology because, according to Brentano, the existence of inner perception enables us to acquire and possess certain and reliable knowledge (Brentano 1973, 140). In inner perception the act of perception and its object are unified, and, therefore, it is not possible to make any mistake as far as our experience is concerned; if I see something, I perceive my act of seeing. I can be wrong about the object I see but I

CZEŻOWSKI’S CONCEPTION OF EXISTENCE

191

cannot be wrong about my inner perception of seeing. In this epistemological context one can provide the following reconstruction of Brentano’s argument that existence is not a property (See Brentano 1973, 210). (1’) If existence is a property, then our first inner perception (the existential judgment saying that there exists a mental act in us) is not possible. (2’) If the first inner perception is not possible, then no inner perception will be possible. (3’) Inner perception exists, and, therefore, it is possible as well. Hence (4’) Existence is not a property. Brentano’s intuitions, as I think, may also be reconstructed in a little modified and more precise way. If existence were a property, then each existential judgment, including the first inner perception, would be a subject-predicate judgment. In order to make any subject-predicate judgment, one must have a concept of a subject and a concept of a predicate. The subject of inner perception would be a mental phenomenon (the concept of such a phenomenon), and the predicate would be the concept of existence. The concept of existence, however, is a very general and abstract concept. Such a concept can be either an innate concept, or, it can be derived from experience. In Brentano’s view the first possibility is not probable (Brentano 1973, 141). However, if the concept of existence were based upon experience, then the first existential judgment would not be the first. Thus, we would obtain a contradiction. But, the contradiction will not arise if the thesis that existence is a property is rejected and regarded as false (See also Gumański 1961, 31– 32). In fact, having rejected the thesis that existence is a property one can refute the view that existential propositions are composed of subject and predicate. Brentano himself summarized his considerations about inner perceptions and judgments as follows:

192

DARIUSZ ŁUKASIEWICZ But it is hard to think of anything more obvious and unmistakable than the fact that a perception is not a conjunction of a concept of the subject and a concept of a predicate, nor does it refer to such a conjunction. Rather, the object of an inner perception is simply a mental phenomenon, and the object of external perception is simply a physical phenomenon, a sound, odor, or the like. (Brentano 1973, 209f.)

One can object to at least two points in Brentano’s argumentation: firstly, to the claim that each experience is accompanied by unshakable knowledge of the existence of such an experience; secondly, to Brentano’s belief that in order to construct a subject-predicate judgment ‘S is P’ it is necessary to have concepts of a subject and of a predicate. If it were really the case, then no perception would be possible at all, except for the possibility that we had had these concepts before we started to perceive. Irrespective of whether Brentano’s arguments are correct, one can conclude that if his claim that existence is not a property were true, then it would be true that existence cannot be presented, i.e., be given in any presentation. This is so because if existence is not a property, then it can neither be given in any partial presentation which presents a particular constituent of an object, nor can it be given in any comprehensive presentation which presents an object as a whole. Around 1874 Brentano, instead of conceiving existence as a property, was inclined to speak about ‘the concept of existence’ (Brentano 1973, 210). The concept of existence is derived from the inner experience with the aid of reflection on true affirmative judgment. However, around 1874 Brentano did not have a clearly elaborated theory of concepts (Chrudzimski 2004, 175). In particular, it was not quite clear what concepts are and how they exist. Brentano said only that concepts are products of certain mental acts, and the basis of mental acts is provided by experimental data. Concepts are neither discovered nor innate, but they are constructed. In traditional logic, concepts have an intension and an extension. Brentano did not say clearly what the content and denotation of the concept of existence are. Speaking about ‘the concept of existence’ also leads to some other problems. Firstly, the concept of existence like any other concept is, as said above, a product of mental acts, but existence of an object is not, or need not, be a product of mental activity. Therefore, there is a difference between the concept of existence and existence itself. Secondly, concepts

CZEŻOWSKI’S CONCEPTION OF EXISTENCE

193

may be treated as a kind of presentations. In such a case the concept of existence would be a presentation of existence. The last conclusion would be incoherent with Brentano’s crucial thesis that existence is not a property, and is not given in any presentation. In Brentano’s later works we find a different conception of existence (Brentano 1930, 27). This conception arose as a by-product of Brentano’s considerations about truth. Existence and non-existence are regarded as judgment-contents. This view was achieved when Brentano tried to join the classic conception of truth and his own (idiogenetic) theory of judgment. According to the classic conception of truth (its strong correspondence version): (9) Judgment is true if and only if it corresponds to the reality. For any true judgment, such as: (10) a does not exist definition (9) leads to: (11) ‘a does not exist’ is true if and only if it corresponds to the reality. However, according to Brentano’s theory of judgment, the negative existential judgment ‘a does not exist’ consists in the rejection of the object a, i.e., in the rejection of reality which the judgment is about, and, in consequence, there is no object a which might correspond to the judgment ‘a does not exist’. Therefore, according to (11), sentence (10) cannot be true. In order to solve this problem and preserve the classical definition of truth, Brentano invented judgment-contents: the existence of something and the non-existence of something. Judgment-contents were to function as truthmakers for existential judgments.

194

DARIUSZ ŁUKASIEWICZ

Twardowski Twardowski in On the Content and Object of Presentations refers directly and indirectly to three of Brentano’s reasons in favor of the idiogenetic theory of judgment and the non-predicative conception of existence. Let us examine how Twardowski modifies and develops Brentano’s views. Firstly, Twardowski makes reference to Brentano’s considerations from 1874, and states that the essence of judgment consists in affirmation or rejection. An affirmed or rejected object is the object of a judgment. Twardowski, however, clearly and firmly says that someone who affirms, or, rejects an object, affirms, or, rejects its existence (Twardowski 1965, 7). The existence of an object is not regarded as an object of a judgment but as judgment-content. In this way Twardowski adopts both one of Brentano’s early views (judgment is a sui generis mental act) and his later conception of existence conceived as judgment-content. The conception of existence as judgment-content emerges in Twardowski’s considerations in the context of the distinction between the content and object of an act, and not, as it was the case in Brentano’s philosophy, in the context of the theory of truth. According to Twardowski, there is a perfect analogy between presentation and judgment (Twardowski 1965, 8)5. Both acts refer to a certain object, and both have content. Because – one can complete Twardowski’s reasoning – presentation and judgment are acts of different kind they have different contents. In fact, the content of a presentation is neither its object nor any of its properties but ‘a picture of an object’, or, ‘presented object’ (Twardowski 1965, 8). Analogically, the content of a judgment is neither an object nor any of its properties but the existence of an object. Twardowski’s view raises some problems. He assumed that ‘to judge’ means the same as ‘to affirm existence’, or, ‘to believe in the existence of an object’ but he only assumed this, and did not show that it is really so. He referred to Brentano’s consideration and clearly stated what the sense of ‘affirmation’ is in the context of the judgment theory. However, as we tried to demonstrate, Brentano’s first reason for holding that existence is not a property and that judging consists in affirmation or rejection of an object is insufficient because the relevant argumentation assumes what is to be demonstrated.

CZEŻOWSKI’S CONCEPTION OF EXISTENCE

195

Apart from this, Twardowski’s claim that existence is a judgmentcontent is misleading. If, as Twardowski holds, there is a full analogy between presentation and judgment, then presentation as well as judgment should have their own separate objects. Presentation has its object but judgment lacks its own object because the object of a given judgment is identical with the object of a given presentation. Next, if the existence of an object is a judgment-content, then judgment-content exists when an act of judging takes place. The same may be said about the content of a presentation; it exists when an act of presentation takes place. Therefore, the existence of a judged object has a relational character. However, there are objects which exist regardless of whether they are judged or not. Further, if we make a judgment that the round square does not exist, then the content of this judgment is the non-existence of the round square. However, the non-existence of the round square as the content of the judgment in question exists. The objection which can be raised here is that the rejection of the existence of the round square entails the affirmation of the existence of the non-existence of the round square, and since the existence of something regarded as judgment-content exists then we also affirm the existence of the existence of the non-existence of the round square and so on ad infinitum. 6 Yet another objection concerns the inner perception of presentations and judgments. If there is a perfect analogy between presentation and judgment, then this analogy has to be taken into account in relation to inner perception of these acts as well. Twardowski was convinced that a presentation can be an object of another presentation although it does not have to be such an object (Twardowski 1965, 52). Let us assume that in presentation P1 presentation P is given as an object. In this case in P1 is also given the content of P. Twardowski himself states that the content of a presentation can be presented as well (Twardowski 1965, 52). One may conclude, because of the analogy between presentations and judgments, that it is possible to present a judgment. One can suppose that in the presentation of a judgment also the content of the judgment will be given, i.e., the existence or the non-existence of the judged object. This, in turn, would mean, that the existence of an object can be presented, and, according to the Brentanian claim, it is the object itself (to be an object is the same as to be given

196

DARIUSZ ŁUKASIEWICZ

in a presentation). Such a conclusion is incoherent with the main point defended by Brentano and Twardowski that existence is not a property. Secondly, Twardowski could not accept all Brentano’s reasons for the thesis that each subject-predicate judgment may be reduced to the existential form. Some difficulties, let us note here again, arise if we take into consideration such a proposition as (1). According to the principle of generalization, if (1) is true, then (4) has to be true as well. Brentano maintained that (4) is true because there exists a Roman god which is a guardian of merchants and thieves, and is imaginatively created by the poets. Twardowski could not approve of such a solution. Such an object as the Roman god which is a guardian of merchants and thieves neither exists in the mind as an immanent object, nor outside the mind as a transcendent object. There exist, at most, presentations of such an object and their contents. The Roman god which is a guardian of merchant and thieves is a non-existent object; it does not exist in any way. Therefore, according to Twardowski, it would be not possible to reduce (1) to (4), even if the principle of reducibility is taken in the weak sense, i.e., under the condition of logical equivalence solely. Thirdly, Twardowski neither mentioned nor developed Brentano’s argument from inner perception for the non-predicative nature of existence. The simplest explanation of this fact is, perhaps, the observation that Twardowski did not accept Brentano’s view on inner perception. It is not the case that each presentation is accompanied by a presentation of the presentation. It is possible that a given presentation can become the object of another presentation but it requires a special and additional mental activity which is not necessarily attached to a presentation. Despite these differences between Twardowski and Brentano, Twardowski like Brentano, was convinced that existence is not a property, and, in consequence, cannot be presented. But, in 1894 he did not put forward any separate arguments in defense of this claim. In the article from 1907 Twardowski puts forward a view that the object of a judgment is an object whose actuality is affirmed, or, rejected, and, the content of a judgment is the actuality, i.e., existence of the judged object. For example, if we take into account the judgment ‘God exists’, God is the object of the judgment, and its content is actuality, i.e., the existence of God (Twardowski 1965, 418). Twardowski put more stress here

CZEŻOWSKI’S CONCEPTION OF EXISTENCE

197

on the objectivity of judgment-content by saying that the existence of an object is the same as its actuality. This, however, is too little, if, for example, the relational nature of existence is to be cancelled (actuality identified with existence is still judgment content). In his significant paper On Actions and Products from 1911 Twardowski makes a distinction between judging (action) and judgment as a product of judging. Judgment as a product is to be taken as the meaning of a sentence (the abstract content of a judgment). It is clear in this context that existence cannot be regarded as judgment-content because this role is taken over by the judgment as the product of judging and the meaning of a sentence. Obviously, the action-product distinction played an essential role in Twardowski’s conception of judgment and existence. The influence of the action-product distinction upon the conception of judgment and existence is best detectable in Twardowski’s epistemological lectures from 1925. Here again Twardowski defended Brentano’s theory of judgment: that judging consists in affirmation or rejection. The actuality (existence) of an object given in presentation is affirmed or rejected. The object of a judgment is identical with the object of a presentation, and the content of a judgment is the meaning of a sentence which expresses a given judgment. The existence of an object is now labelled not the ‘content of judgment’ but its ground (‘osnowa’). In order to defend Brentano’s conception of judgment, Twardowski provides arguments which are to support the view that existence is not a property, and, hence, that existential judgment cannot be regarded as a subject-predicate judgment. Let us underline this point: for Twardowski the most important matter was to demonstrate that Brentano’s theory of judgment, and not Aristotle’s theory, was correct (Twardowski 1975, 260). The nature of existence became the critical point in this context. If existence were a property, then existential judgments would be, in fact, subject-predicate judgments. Twardowski emphasized that, according to Brentano’s theory of judgment, judging does not consist in bringing concepts into a whole but it rests on affirmation (rejection) of something. In Twardowski’s view it is not important whether the object of a judgment is simple or complex. It seems, however, that it is not essential for the preservation of Brentano’s conception of judgment whether existence is a property or not. We may present the judgment ‘a exists’ and interpret it as the judgment ‘a is an existing entity’ or ‘a has the property ex-

198

DARIUSZ ŁUKASIEWICZ

istence’. A presented judgment is still not a judgment. A presentation of a judgment will become a judgment, if we affirm, or, reject the presented judgment, for example, ‘a is an existing entity.’ Affirmation would consist here in taking the presented judgment as true (we may call it the ‘propositional content’), and rejection would consist in taking the propositional content as false. In this conception the essence of judging is the same as in Brentano’s and Twardowski’s case; judging consists in affirmation or rejection, and not in bringing concepts into a whole. The most crucial difference is of course that affirmation (rejection) would not concern the existence of an object but the propositional content. Existence is treated here as a property but it would not have to violate the view that existence cannot be presented because it could be regarded as an unobservable property of an object. There is no organ of existential perception but this fact does not make it logically impossible to view existence as a property. If this approach to judgment were correct, then one could state that Brentano’s theory of judgment and the predicative (resp. non-predicative) conception of existence are logically independent. Existence might be treated as a nonproperty, if there were some other reasons justifying the non-predicative nature of existence. In fact, Twardowski tried to provide such reasons. He argued that in the classical Aristotelian theory of judgment we always combine different concepts, for example, Desdemona and Cassio’s love, and in the existential judgments we add always the same concept (existence) to a given one. However, this observation alone does not suffice for viewing existence in the non-predicative manner. Also, Twardowski pointed out that the combination of concepts itself does not constitute a judgment because one can imagine such a combination (Martians, existence) which results in the assumption ‘Martians exist’ and not in judgment (Twardowski 1975, 264). Twardowski is obviously right with regard to this point but one can affirm or reject the assumption in question and make a judgment. Thus, this observation by Twardowski is not satisfactory either. Another reason is the following one. It is a mistake to treat existence as a property because from the premise that existence is a property there follow false consequences. Twardowski means here Anselm’s ontological argument and mentions Kant’s criticism of this argument (Twardowski 1975, 265). However, let us note, that the ontological argument is accompanied

CZEŻOWSKI’S CONCEPTION OF EXISTENCE

199

also by other premises, saying, for example, that existence is a perfection of an object; a non-existing object is less perfect than an existing one, and that it is allowed to infer conclusions concerning the existence of an object from the analysis of its concept. If these assumptions are made, then there arise difficulties pointed out by the opponents of the ontological argument. The thesis that existence is a property is not sufficient in itself to the construction of the reasoning labeled as the ‘ontological argument’. Therefore, this objection by Twardowski’s to the predicative nature of existence is not sound either. Twardowski also provides the argument that existence is not a relational (relative) property of an object which may be attributed to a given object only if it is in relation to the true affirmative judgment (Twardowski 1975, 266–267). He correctly notes that the existence of an object is not a relative property of an object, but being the object of a judgment can be such a property if a given object is judged. The existence of an object itself is independent of any judgments made about a given object. From this, however, does not follow that existence is not, or, cannot be, a non-relational property of an object.

Czeżowski’s pre-Brentanian Philosophy (1918-1930) As we said at the beginning, Czeżowski was one of Twardowski’s closest disciples. There is a view, however, that he took from Twardowski only the idea of doing philosophy by means of a precise analysis of language, and that he accepted some of his pedagogical views. Apart from these points there were crucial differences between them. Czeżowski was convinced that the fundamental philosophical science is logic; and Twardowski saw descriptive psychology as the most important philosophical science (Wiśniewski 1989, 260). But I do not think this view is quite correct, because, firstly, Czeżowski did not reject psychological analysis. He used this method in his investigations, and he achieved significant results, especially in the description of the structure of ethical and aesthetic evaluations (See, for example, Łukasiewicz 2002, 280–283). Secondly, logic was for Czeżowski an instrument in philosophical investigations, and it served as the criterion of scientific validity. So logic was not a substitute

200

DARIUSZ ŁUKASIEWICZ

for philosophy. Philosophy provided data for logical analysis. Such an understanding of the main factors in Czeżowski’s philosophizing is also confirmed by his analysis of existence. One can distinguish two phases in Czeżowski’s philosophy. The first (early) phase (1918-1930) had been under the influence of Frege, Meinong and Russell, and this phase may be called the ‘pre-Brentanian’ phase. The second phase (Brentanian) was influenced mainly by Brentano and Twardowski. In brief, the most characteristic of Czeżowski’s ideas of the early phase are: (a) Existence is the most universal and primitive property. (b) Each judgment (existential judgments included) can be reduced to the relational form ‘aRb’. (c) Judgments assert states of affairs and states of affairs are objects of judgments; a state of affair is composed of objects and a relation obtaining between them. (d) A judgment (proposition) is the content of a sentence expressing it, and the state of affairs asserted in the sentence is the meaning of it.7 (e) Truth is the denotation of a true sentence, and falsity is the denotation of a false sentence. Czeżowski’s view that relational judgment is the most fundamental form of judgment is mainly rooted in B. Russell’s and L. Couturat’s ideas which Czeżowski was dealing with in his dissertation on class paradox (Czeżowski 1918c) and habilitation “Variables and Functions” (Czeżowski 1920). Russell, as is well known, having rejected the neo-Hegelian idealism, refuted the subject-predicate theory of judgment. He was convinced that this theory of judgment is one of the premises of idealism because it reduces relations to monadic properties of objects. According to such a view, relations are solely mind products, and, therefore their existence is dependent on the mind. For Russell the objectivity of relations was a guarantee of realism, the defense of which played an important role in his philosophy. This is why Russell emphasized the importance of relational propositions in logic and metaphysics (Hylton 1990, 155). The axiom put forward by L. Couturat in Les principies des mathematiques might have influenced Czeżowski’s apprehension of judgments as well (Couturat

CZEŻOWSKI’S CONCEPTION OF EXISTENCE

201

1905, 33). Couturat’s axiom states that there is a relation between any two objects a and b determined by these objects. Czeżowski mentioned this axiom in his habilitation “Variables and Functions” (Czeżowski 1920, 33). Russell’s and Couturat’s views also exerted the influence on the way Czeżowski conceived the very nature of relations asserted in relational propositions. For example, Czeżowski says that all propositions about any object can be reduced to the proposition saying that a given object is an element of a certain class (Czeżowski 1918c, 32). This view may serve as evidence of Czeżowski’s acceptance of the Principle of Reducibility, formulated by Russell and taken over by Czeżowski from Couturat. The Principle of Reducibility says that each propositional function φ(x) corresponds to the proposition which has the general form x ε a (a is the class of objects satisfying, i.e., making true the function φ(x), and the symbol ‘ε’ denotes the relation of membership obtaining between an individual and the class a) (Couturat 1912, 160)8. After approving of (b) Czeżowski could easily accept (c). For, if each proposition can be reduced to the relational form, then a complex entity should be regarded as the object of a proposition. Such an entity is called by Russell a ‘fact’ (by Husserl a ‘state of affairs’), and Meinong, whose ideas were known to Czeżowski too, labelled it ‘Objectiv’. Czeżowski, perhaps under J. Łukasiewicz’s influence, took from Meinong the view described in thesis (d). The circumstance that a state of affairs is treated as the meaning of a sentence, and the proposition is the content of a sentence, could be regarded by Czeżowski as support for the objectivity of meaning. Thesis (e) was grounded in Frege’s conception of truth, and it served as a criterion for demarcation between names and sentences; names differ from sentences because they have different denotations. Czeżowski’s definitions of a sentence and of truth are exemplifications of his belief in the essential role of relational propositions. A sentence is an expression such as, for example, ‘Cain killed Abel’, or ‘The whale is a mammal’. Briefly, (12) A sentence is an expression of the form aRb in which a and b are names of objects and R is a name of a relation (Czeżowski 1918b, 112).

202

DARIUSZ ŁUKASIEWICZ

The definition of truth says that (13) The sentence ‘aRb’ is true if a is in relation R to b (Czeżowski 1918b, 112). Thesis (a) can be interpreted as follows. The sentence: (14) a exists is synonymous and logically equivalent to: (15) a is existing. The word ‘is’ in (15) denotes the relation of membership between the referent of the name ‘a’ and the class of existing objects, i.e., objects which have the property of existence, and the word ‘exists’ in (14) is synonymous with ‘is existing’ occurring in (15). It is possible to interpret the word ‘is’ in such a way that it denotes the relation of inherence, or characterizing: the object a is characterized by a property F. In this case (15) could be translated into: (16) There is a relation of characterizing between the object a and the property of existence. Sentences (15) and (16) would be logically equivalent because it is possible to define the relation of a’s membership to the class of existent objects, say the class X, with the aid of the relation of characterizing: (17) a is a member of X if and only if a is characterized by the property existence. Czeżowski himself describes his view by saying that the existential proposition ‘a exists’ is equivalent with the proposition ‘a is such as anything’ (Czeżowski 1918c, 32).9 I think proposition

CZEŻOWSKI’S CONCEPTION OF EXISTENCE

203

(18) a is such as anything is to be understood as saying that a is a being (object). The word ‘is’ in (18) would stand for the membership relation obtaining between the object called ‘a’ and the universe of all objects. Because of (17) and the transitivity of equivalence, sentences (14) and (18) are equivalent, which Czeżowski, in fact, admitted. Czeżowski treated the word ‘exist’ in his early writings univocally; existence is a universal property of each object, and there are no kinds of existence, contrary to, for example, colors – there is more than one shade of redness. He also assumes a clear ontological premise contained in the proposition saying that expressions ‘existent’ and ‘object’ have the same denotation. From this ontological premise it follows that if something does not exist, then it is not an object, and, if one may say so, it is, in fact, just nothing. As our aim here is not to provide an analysis of the coherence and soundness of Czeżowski’s claims but to describe the changes which occurred in his views on existence, we put aside a detailed inspection of his propositions. However, one should point out the difficulties that arise in his early writings as a result of the assimilation of some Brentanian theses. In Czeżowski’s early logical writings there are two propositions belonging to the Brentanian descriptive psychology. The first of them says that each presentation has its object determined by its content (Czeżowski 1918a). Czeżowski proposed to make a sharp demarcation line between the scope of logic and the scope of psychology by distinguishing between denotations of names and objects of presentations. Names would be a subject matter of logic, and presentations would be a subject matter of psychology. It is possible that there are names lacking denotation, for example, ‘centaur’ but it is impossible that there are presentations lacking objects. A centaur is not a referent of the name ‘centaur’ because there are no centaurs in the world but a centaur is a correlate of the presentation of a centaur (Czeżowski 1918a, 108). It follows that the denotation of a name (subject matter of logic) and the class of objects of a presentation (subject matter of psychology) are not identical (a centaur is an object but it does not exist). However, it would be incoherent with the ontological premise underlying the analysis of existential propositions which says that the expressions ‘existent’ and ‘object’ have the same denotation.

204

DARIUSZ ŁUKASIEWICZ

The second Brentanian proposition says that judgment consists in affirmation or rejection (Czeżowski 1918b, 112). Czeżowski, however, contrary to Brentano and Twardowski, maintained that neither the existence of an object nor the object itself is affirmed or rejected in the act of judging. Judgment has his own propositional correlate, i.e. a relational state of affairs.

Czeżowski’s Brentanian Philosophy In the 1930s Czeżowski subscribed to Twardowski’s conception of judgment. Judging consists in affirmation or rejection of the existence of an object given in a presentation (Czeżowski 1933). Czeżowski’s acceptance of the Brentanian theory of judgment went together with his refutation of the thesis that existence is a property (Czeżowski 1938, 4). We tried to point out that the acceptance of the Brentanian conception of judgment does not entail the acceptance of the non-predicative concept of existence. Czeżowski did not mention reasons provided by Brentano or Twardowski in defense of the non-predicative nature of existence, and did not use them, except for the argument concerning the reducibility of categorical judgments to existential form. Czeżowski’s departure from the relational conception of judgment and predicative conception of existence was not sudden and unexplainable. Czeżowski, as has been said above, regarded judgment as affirmation, or rejection, already in his early logical writings. The change which took place was that it is not a state of affairs which is affirmed, or rejected in the judgment but the existence of a state of affairs, or the existence of a simple non-propositional object. There are a few reasons for this change of view. Firstly, one regards sentence in the sentential calculus as a primitive entity and the target of analysis. This fact, in Czeżowski’s view, confirms Brentano’s and Twardowski’s conceptions of judgment because judgment is treated by them as a mental act sui generis (Czeżowski 1925). Secondly, Czeżowski came to the conclusion that each judgment, except for judgments about fictional objects, can be reduced to an existential judgment conceived in the Brentanian way. His own studies on Aristotle’s logic assured him in this respect (Czeżowski 1927). In the late 1940s Cze-

CZEŻOWSKI’S CONCEPTION OF EXISTENCE

205

żowski analyzed sentences about fictitious objects such as (1) and (2). He treated (1) as false according to Russell’s theory of description. Sentence (2) ‘A centaur is a poetic fiction’ is, according to him, true and synonymous with the sentence: ‘No living creature is a centaur’. (Czeżowski 1948, 76; Czeżowski 2004, 85). Thirdly, he argued that by the presentation of an object we are not able to draw any conclusions about its existence, or non-existence, because, although we know the object’s properties by the content of a presentation, there is nothing in the content of a presentation that could give us a hint as to the existence of an object (Czeżowski 1938, 4). According to Czeżowski, the problem of existence is posed and decided by the judgment (in Czeżowski’s terminology: belief)10. Czeżowski pointed out two additional reasons (absent in Brentano’s and Twardowski’s considerations) for the thesis that existence and other transcendental concepts are not properties11. The first one was provided by the analysis of the syntactic structure of expressions composed of such words as ‘true’, ‘good’, ‘necessary’ and ‘beautiful’, and the second reason was found in the logic of quantifiers. Briefly, with regard to the first reason, existence is not a property because it is not symbolized in language by a predicate. Such words as ‘true’, ‘good’, ‘necessary’ and ‘beautiful’ are only morphologically similar to predicates but, in fact, they are not predicates. They are sentential functors because they occur in such sentential constructions as, for example, ‘It is necessary that…’, ‘It is good that…’, ‘It is beautiful that…’, or ‘It is true that…’. It is of course permissible to say: (19) The rain is necessary and it is possible to say: (20) The rain is torrential. (19) is equivalent to: (21) It is necessary that it is raining.

DARIUSZ ŁUKASIEWICZ

206

However, it is not possible to interpret (20) in the same way because the linguistic construction ‘It is torrential that…’ cannot result in any sentence (Czeżowski 1965, 38). According to Czeżowski, the sentence ‘a exists’ should be translated into the following sentence: (22) It is true that some x is a or into: (23) There is such x that x is a. The word ‘is’ occurring in the sentence ‘Some x is a’ denotes the relation of membership if a is a general term, or the relation of identity if a is a singular term. He says that: In all these examples there occurs a sentence composed of modus and dictum (if we use classical terminology); modus is an expression ‘It is necessary that…’, ‘It is true that…’ etc., dictum is the sentence following modus. We call today modus a sentential functor. The circumstance that modal functors (necessary, possible), the functor of assertion (it is true that…) and the functor of evaluation (good, beautiful) do require as their complement a sentence (and not a name, as other adjectives do when they play the role of an attribute) shows that these modi cannot be given in presentations but that they are asserted in propositions. Anyway, it has been well known for a long time – Hume and Kant were conscious of it – that they (modi) cannot be given in any presentation, and even that these expressions are ‘contentless’; they express solely someone’s reaction to a certain state of affairs. (Czeżowski 1965, 69)

Evidence for the non-predicative character of existence is also provided by modern logic, according to which, the concept of existence is connected with particular quantifiers, and it serves to define the non-emptiness of the general term which is a predicate in the sentence represented by the propositional function ‘x is P’. The propositional function ‘x is P’ is synonymous with (24) There exists such x that x is P

CZEŻOWSKI’S CONCEPTION OF EXISTENCE

207

expressed in the language of standard predicate logic as (25) ∑x Px. ‘P’ in the function ‘x is P’ stands for a property P of an object represented by x. Thus the existence of P is the existence of a property, or, to put it in Aristotle’s terms, the existence of P is the existence of the second substance which exits in the first substance. In this context one should distinguish the existence of individuals – introduced as a primitive concept – and the existence of properties defined by the existence of individuals (Czeżowski 1965, 64). However, to be precise, for Czeżowski, it is not the case that existence is not a property, because it is represented by a quantifier, and a quantifier is not a predicate. For Czeżowski it is not the existential quantifier which is a logical symbol of existence but it is the sentential functor of assertion ‘It is true that…’. Czeżowski says that The concept of existence is bound with the concept of truth by the classical definition of truth, according to which a proposition is true if and only if there exists that which is asserted by that proposition. (Czeżowski 1965, 69)

Thus, it would even be possible to reject the existential interpretation of the particular quantifier and to preserve the thesis about the non-predicative nature of existence resting solely on the classical definition of truth and the syntactic analysis of language. Czeżowski, in fact, accepted both the existential interpretation of the particular quantifier and the classical definition of truth. The thesis about the non-predicative nature of existence is supported on the one hand by psychological analyses of presentations whose results are coherent with Hume’s and Kant’s intuitions, and, on the other, by the analysis of natural language and the symbolic language of modern logic. Taking these facts into account one can conclude that by defending the non-predicative nature of existence Czeżowski posed the problem of existence in a wider context than Brentano and Twardowski did. For them the problem of existence appeared directly in the context of judgment-theory, and some solutions concerning the nature of existence played an important role in defense of Brentano’s theory of judgment. There is no doubt, how-

208

DARIUSZ ŁUKASIEWICZ

ever, that the essential stimulus for Czeżowski’s abandoning the predicative nature of existence was the Brentanian theory of judgment and the classical definition of truth defended vigorously by Twardowski and, for a time, by Brentano. An entirely different and open question is whether Czeżowski’s arguments supporting the non-predicative nature of existence are conclusive. Czeżowski’s argumentation for the non-predicative nature of existence can be questioned by criticism of the premises which are involved in it. One can discuss: (i) the classical definition of truth (ii) the relation of correspondence between the concept of truth and the concept of existence (iii) the ontological commitment of the ‘particular quantifier’ (iv) the role of language analysis for the ontological conclusions (v) the relation between the thesis that existence cannot be presented and the thesis that existence is not a property. The definition of truth which Czeżowski proposed in his late works says that (26) A sentence is true if and only if it asserts that what exists or negates that what does not exist. (Czeżowski 1946, 65) (26) was an integral part of Czeżowski’s mature epistemology taken from Twardowski and grounded in the idiogenetic theory of judgment, and a direct consequence of (26) is (ii). In Czeżowski’s case, (ii) was connected with the belief (one may call it the ‘Russellian belief’) in a ‘robust sense of reality’ saying that all objects exist, and that there are no non-existent objects (Butchvarov 1985, 408). If this belief was rejected, then the following (let us say Meinongian propositions) would be true: (27) The golden mountain is golden. Or:

CZEŻOWSKI’S CONCEPTION OF EXISTENCE

209

(28) The round square is round. According to this Russellian belief, the intuitively true proposition: (29) The round square does not exist would say that there is no such an object as the round square or, on the semantical reading, the expression ‘the round square’ is not a name because it does not name anything. It is characteristic of Czeżowski that he always held the thesis about a ‘robust sense of reality’, and it was one of the most important differences between him and Twardowski; we may say that Czeżowski’s Brentanism was obviously anti-Meinongian. Thus, for Czeżowski both (27) and (28) were false.12 The rejection of (ii) would make impossible the claim that the sentential functor of assertion ‘It is true that…’ is the functor of existence. It is worthy of note that by negating the Russellian belief the Meinongian thesis that there are objects which do not exist would have to be true13. But let us repeat: Czeżowski never decided to make such a step. It was Czeżowski’s disciple Leon Gumański and later on Jacek J. Jadacki who did it in Poland. For, as I think, the belief in a ‘robust sense of reality’ as well as lack of this belief is not an essential constituent of Brentanism Czeżowski’s sympathy with it is not evidence of his reservations about Brentanism, in regard to the issue of existence. The acceptance of the Russellian belief entailed the acceptance of (iii) – the acknowledgment of the existential sense of the particular quantifier. Therefore, it is not surprising that someone who rejected the belief in a ‘robust sense of reality’ rejected (iii) as well. L. Gumański who did reject this was convinced that the formula ‘∑x Fx’ means the same as the expression ‘For some x: x is F’. The expressions ‘for some x’ and ‘there exists x such as …’ are not synonyms (Gumański 1961, 92). Jadacki reads ‘∑x Fx’ as saying that “There is a name which put in the expression ‘Fx’ turns it into a true proposition” (the so-called substitutional reading of quantifiers) (Jadacki 1980, 54).14 Anyway, by rejecting (iii) one can argue as follows: if existence is not a quantifier, then it is a first-level predicate, and it is a first-level property.

210

DARIUSZ ŁUKASIEWICZ

Among Polish philosophers this route was chosen in the most decisive way by Gumański. By rejecting (iv) one can give up Czeżowski’s argument expressed in sentences (19)–(21). Brentano himself was suspicious about to the role of linguistic analysis in philosophy and, in particular, about drawing ontological conclusions from the analysis of language15. For Czeżowski, however, the analysis of language was the basic device in philosophical work, and he did not hesitate to draw conclusions from such analyses. The same view concerning the role of language analysis was held by his disciple – Leon Gumański. The latter, however, came to conclusions contradictory to Czeżowski’s; existence for Gumański is a (first-level) property (Gumański 1961, 55). This is so because Gumański rejected the thesis about the correspondence between truth and existence, and the belief in a ‘robust sense of reality’. This is why, according to Gumański, the functor of assertion does not play the role of the functor of existence. It is, of course, possible to challenge thesis (v) saying that if existence is not presentable, then it is not a property. It was Gumański again who did this in Poland. It is true, according to him, that existence cannot be given in any presentation but it is not true that existence is not a property. There are many properties of an object which cannot be presented, for example, the presence or location. Existence should be treated as an object’s unobservable property, i.e., a property which cannot be perceived with the aid of any separate sense organ because there is no sense organ of existential perception. In Gumański’s view, existence cannot be inferred from other properties (propositions about some properties) either; we know that an object exists by means of a special kind of existential intuition which is grounded in sensory data provided by all the sense organs that we have.16 Thus, our short review of possible objections to Czeżowski’s arguments for the non-predicative concept of existence shows that there are, in fact, two crucial premises supporting his view: the classical conception of truth rooted in Aristotle’s philosophy accompanied by the Russellian belief in a ‘robust sense of reality’, and thesis (v) rooted mainly in the tradition of the Anglo-Saxon empiricism (which Kant approved of as well). Both these conceptions played an important role for Czeżowski’s original theory, according to which, existence (like good, beauty, necessity and possibility) is a transcendental concept which designates the object’s mode of being

CZEŻOWSKI’S CONCEPTION OF EXISTENCE

211

(modi essendi) asserted in a proposition17. The proposition that existence is a transcendental concept does not, of course, follow merely from the proposition that it is not a property. Existence has a transcendental nature because it is predicated in a sentence of each object (a ‘robust sense of reality: each object exists; something which does not exist is not an object at all), and existence is not a real predicate (following Kant’s terminology) because, as Hume already taught, it adds nothing to the object’s description18. There is still another interesting question concerning the ontological nature of transcendentalia: what is the difference between modi essendi and Sachverhalte (states of affair)? One possible answer would be that modi essendi cannot be presented, and Sachverhalte can in principle be given in presentation19. Both these main ideas are not in fact typically Brentanian, although the classical definition of truth establishing the relation of truth and existence is bound up with the Brentanian idiogenetic (non-reductionist) conception of judgment. Both appeared in Czeżowski’s philosophy when he abandoned the relational theory of judgment, and approved of the idiogenetic conception of judgment whose correctness was confirmed, at least for Czeżowski, by modern logic. Perhaps, Czeżowski’s belief in the crucial role of logic for philosophical investigations, so typical of the whole Lvov-Warsaw School, was the most important factor deciding that he subscribed to the non-predicative and transcendental concept of existence.

Notes ∗

I would like to express gratitude to my wife Elżbieta both for reading and improving the present paper. I’m also indebted to Arkadiusz Chrudzimski for his helpful remarks. The remaining shortcomings are the author’s own responsibility. 1 In what follows I employ the terms ‘judgment’ and ‘proposition’ interchangeably, except for places where the distinction between purely psychological and logical issues is crucial. In these places ‘judgment’ stands for a mental entity and ‘proposition’ stands for an abstract, non-mental object. 2 ‘a’ is a singular or general term. 3 The affirmation of the whole: the existence of the a’s existence would amount to the affirmation of the part: the existence of a.

212 4

DARIUSZ ŁUKASIEWICZ

The negative existential judgment ‘The round square does not exist’ does not concern an existing object but the non-existing round square. 5 This claim of Twardowski’s was criticized by Jan Woleński in (Woleński 1990). 6 Twardowski in (Twardowski 1965, 418; Twardowski 1975, 261) said that all existential judgments have the same content or ground: the existence of an object. The existence of an object is affirmed in an affirmative existential judgment and rejected in a negative existential judgment. This would mean that the non-existence of an object is not regarded by Twardowski as the ground (content) of a negative existential judgment. The non-existence of an object would be just nothing. 7 Czeżowski understands the term ‘judgment’ as synonymous with the term ‘proposition’. 8 The word ‘is’ occurring in such sentences as ‘a is b’ denotes the relation of inclusion between the class a and the class b, for example, in the sentence ‘A whale is a mammal’ the word ‘is’ denotes the relation of inclusion between the class of whales and the class of mammals. 9 To be more precise, he says: ‘Sąd egzystencjalny “P jest istnieje”, w którym przypisuje się przedmiotowi [P] najogólniejszą, najbardziej pierwotną cechę istnienia, jest równoważny z sądem P jest takie jak cokolwiek’ (the existential proposition ‘P exists’ which attributed existence, i.e., the most universal and primitive property, to object P is equivalent with the proposition ‘P is such as anything’) (Czeżowski 1918c, 32). All quotations from Czeżowski appearing in this paper are my translations from Polish into English. 10 This point may be objected to because if we think of the round square, for example, and detect contradictory elements in the content of its presentation, we are able to say that the round square does not exist. 11 In his later works Czeżowski regarded existence, and also good, beauty, necessity and possibility as transcendental concepts (transcendentalia) (Czeżowski 1965, Czeżowski 1977). Czeżowski’s theory of transcendental concepts was developed by Jan Woleński in (Woleński 2004). 12 He changed his mind in one of his last works (Czeżowski 1969) in such a way that he would be ready to consider (27) but not (28) to be true. It was possible because Czeżowski extended the scope of the term ‘existent’ and counted fictitious objects as existent. The existence of fictitious objects was related to the existence of literary works. I discuss the issue in more detail in (Łukasiewicz 2004). 13 If we reject the Russellian thesis, then it is true that there are objects such as the golden mountain or the round square which do not exist, and then the propositions (27) and (28) could be regarded as true. This would mean, however, that there is no relation between truth and existence because there are true affirmative propositions about nonexistent objects. 14 Jadacki, influenced by Marian Przełęcki’s criticism (Przełęcki 1980), gave up this view and after some years proposed a more neutral interpretation of ‘∑x Fx’ saying that “There obtains such x that Fx’ (Jadacki 1996, 55). The notion of ‘obtaining’ means ‘being an element of a certain set’, and it must be distinguished from the concept of existence as defined by having position in place and time.

CZEŻOWSKI’S CONCEPTION OF EXISTENCE

213

15

He says, for example, “And so I hope that people will finally, once and for all, stop confusing linguistic differences with differences in thought” (Brentano 1973, 220). 16 Gumański described the existence of an object in terms of its location in space and time; objects lacking such a location do not exist and they are labelled ‘abstract’. Numbers or fictitious objects are abstract objects (Gumański 1961, 82). 17 The view that existence is a transcendental concept is proposed by Butchvarov too. However, according to him, a transcendental concept is a concept which is a classificatory concept and which does not designate anything (Butchvarov 1985, 408). 18 There is, of course, the question how to reconcile the Brentanian claim, adopted also by Czeżowski, that being an object is equivalent to being presentable with the Russellian thesis about a ‘robust sense of reality’. This is so because certainly not everything what can be thought of exists, and, in consequence, not everything thinkable is an object in the Russellian sense. This question makes direct reference to the cornerstone of the Brentanian philosophy, i.e., to the theory of intentionality. According to this theory, each mental act is directed toward an object intended in the act. But not every object has to exist. Czeżowski stressed many times that acts such as thinking about the round square or about the centaur have their intentional objects such as the non-existing round square or the non-existing centaur (See, for example, Czeżowski 1959, 4). But I think that in Czeżowski’s case speaking about intentional non-existing objects is, as Husserl used to say, ‘merely figurative’. There exists a mental act (an intention) referring to something, or, to put it differently, a mental act which has the property of being referred to but there is no non-existing object of a given act. The function of the target of the intention is played by its content. I do not mean to say that this is a good solution but it was a solution adopted for a time by Czeżowski. I discuss this problem in a detailed way in (Łukasiewicz 2004). 19 Czeżowski states that the concept of existence, like the concepts of necessity, possibility, good and beauty, is a transcendental concept similar to the concept of being (ens). However, these former concepts differ from the concept of ens in that they can be negated, and their negation is not empty while ens cannot be negated (Czeżowski 2004, 190). The negation of necessity is contingency, the negation of good is evil, and the negation of existence is ‘lack of existence’. It is not, however, clear how to understand Czeżowski’s proposition that the negation of existence is not empty since it is ‘lack of existence’. A possible solution is that the lack of existence is just the non-existence of something.

References Brentano, Franz 1973. Psychology from an Empirical Standpoint, Eng. trans. by Antos C. Rancurello, D. B. Terrell and Linda L. McAlister, New York: Routledge and Kegan Paul, Humanities Press. Brentano, Franz 1930. Wahrheit und Evidenz, ed. by O. Kraus, Leipzig: Felix Meiner. Eng. trans. ed. by R. M. Chisholm, The True and the Evident, London: Routledge and Kegan Paul, 1961.

214

DARIUSZ ŁUKASIEWICZ

Butchvarov, Panayot 1985. “Our Robust Sense of Reality”, in Haller Rudolf (ed.) NonExistence and Predication, Amsterdam: Rodopi, 403–421. Chrudzimski, Arkadiusz 2004. Die Ontologie Franz Brentanos, Dordrecht: Kluwer Academic Publishers. Chrudzimski, Arkadiusz and Huemer, Wolfgang (eds.) 2004. Phenomenology and Analysis, Frankfurt/Lancaster: ontos Verlag. Couturat, Louis 1905. Les principes des mathematiques, Paris. Couturat, Louis 1912. “Die Prinzipien der Logik”, in Arnold Ruge (ed.) Encyclopädie der philosophischen Wissenschaften, Tübingen: Verlag von J.C.B. Mohr (Paul Siebeck). Czeżowski, Tadeusz 1918a. “Imiona i zdania”, Przegląd Filozoficzny, 101–109. Czeżowski, Tadeusz 1918b. “O zdaniach bez treści”, Przegląd Filozoficzny, 110–120. Czeżowski, Tadeusz 1918c. Teoria klas, Lwów: Archiwum Naukowe Wydawnictw Tow. Dla Popierania Nauki Polskiej. Czeżowski, Tadeusz 1920. “Zmienne i funkcje”, Przegląd Filozoficzny, 157–173. Czeżowski, Tadeusz 1925. “Teoria pojęć Kazimierza Twardowskiego”, Przegląd Filozoficzny, 106–110. Czeżowski, Tadeusz 1927. Klasyczna nauka o sądzie i wniosku w świetle logiki współczesnej, Wilno: J. Zawadzki. Czeżowski, Tadeusz 1933. “Spostrzeżenia i przypomnienia”, Kwartalnik Psychologiczny, 237–244. Czeżowski, Tadeusz 1938. Propedeutyka filozofii. Podręcznik dla II klasy wszystkich wydziałów w liceach ogólnokształcących, Lwów: S. Jakubowski. Czeżowski, Tadeusz 1946. Główne zasady nauk filozoficznych, 2nd ed., Toruń: Wydawnictwo Uniwersytetu Mikołaja Kopernika, T. Szczęsny. Czeżowski, Tadeusz 1948. O metafizyce, jej kierunkach i zagadnieniach, 1st ed., Toruń: Wydawnictwo Uniwersytetu Mikołaja Kopernika, T. Szczęsny. Czeżowski, Tadeusz 1959. Główne zasady nauk filozoficznych, 3rd ed., Wrocław: Zakład Narodowy im. Ossolińskich. Czeżowski, Tadeusz 1965. Filozofia na rozdrożu, Warszawa: PWN. Czeżowski, Tadeusz 1969. “Pojęcie prawdziwości w odniesieniu do utworów literackich”, in Odczyty Filozoficzne, 2nd ed., Toruń: Towarzystwo Naukowe w Toruniu, 224–226. Czeżowski, Tadeusz 1977. “Transcendtentalia – przyczynek do ontologii”, Ruch Filozoficzny 1–2, 54–56. Repr. in Czeżowski 2004, 188–192. Czeżowski, Tadeusz 2004. O metafizyce, jej kierunkach i zagadnieniach, 2nd ed., Kęty: Wydawnictwo Antyk. Gumański, Leon 1961. Elementy sądu a istnienie, Toruń: Towarzystwo Naukowe w Toruniu. Hylton, Peter 1990. Russell, Idealism, and the Emergence of Analytic Philosophy, Oxford: Clarendon Press.

CZEŻOWSKI’S CONCEPTION OF EXISTENCE

215

Jadacki, Jacek J. 1980. “Spiritus metaphysicae in corpore logicorum czyli o dziedzinie przedmiotowej języka i starej zagadce bytu”, Studia Filozoficzne 9, 111–140. Jadacki, Jacek J. 1996. Metafizyka i semiotyka, Warszawa: Wydział Filozofii i Socjologii UW. Łukasiewicz, Dariusz 2002. Filozofia Tadeusza Czeżowskiego, Bydgoszcz: Wydawnictwo Akademii Bydgoskiej. Łukasiewicz, Dariusz 2004. “Tadeusz Czeżowski’s Approach to the Intentionality and Ontology of Fiction”, Reports on Philosophy, 22, 142–161. Przełęcki, Marian 1980. “Nie ma tego, co nie istnieje”, Studia Filozoficzne, 9, 141– 148. Twardowski, Kazimierz 1894. Zur Lehre von Inhalt und Gegenstand der Vorstellungen. Eine Psychologische Untersuchung, Wien: Hölder. Eng trans. in Twardowski 1977; Polish trans. “O treści i przedmiocie przedstawień”, in Twardowski 1965, 3–91. Twardowski, Kazimierz 1911. “O czynnościach i wytworach – Kilka uwag z pogranicza psychologii, gramatyki i logiki”, Kraków: Gubrynowicz i syn, 1911, pp. 33 (extract from Księga pamiątkowa ku uczczeniu 250-tej rocznicy założenia Uniwersytetu Lwowskiego przez króla Jana Kazimierza vol. II, Lwów, Nakładem Uniwersytetu Lwowskiego, 1912) pp. 33. Eng. trans. in On Actions, Products and Other Topics in philosophy, Johannes Brandl and Jan Woleński eds., Amsterdam, Rodopi, 1999, 103–132. Twardowski, Kazimierz 1927. “O idio- i allogenetycznych teoriach sądu”, in Kazimierz Twardowski, Rozprawy i artykuły filozoficzne, Lwów: Księgarnia S.A. Książnica-Atlas T.N.S.W. 418–420. Second edition in Twardowski 1965, 198– 200. Twardowski, Kazimierz 1965. Wybrane pisma filozoficzne, Warszawa: PWN. Twardowski, Kazimierz 1975. “Teoria poznania”, Archiwum Historii Filozofii i Myśli Społecznej, 21, 244–299. Twardowski, Casimir 1977. On the Content and Object of Presentation. A Psychological Investigation, (translated by Reinhardt Grossmann), The Hague: Martinus Nijhoff. Wiśniewski, Ryszard 1989. “Doświadczenie aksjologiczne a teorie etyczne w koncepcji Tadeusza Czeżowskiego”, in Z. Czarnecki, S. Soldenhoff (eds.) 1989, Człowiek i wartości moralne. Studia z dziejów niezależnej myśli etycznej, Lublin: Wydawnictwo Lubelskie, 259–287. Woleński, Jan 1990. “Marty and the Lvov-Warsaw School” in Mulligan Kevin (ed.) Mind, Meaning and Metaphysics, Dordrecht: Kluwer Academic Publishers, 215– 223. Woleński, Jan 2004. “Malum, Transcendentalia and Logic” in Chrudzimski and Huemer 2004, (eds.), 359–370.

Brentanism and the Rise of Formal Semantics JAN WOLEŃSKI

Formal semantics as a part of logic arose in Poland mainly as a result of Tarski’s impetus. In order to explain what is meant by ‘formal’ in this context, some conceptual distinctions are useful (I omit a discussion of ‘formal’ as used in grammatical theories of natural language considered as a part of linguistics; I will be concerned solely with logical semantics). It is customary to distinguish semantics sensu largo and semantics sensu stricto. The latter is usually conceived as the relation between language and what language is about. In the wide sense, semantics consists of syntax, semantics sensu stricto and pragmatics. This tripartite division of semantics sensu largo was introduced by Morris (see Morris 1938, p. 84). Carnap (see Carnap 1939, p. 146) gives the following characterization of particular subfields of semantics in the wide sense: If in investigations explicit reference is made to the speaker, or to put it in more general terms, to the user of a language, then we assign it to the field of pragmatics. […]. If we abstract from the user of the language and analyze only the expressions and their designata, we are in the field of semantics. And, if finally, we abstract from the designata also and analyze only the relations between the expressions, we are in (logical) syntax. The whole science of language, consisting of the three parts mentioned, is called semiotic.

An illustration, simplified to some extent, can be offered in order to clarify some essential features of semantics sensu largo. Proceeding from simpler to more complex situations we have, first, language itself: that is, a collection of expressions. To give the syntax of such a collection, one must describe expressions in terms of their admissible forms and of the relations between them. For example if we say that an expression ‘A’ occurs in (or is a part of) the expression ‘A ∧ B’, we make a statement about the syntax of the language of propositional calculus. Let us agree to call such statements ‘syntactic’. Similarly, if we say that ‘London’ is the first word in the sentence ‘London is the capital of UK’, we make a syntactic statement about ordinary English. In general, syntactic statements concern relations within Actions, Products, and Things. Brentano and Polish Philosophy. Arkadiusz Chrudzimski and Dariusz Łukasiewicz (eds.), Frankfurt: ontos, 2006, 217–232.

218

JAN WOLEŃSKI

languages. Secondly, if one says that ‘London’ refers to London or that the form ‘x is a prime number’ is satisfied by the number 2, one makes semantic statements about English. Finally, pragmatic statements take into account the attitudes of users of languages expressed by such words as ‘asserting’, ‘asking’ or ‘expressing’. For example, the sentences ‘Russell asserted that Wittgenstein was a genius’ or ‘Heidegger asked whether Das Nichts nichtet’ represent pragmatic statements about English and German. Semantics is also conceived as the theory of meaning. This creates a confusion pointed out by Quine (see Quine 1953, p. 130): When the cleavage between meaning and reference is properly heeded […], the problems of what is loosely called semantics become separated into two provinces so fundamentally distinct as not to deserve a joint appellation at all. They may be called the theory of meaning and the theory of reference. ‘Semantics’ would be a good name for the theory of meaning, were it not for the fact that some of the best works in so-called semantics, notably, Tarski’s, belong to the theory of reference. The main concepts in the theory of meaning, apart from meaning itself, are synonymy (or sameness of meaning), significance (or possession of meaning), and analyticity (or truth in virtue of meaning). Another is entailment, or analyticity of the conditional. The main concepts in the theory of reference are naming, truth, denotation (or truth-of), and extension. Another is the notion of values of variables.

What Tarski himself understood by semantics is indicated in the following passages (Tarski 1933/1936; the Polish original appeared in 1933, its German version was published in 1936, the English translation is based on the German text, p. 252, 1936a, p. 401), which clearly suggests that he thought about the theory of reference: […] we attempted to go further and to construct […] definitions of certain concepts belonging to another domain, namely that called the semantics of language – i.e. such concepts as satisfaction, denoting, truth, definability, and so on. A characteristic feature of the semantical concepts is that they give expression to certain relations between the expressions of language and the objects about which these expressions speak, or that by means of such relations they characterize certain classes of expressions or other objects. We could also say (making use of the suppositio materialis) that concepts serve to set up the correlation between names of expressions and the expressions themselves.

BRENTANISM AND THE RISE OF FORMAL SEMANTICS

219

The word ‘semantics’ is used here in a narrower sense than usual. We shall understand by semantics the considerations concerning those concepts which, roughly speaking, express certain connexions between the expressions of a language and the objects and states of affairs referred to by these expressions. As typical of semantical concepts we can mention the concepts of denotation, satisfaction, and definition, […]. The concept of truth also – and this is not commonly recognized – is to be included here, at least in its classical interpretation.

The distinctive feature of logical semantics consists in employing methods derived from logic and mathematics. Roughly speaking, this field in its contemporary understanding studies relations holding between formalized languages (or better: theories expressed in such languages) and their models generated by interpretations of logical and extra-logical signs. A short narrative of the birth formal semantics would run as follows (see Woleński 1999 for a more detailed account). Philosophers were always interested in language and its relations to the world. These interests became much deeper after the birth of mathematical logic, which provided a suitable formal framework for semantic investigations. Frege and Russell, the fathers of mathematical logic, made essential contributions to semantics and philosophy of language, but did not achieve a general theory (sense, reference, descriptions, etc.). Wittgenstein and the Vienna Circle continued their research and even limited philosophy to considerations about language. However, under the influence of Hilbert and his program of formal metamathematics, logical empiricists limited their work to syntax; I am speaking about the early Vienna Circle – the situation changed in the midthirties, when Carnap converted to semantics under Tarski’s influence. Moreover, the early stage did not sufficiently distinguish syntax, semantics and pragmatics. The discovery of antinomies made semantics suspicious: it could lead to inconsistencies. Tarski, continuing and combining ideas stemming from Russell’s theory of logical types and some views of Leśniewski, found a successful way of solving antinomies by the language/metalanguage distinction. It was the key to defining semantic notions for a fixed language L in its metalanguage; the semantic definition of truth became a paradigmatic case. These facts and achievements opened the gate for logical semantics as a separate and well-established branch of logic and philosophy.

220

JAN WOLEŃSKI

Tarski himself shared this picture and diagnosis (Tarski 1933/1936, p. 252, Tarski 1936a, p. 401): For a long time the semantical concepts have had an evil reputation among specialists in the study of language. They have resisted all attempts to define their meaning exactly, and the properties of these concepts, apparently so clear in their content, have led to paradoxes and antinomies. Concepts from the domain of semantics have traditionally played a prominent part in the discussions of philosophers, logicians and philologists. Nevertheless they have long been regarded with a certain scepticism. From the historical point of view this scepticism is well founded; for, although the content of the semantical concepts, as they occur in colloquial language, is clear enough, yet all attempts to characterize this content more precisely have failed, and various discussions in which these concepts appeared and which were based on quite plausible and seemingly evident premises, have often lead to paradoxes and antinomies. It suffices to mention here the antinomy of the liar, the Grelling-Nelson antinomy of heterological terms and the Richard antinomy of definability.

After twenty years, he also stressed the role of yet another factor (Tarski 1954, p. 713): As an essential contribution of the Polish school to the development of metamathematics one can regard the fact that from the very beginning it admitted into metamathematical research all fruitful methods, whether finitary or not.

In fact, the conscious access of Polish logicians to infinitistic tools of reasoning constituted an important circumstance in the development of formal semantics. For instance, Hilbert admitted infinitary concepts and methods in mathematics, but excluded them from metamathematics. Since semantics, at least associated with classical logic, requires infinitary techniques, it could not belong to metalogic, contrary to syntax, which can be done by purely combinatorial (effective, constructive, finitary) methods. It was one of the sources of syntacticism as a dogma of the Vienna Circle. I do not oppose the description of the birth of formal semantics presented above. However, I think that it does not provide the whole story because it neglects the philosophical climate at the time Tarski was working. One particular point should be perhaps stressed. Although nobody denies that antinomies, in particular Russell’s paradox of the set of all sets which

BRENTANISM AND THE RISE OF FORMAL SEMANTICS

221

do not belong to themselves, provided a serious impetus for logical research, yet it is a well-known fact that inconsistencies did not much impress Brouwer, Hilbert or Zermelo. Leśniewski (see Leśniewski 1929, p. 421; quoted below) maintained that the simple theory of logical types (its linguistic version is very close to Tarski’s hierarchy of languages and metalanguages and the theory of semantic categories) would have been invented even without the discovery of antinomies. Although we have no reasons to doubt Tarski’s claim that for him (as well as other logicians) semantic antinomies constituted a challenge and thereby contributed to the invention of formal semantics, other factors must be taken into account as well. Some have suggested that Husserl’s ideas played a very important role in Poland. Martin Kusch sees the history through “Husserlian” glasses (Kusch 1989, p. 60): Concerning investigations three and four [in Husserl’s Logical Investigations – J.W.] the first noteworthy fact relates to the historical role these studies played in the semantical approaches to formal logic. As is well-known, the main gate through which these ideas entered modern logic was the work of Tarski and other Polish logicians. […] it is remarkable that it was Husserl’s Logical Investigations and especially the third and fourth investigation that exerted a strong influence in Warsaw between the two wars. This influence has been described as comparable to the influence of Wittgenstein’s Tractatus in Vienna in the twenties and thirties. […]. It can be considered as indirect evidence for attributing to Husserl the calculus conception that a precise formal semantical theory was developed where his influence was the strongest. And this influence did not only remain on the abstract unspecified level. […]. Ajdukiewicz’s and Leśniewski’s seminal work on categorical grammar had as its starting point Husserl’s fourth investigation concerning the ideal logical grammar.

Clearly, the idea of semantic categories was very influential in Poland. It is explicitly documented by references in the works of Ajdukiewicz, Leśniewski and Tarski. For example, Leśniewski wrote (Leśniewski 1929, pp. 421–422): In 1922 I outlined a concept of semantical categories as a replacement for the hierarchy of types, which is quite unintuitive to me. Frankly, I would still feel obliged to accept this concept even if there were no antinomies at all. From a formal point

222

JAN WOLEŃSKI of view, my concept of semantic categories is closely related to the well-known theory of types […] especially for their theoretical consequences. Intuitively, however, the concept is more easily related to the thread of tradition running through Aristotle’s categories, the parts of speech of traditional grammar, and Husserl’s meaning categories.

Yet I do not agree that, inasmuch as the general philosophical environment is concerned, Husserl’s influence in Poland contributed very much to the actual development of semantics. In fact, the fathers of the Polish school of logic had several and rather strong reservations concerning Husserl. Certainly, he was not regarded in Poland as a pattern of clarity and rigour. Leśniewski perceived his own development in the following way (Leśniewski 1927, p. 181): Stepped in the influence of John Stuart Mill in which I mainly grew up, and ‘conditioned’ by the problems of ‘universal-grammar’ and of logico-semantics in the style of […] Husserl and by the exponents of so-called Austrian School, I intellectually attacked the foundations of ‘logistic’ from this point of view.

It is clear that Leśniewski very strongly contrasted the Husserlian version of semantics with that based on logistic (= mathematical logic). The most critical Polish opinion (coming from the Lvov-Warsaw school) about Husserl is voiced by Łukasiewicz, although he followed Husserlian criticism of psychologism in logic at the beginning of his philosophical career. According to Łukasiewicz (Łukasiewicz 1949, p. 57/58): The first volume of Husserl’s Logical Investigations impressed the Lvov philosophical circle very much; in particular, it impressed me [it is an allusion to the earlier mentioned problem of psychologism – J.W]. I had not liked psychologism for some time, and I entirely rejected this view after reading Husserl. However, I became disappointed with the second volume of the Logical Investigations. Once more I encountered the obscure philosophical talk which always repelled me from German philosophers. I wondered that such a deep difference could occur between two volumes of the same work. Later, I realized that it was not Husserl who spoke to me in the first volume of the Logical Investigations, but someone else, who was used by Husserl in his book and was much greater than he, namely Gottlob Frege.

BRENTANISM AND THE RISE OF FORMAL SEMANTICS

223

Of course, it might be that Leśniewski’s and Łukasiewicz’s critical remarks about Husserl and phenomenology were exaggerated. Contemporary commentators very often see Husserl as much closer to formal or even mathematical logic than was the case in the 1920‘s. I will not address this uneasy interpretative problem, except to remark that the recently published collection of Husserl’s lectures about logic (see Husserl 2001, Husserl 2001a, Husserl 2003) seem to testify to the fact that his knowledge of mathematical logic was rather poor and not up to date. Leaving aside the question of Husserl’s maturity in logic, its problems and development, Leśniewski’s and Łukasiewicz’s opinions about Husserl, even if they are unfair to some extent, give rather strong evidence that a direct Husserlian influence on the Polish logical school should not be exaggerated but restricted to those points explicitly noted in relevant sources, as in the case of the concept of semantic category and criticism of psychologism in logic and its philosophy. As far as the matter concerns the concept of language as calculus, as contrasted with the concept of language as universal medium (this point is raised by Kusch), there was more of a parallelism (see Woleński 1997) than a following of Husserl in Poland. In fact, the Polish school thought of language as a reinterpretable calculus, the idea entirely rejected by Husserl. I take it that the generally favourably philosophical climate for the rise and development of logical semantics in Poland was connected with the Brentanism, introduced into Polish philosophy by Twardowski. This claim must be properly understood. I do not suggest that Polish logicians used Brentano’s formal views, for example, his interpretation of all sentences as existential or that they shared all his more general philosophical opinions. The quotation from Leśniewski (see above) clearly expresses his criticism of the Austrian school and its idea of grammar. Łukasiewicz accused Brentano of strong psychologism and modern scholasticism. In fact, Łukasiewicz regarded Bolzano as a much better philosopher than Brentano. Neither do I suggest that Tarski was quite conscious of his Brentanist roots. But the following circumstances should be noted. Tarski’s rather deep interests in philosophy are well known (see Woleński 1993, Woleński 1995 on Tarski as a philosopher). Łukasiewicz and Leśniewski, Tarski’s main teachers, studied with Twardowski, began their work as philosophers and can arguably be regarded as Brentano’s grandsons at least in their early

224

JAN WOLEŃSKI

academic careers (i.e. before they became mathematical logicians). Hence, it is not implausible that Tarski was indeed influenced by ideas going back to Brentano. We have a document in which Tarski points out his connection with the Twardowskian tradition (Tarski 1992, p. 20): Almost all researchers, who pursue the philosophy of exact sciences in Poland are indirectly or directly the disciples of Twardowski, although his works could hardly be counted within this domain.

Doubtless, the author of these words had to consider himself, at most, an “indirect disciple” and although the quoted passage says nothing about his views but rather stresses a sociological fact, it gives indirect evidence about Tarski’s intellectual connections. On the other hand, it seems that Tarski followed Twardowski very closely in at least one question, although he does so without mentioning the latter. Let me examine the following passage (Tarski 1933/1936, p. 253; bold mine): In the course of our investigations we have repeatedly encountered similar phenomena: the impossibility of grasping the simultaneous dependence between objects which belong to infinitely many semantical categories; the lack of terms of ‘infinite order’; the impossibility of including, in one process of definition, infinitely many concepts, and so on. […]. I do not believe that these phenomena can be viewed as a symptom of the formal incompleteness of the actually existing phenomena – their cause is to be sought rather in the nature of language itself: language, which is a product of human activity, necessarily possesses a ‘finitistic’ character, and cannot serve as an adequate tool for the investigation of facts, or for the construction of concepts, of an eminently ‘infinitistic’ character.

Twardowski, drawing on his early studies in descriptive psychology within the Brentanian program, had introduced a very important distinction, namely that of actions and products, which became standard in Poland. It was applied by him as well as by his students to various problems, in particular for the purpose of explaining how language works. This was a generalization of another distinction of his, namely that between the content and the object of a presentation. Thus, we can say that Tarski employed the concepts of actions and products when he elaborated the limitations of finitary means in covering infinitistic phenomena. Although he did not men-

BRENTANISM AND THE RISE OF FORMAL SEMANTICS

225

tion Twardowski, Tarski had to know who introduced the distinction in question. Now if we find that Tarski followed Twardowski once, why do not look for other examples, even if we have no written documentation? Two things come to mind in this context. The first point consists in general acceptance by Polish philosophers and logicians of Brentatno’s view that mental phenomena are intentional, that is, directed to objects. Since, according to most Polish philosophers, these phenomena manifest themselves exclusively or at least mainly through linguistic performances, language and its elements are also intentional in the sense that particular expressions (at least those serving as names, predicates and sentences) refer to the objects in the world. Applying Twardowski’s distinction, expressions can be considered products of definite actions which supervene on them. Thus, if actions are intentional, this feature should also be present in their products. The conceptual succession is this: linguistic acts cum fundamentum in mental acts (actions) → linguistic expressions. This Brentanian picture shows how semantic features of expressions are rooted in referential (intentional) properties of linguistic actions. The perspective here is much more Brentanian than Husserlian because, at least in Twardowski’s interpretation of intentionality, intentional objects are mostly real. Eventually, one can compare this picture with the Husserlian theory of meaning-intention (see Woleński 1997), although Husserl’s position was much more sophisticated (which does not mean that it was better). Anyway, Twardowski himself takes this route in his treatment of names as counterparts of presentations (see Twardowski 1894; see Woleński 2002 for further details) and it is plausible to think that his analysis, although modest when compared to the investigations of his followers, provided a pattern of doing semantics. The next point concerns the thesis on the primacy of the intentional stated by Chisholm (Chisholm 1986, p. 13/14): I have sometimes used the expression, “the thesis of the primacy of the intentional”, by which I mean the view that the intentional relations involved in using language are to be understood in terms of the intentional relations involved in thinking, and not conversely.

226

JAN WOLEŃSKI

Although no Polish philosopher formulated this thesis in a precise manner, I think that it can be attributed to Twardowski himself as well as to his older students (notably, Ajdukiewicz, Czeżowski, Kotarbiński, Leśniewski, Łukasiewicz), at least before the Lvov-Warsaw School became mostly influenced by the new logic. The last reservation does not mean that Polish logicians rejected the thesis on the primacy of the intentional but rather that they lost interests in psychological considerations. Tarski himself probably had no view about the relation between psychological intentionality and referential relations of terms and sentences. On the other hand, he certainly accepted the thesis of the primacy of semantics over syntax: the semantic relations involved in using language are primary and not reducible to syntactic properties of linguistic expressions. Although Tarski’s view is not an exact counterpart to Chisholm’s thesis, the path from intentionality to the essential features of contemporary thinking about language is fairly clear in this perspective. This feature consists in the fundamental place of semantics in metalogic. Thus, it is possible that Brentanism contributed to the semantic revolution of the 1930’s. The second point concerns the role of meaning in semantics. It is a subtle question and depends on how the domain of logical semantics is delineated. If both parts of semantics mentioned by Quine (theory of reference and theory of meaning) are admitted, formal analysis covers reference as well as meaning, but when only referential relations are taken into consideration, the concept of meaning is ruled out. The restriction of the field of logical semantics to the theory of reference is understandable, because the concept of meaning involves the well-known logical and philosophical difficulties. Nevertheless, it is difficult to imagine philosophical work ignoring what linguistic expressions mean. If we assume that the tripartite division of semantics sensu largo into syntax, semantics and pragmatics is exhaustive, only three possibilities remain: (a) the concept of meaning is added to syntax; (b) the concept of meaning is added to semantics sensu stricto; or (c) the concept of meaning is added to pragmatics. The first possibility was attempted by logical empiricists, but without success. Solution (b) requires either the concept of meaning as a new semantic primitive or its reduction to the concept of reference. It seems that supplementing the semantic vocabulary by the concept of meaning introduces dualism to semantics sensu stricto, which is neither elegant nor easy to explain. At the

BRENTANISM AND THE RISE OF FORMAL SEMANTICS

227

formal level, this situation motivates logical intensional semantics, but I ignore this path (it raises several difficulties, but I cannot enter into details here). On the other hand, the reduction of the concept of meaning to referential relations is at odds with the well-known fact that various intensions (meanings) can correspond to the same reference (extension). If (a) and (b) are rejected, the only solution consists in viewing meaning as a pragmatic category. In this perspective, the meaning of expressions is the property determining how they are understood in communication. Probably all known theories of meaning (mentalistic, behaviouristic, referential, subjective, objective, etc.) sooner or later agree that the question as to how people understand expressions constitute the main test for identification of meanings. Tarski clearly saw the problem of meaning and pointed out this fact quite explicitly (Tarski 1933/1936, p. 166/167): It remains perhaps to add that we are not interested here in ‘formal’ language and sciences in one special sense of the word ‘formal’, namely sciences to the signs and expressions of which no material [in Polish original ‘intuicyjnego’, that is, ‘intuitive’ – J.W.] sense is attached. For such sciences the problem here discussed [the problem of truth – J.W.] has no relevance, it is not even meaningful. We shall always ascribe quite concrete and for us, intelligible meanings to signs which occur in them have been translated into colloquial language. The sentences which are always guided by the principles that when such rules are applied to true sentences the sentences obtained by their use should also be true.

This passage is important for understanding Tarski’s position. And it is also puzzling given Tarski’s treatment of semantics as referential in Quine’s sense. I will not enter into the question as to how Tarski understood the concept of meaning, except to make a few remarks. He was sceptical about the prospects of a satisfactory definition of meaning. For this reason, Tarski relativised semantic notions to language, taking for granted that it consists of meaningful expressions. One can guess that he approved of a pragmatic conception of meaning, but left its clarification for philosophers. Yet Tarski seriously took the meaning-parameter of language to explain why semantic problems are intelligible at all for interpreted languages.

228

JAN WOLEŃSKI

But what does it mean that expressions have an intuitive sense and that we ascribe to signs “quite concrete and, for us, intelligible meanings”? Although we can say that referential relations presuppose that expressions are equipped with meaning, this wording does not explain very much. Doubtless, Tarski’s treatment of language and meaning was strongly influenced by the following Leśniewskian tenet (Leśniewski 1929, p. 487): Having no predilection for ‘various mathematical games’ that consist in writing out according to one or another conventional rule various more or less picturesque formulae which need not be meaningful or even – as some of the ‘mathematical gamers’ might prefer – which should necessarily be meaningless, I would not have taken the trouble to systematize and to often check quite scrupulously the directives of my system, had I not imputed to its theses a certain specific and completely determined sense, in virtue of which its axioms, definitions and final directives […] have for me an irresistible intuitive validity. I see no contradiction therefore, in saying that I advocate a rather radical ‘formalism’ in the construction of my system even though I am an obdurate ‘intuitionist’. Having endeavoured to express my thoughts on various particular topics by representing them as a series of propositions meaningful in various deductive theories, and to derive one proposition from others in a way that would harmonize with the way I finally considered intuitively binding, I know no method more effective for acquainting the reader with my logical intuitions than the method of formalizing any deductive theory to be set forth. By no means do theories under the influence of such formalizations cease to consist of genuinely meaningful propositions which for me are intuitively valid. But I always view the method of carrying out mathematical deduction on an ‘intuitionistic’ basis of various logical secrets as considerably less expedient method.

This passage expresses various things, but intuitive (I prefer ‘intuitive’ over ‘intuitionistic’ in this context) formalism is perhaps the most important here. To put things simply, it gives a good general foundation for Tarski’s view that the formalized and the meaningful are not mutually at odds, but just coherent features of certain linguistic systems. If Tarski inherited his treatment of language and meaning from his most important teacher Leśniewski, much could be learned of the philosophical background of the former by taking into account the latter’s position. But, apparently, this is not an easy task, for Leśniewski strongly and

BRENTANISM AND THE RISE OF FORMAL SEMANTICS

229

even ceremonially condemned his early views as sins of youth (Lesniewski 1927, p. 197/198): Living intellectually beyond the sphere of the valuable achievements of the exponents of ‘Mathematical Logic’ and yielding to many destructive habits resulting from the one sided, ‘philosophical’ – grammatical culture, I struggled in the works mentioned [that is, published by him in 1911–1913 – J.W.] with a number of problems which were beyond my powers at that time, discovering already-discovered Americas on the way. I have mentioned those works desiring to point out that I regret that they have appeared in print, and formally ‘repudiate’ them herewith. Though I have already done this within the university faculty, affirming the bankruptcy of the ‘philosophical’-grammatical work of the initial period of my work.

Yet this self-criticism does not necessarily concern all of his “initial” positions. For instance, Leśniewski kept his early views about eternity and sempiternality of truth and the thesis that all sentences with empty names in the subject position are false. If we take the recently quoted passage cum grano salis, looking for Brentanian traces of intuitive formalism is not hopeless. In fact, it is not difficult to find them. According to Brentano, we grasp the content of mental acts directly as intuitively evident. Meanings or sense functions, according to Polish Brentanists, are the linguistic counterpart to the content of mental acts and this parallelism sufficiently supports the thesis that the former is also intuitively grasped. If we have no reasons to doubt what expressions mean or whether they are correct from a certain point of view, for instance, as far as the matter concerns their coherence (it is related to the problem of antinomies), we can rely on our linguistic intuition. Kotarbiński very nicely formulated a similar point (Kotarbiński 1966, p. 408): Logical semantics is also deeply interested in the problem of the essence of the a priori, in doing which it analyses the supposition that all a priori knowledge is nothing but self-evident knowledge that can be derived without recourse to empirical data.

Although Kotarbiński’s remark does not concern semantics but rather epistemology, we find here the same trust in the self-evidence of meaning as is the case in Leśniewski or Tarski.

230

JAN WOLEŃSKI

To close, I do not think that the problem of how meaning matters to semantics sensu stricto is only of historical interest. Doubtless, it does not influence formal work in semantics. For general studies, it is enough to assume that we have a fixed interpretation of language in question, but if a concrete model is investigated, it is always given in advance. Thus, the problem of how an interpretation is generated has no real importance for working model-theorists and they can say that logical semantics investigates purely referential relations between formalized theories and their models. However, philosophers are mostly interested in a more general setting of formal semantics. This also involves the question as to how we reach an interpretation. It is a trivial fact that any interpretation of a language L must be expressed in its metalanguage ML, that is, in another, richer linguistic apparatus. Another trivial fact consists in the impossibility of formalizing everything. However, we encounter here a non-trivial consequence that the interpretation of L does not suffice for the full understanding of the vocabulary of ML, which must remain partly informal and used in accordance with intuitions, eventually corrected by additional steps. Radical conventionalism does not offer a sound solution, because even if we say that an interpretation is arbitrary, we must know what the words used in conventions mean. It seems that conventionalism inevitably collapses into a vicious circle. Perhaps it is no accident that Leśniewski represented a decisive anti-conventionalist attitude. He was not against introducing conventions as governing linguistic usages but rather against conventionalism as the ultimate explanation of how languages work. The same attitude can be attributed to Tarski and it has clear Brentanist roots. I do not claim that this philosophy of semantics explains or solves everything. In particular, passing from psychological certainty to “an irresistible intuitive validity” is still a problem. PS. I am grateful for Sandra Lapointe for helping me in preparing this paper.

BRENTANISM AND THE RISE OF FORMAL SEMANTICS

231

References Carnap, Rudolf 1939. The Foundations of Logic and Mathematics (International Encyclopedia of the Unified Science, ed. by O. Neurath, R. Carnap and Ch. Morris, vol. 1, no 3), University of Chicago Press, Chicago; 2nd ed. (the cloth edition of the whole Encyclopedia), University of Chicago Press, Chicago 1955. Chisholm Roderick M. 1986. “Self-Profile”, in Roderick M. Chisholm, ed. By R. Bogdan, D. Reidel Publishing Company, Dordrecht, pp. 3–77. Husserl, Edmund 2001. Logik Vorlesung 1896, ed. by E. Schumann, Kluwer Academic Publishers, Dordrecht. Husserl, Edmund 2001a. Logik Vorlesung 1902/03, ed. by E. Schumann, Kluwer Academic Publishers, Dordrecht. Husserl, Edmund 2003. Logik Vorlesung 1908/09, ed. by E. Schumann, Kluwer Academic Publishers, Dordrecht. Kotarbiński, Tadeusz 1966. Gnosiology. The Scientific Approach to the Theory of Knowledge, Oxford, Pergamon Press, Wroclaw, Ossolineum. Kusch, Martin 1989. Language as Calculus vs. Language as Universal Medium. A Study in Husserl, Heidegger and Gadamer, Kluwer Academic Publishers, Dordrecht. Leśniewski, Stanisław 1927. “O podstawach matematyki I” (On the Foundations of Mathematics), Przegląd Filozoficzny XXX, pp. 164–206; Eng. tr. in Leśniewski 1992, pp. 174–226. Leśniewski, Stanisław 1929. “Grundzüge eines neuen Systems der Grundlagen der Mathematik”, Fundamenta Mathematicae”, XIV, pp. 1–81; Eng. tr in Leśniewski 1992, pp. 410–605. Leśniewski, Stanisław 1992. Collected Works, ed. and tr. by S. J. Surma, J. Strzednicki, and D. I. Barnett, Kluwer Academic Publishers, Dordrecht. Łukasiewicz, Jan 1949. Diary (unpublished) (deposited in the Archive of the University of Warsaw), Warszawa. Morris, Charles 1938. The Foundations of the Theory of Signs (International Encyclopedia of the Unified Science, ed. by O. Neurath, R. Carnap and Ch. Morris, vol. 1, no 2), University of Chcicago Press, Chicago. Quine, Willard Van Orman 1953. From a Logical Point of View, Harvard University Press, Harvard. Tarski, Alfred 1933. Pojęcie prawdy w językach nauk dedukcyjnych (The Concept of Truth in Languages of Deductive Sciences), Towarzystwo Naukowe Warszawskie, Warszawa. Tarski, Alfred 1936. “Der Wahrheitsbegriff in den formalisierten Sprachen”, Studia Philosophica I, pp. 152–278; repr. in Tarski 1986, vol. 2, pp. 51–198; Eng. tr. by J. H. Woodger in Tarski 1983, pp. 152–278.

232

JAN WOLEŃSKI

Tarski, Alfred 1936a. “Grundlegung der wissenschaftlichen Semantik”, Actes du Congrés International du Philosophie Scientifique, vol. 3, Hermann, Paris 1938, pp. 1–8; repr. In Tarski 1986, vol. 2, pp. 259–268; Eng. tr. (by J. H. Woodger) in Tarski 1983, pp. 401–408. Tarski, Alfred 1954. “[Contribution to the Discussion of P. Bernays, Zur Beurteilung der Situation in in der beweistheoretischen Forschung]”, Revue Internationale de Philosophie 8, p. 16–20; repr. in Tarski 1986, vol. 4, pp. 713–714. Tarski, Alfred 1983. Logic, Semantics, Mathematics. Papers from 1923 to 1938, 2nd ed., ed. by J. Corcoran, Hackett Publishing Company, Indianapolis. Tarski, Alfred 1986. Collected Papers, vol. 1–4, ed. by S. R. Givant and R. N. McKenzie, Birkhäuser, Basel. Tarski, Alfred 1992. “Drei Briefe an Otto Neurath” (with Eng. tr. by J. Tarski), Grazer Philosophische Studien 43, pp. 1–32. Twardowski Kasimir 1894. Zur Lehre vom Inhalt und Gegenstand der Vorstellungen. Eine psychologische Untersuchung, Hölder; sec. ed., Philosophia Verlag, München; Eng. tr. By R. Grossmann, On the Content and Object of Presentation, M. Nijhoff, Den Haag 1977. Twardowski, Kasimir 1912, “O czynnościach I wytworach. Kilka uwag z pogranicza psychologii, gramatyki I logiki” (On Actions and Products. Some Remarks from the Boderline of Psychology, Grammar and Logic), in Księga Pamiątkowa ku uczczeniu 250 rocznicy założenia Uniwersytetu Lwowskiego przez króla Jana Kazimierza (The Memorial Book for Celebration the 250th Anniversary of Establishing the Lvov University by the King John Casimir II, Uniwersytet Lwowski, Lwów; Eng. tr. by A. Szylewicz, in K. Twardowski, On Actions, Products and Other Topics in Philosophy, ed. by J. Brandl and J. Woleński, Rodopi, Amsterdam 1999, pp. 104–132. Woleński, Jan 1993. “Tarski as a Philosopher”, in Polish Scientific Philosophy: The Lvov-Warsaw School, ed. by F. Coniglione, R. Poli, and J. Woleński, Rodopi, Amsterdam, pp. 319–338. Woleński, Jan 1995. “On Tarski’s Background”, in From Dedekind to Gödel, ed. by J. Hintikka, Kluwer Academic Publishers, Dordrecht , pp. 331–341; repr. in Woleński 1999a, pp. 126–133. Woleński, Jan 1997. “Husserl and the Development of Formal Semantics”, Philosophia Scientiae. Travaux d’ histoire et de philosophie des sciences v. II, cahier 4, pp. 151– 158. Woleński, Jan 1999. “Semantic Revolution: Rudolf Carnap, Kurt Gödel, Alfred Tarski”, in J. Woleński and E. Köhler, Tarski and the Vienna Circle, Kluwer Academic Publishers, Dordrecht , pp. 1–15. Woleński, Jan 2002. “From Intentionality to Formal Semantics (From Twardowski to Tarski)”, Erkenntnis 56, pp. 9–27.

Contributors Arianna Betti (born 1970) studied philosophy in Italy, Poland and Austria. Currently a postdoctoral researcher at the Vrije Universiteit Amsterdam, her interests centre on Austrian and Polish Philosophy, the history and philosophy of logic, and metaphysics. She is working on a book on truth and time both from a systematic and a historical perspective. Arkadiusz Chrudzimski (born 1967) is Associate Professor of Philosophy at the University of Szczecin and Research Assistant at the University of Salzburg. He is the autor of: Die Erkenntnistheorie von Roman Ingarden (Kluwer 1999), Intentionalitätstheorie beim frühen Brentano (Kluwer 2001), Die Ontologie Franz Brentanos (Kluwer 2004), and Intentionalität, Zeitbewusstsein und Intersubjektivität. Studien zur Phänomenologie von Brentano bis Ingarden (Ontos-Verlag 2005). He is currently working on a book on Meinong’s Gegenstandstheorie and on a defense of the ontology of intentional objects. Dale Jacquette is Professor of Philosophy at The Pennsylvania State University. He is the author of Philosophy of Mind (1994), Meinongian Logic: The Semantics of Existence and Nonexistence (1996), Wittgenstein’s Thought in Transition (1998), and On Boole (2002), and has recently edited The Cambridge Companion to Brentano (2004). Dariusz Łukasiewicz, born 1965, obtained his Ph.D. 1995 at the Jagiellonian University (Kraków, Poland) and his habilitation 2003 at the same university. Associate Professor at Casimir the Great University in Bydgoszcz (Poland). He has published articles in analytic phenomenology, the Lvov-Warsaw School philosophy and philosophy of religion. Books: Filozofia Tadeusza Czeżowskiego (Tadeusz Czeżowski’s Philosophy), Stany Rzeczy i Prawda (States of Affairs and Truth). He is currently working on Polish Brentanism and the philosophy of time in contemporary analytic philosophy of religion. Maria van der Schaar is lecturer at the faculty of philosophy of Leiden university. She wrote her Ph.D. thesis in 1991 under Gabriel Nuchelmans

234

CONTRIBUTORS

on the influence of Brentano and Twardowski on G. F. Stout and analytic philosophy. She is working on the theory of judgement both from a systematic and a historical point of view. She has published, for example, on judgement and philosophy of logic in the writings of Brentano (Brentano Studien, 10, 2002/03), Twardowski (Axiomathes, 3, 1996, and, together with Arianna Betti, Grazer Philosophische Studien, 67, 2004), Stout (Poznań Studies, 82, 2004) and Chisholm (Library of Living Philosophers on Chisholm, 25, 1997), and is currently doing systematic work on assertion, the linguistic counterpart of judgement, and knowledge. Peter Simons studied Mathematics and Philosophy at Manchester. He worked at the University of Salzburg from 1980 to 1995, since when he has been Professor at the University of Leeds. Besides Austrian and Polish philosophy, he specializes in metaphysics and the philosophy and history of logic. He is Director of the Franz Brentano Foundation. Barry Smith is the author of some 400 scientific publications, including 17 authored or edited books. He is also the editor of The Monist: An International Quarterly Journal of General Philosophical Inquiry. His current research is in the areas of ontology, and specifically on applications of ontology in the field of biomedical informatics. In 2001 he was awarded the Wolfgang Paul Prize of the Alexander von Humboldt Foundation. Jan Woleński, born in 1940, obtained his Ph.D. 1968 at the Jagiellonian University (Kraków, Poland) and his habilitation 1972 at the same university, Professor of Philosophy at the Jagiellonian University since 1990, current research: philosophical logic, epistemology, history of logic and philosophy, particularly in Poland. Books in English: Logic and Philosophy in the Lvov-Warsaw School (Dordrecht: Kluwer 1999), Essays in the History of Logic and Logical Philosophy (Kraków: Jagiellonian University). He is a member of Institut Internationale de Philosophie and the Polish Academy of Arts and Sciences.

Index of Names

Ajdukiewicz, K., 56, 68, 84, 97, 103f., 125f., 171, 221, 226 Albert the Great, 176 Albertazzi, L., 30 Anselm of Canterbury, 183f. Aquinas, Thomas, 50–52, 162, 175f., 183f. Aristotle, 11, 26f., 50, 52, 64f., 70, 92, 94, 129, 138f., 141f., 147, 150, 156f., 160f., 164, 166f., 169, 171, 173, 175f., 183–185, 198 Avé-Lallemant, E., 76 Batóg, T., 72, 73, 77 Bednarowski, W., 175 Bell, D., 74 Bergmann, H., 58 Bergmann, J., 51 Bergson, H., 152 Berkeley, G., 11, 12f., 15, 26, 31 Bernays, P., 118 Betti, A., 51, 72f., 76f. Biegański, W., 72, 77 Bolzano, B., 37, 44, 49, 55–81, 168, 174, 223 Boole, G., 66, 97 Boolos, G., 103, 104 Borowski, M., 67 Brandl, J., 51 Brentano, F. 9–33, 38, 43f., 51, 55–57, 60, 64, 67, 71f., 77, 83, 86, 89f., 92–94, 96, 98, 101f., 103f. , 107f., 115–119, 121–125, 133, 157–170, 172–176, 183, 185–187, 189–194, 196–198, 200, 203f., 207–209, 211, 213, 223–226, 229f. Brouwer, L. E. J., 221 Brugmann, K., 51 Buczyńska-Garewicz, H., 51 Butchvarov, P., 208, 213 Cajetan, Thomas 175 Cantor, G., 75, 83, 123, 171 Canty, J. T., 173

Carnap, R., 95, 171, 217, 219 Carroll, L., 118 Chrudzimski, A., 72, 77, 102, 185, 192, 211, 214 Chisholm, R. M., 141, 165, 173–175, 225f. Chwistek, L., 70, 76 Cohen, R., 170 Comte, A., 176 Cornelius, H., 175 Coutrant, L., 75, 200, 201 Czeżowski, T., 57, 183–215, 226 Davidson, D., 137, 170, 173 Dąmbska, I., 52 Dedekind, J. W. R., 75 Descartes, R., 26, 133, 165 Drewnowski, J. F., 73, 77 Dummett, M., 35 Einstein, A., 167 Exner, F., 73 Findlay, J. N., 31f. Frege, G., 16, 22, 32, 35f., 83f., 89, 95, 97, 101, 112, 200f., 219 Fraenkel, A., 59 Fränklówna, M., 73, 77 Frege, G., 57, 63, 76, 124, 168, 171 Galton, A., 175 Głombik, C., 76 Goodman, N., 167f. Gödel, K., 76 Griffin, N., 32 Grossmann, R., 29f. Grzegorczyk, A., 163 Gumański, L., 187, 191, 209, 210, 213 Hamilton, W., 66 Heidegger, M., 218 Heraclitus, 152 Hilbert, D., 118, 219–221 Hoeksema, J., 175 Höfler, A., 10, 13, 24, 27, 29f., 61, 74, 172

236

INDEX OF NAMES

Hume, D., 11–16, 18, 26–29, 31, 152, 163, 206f., 211 Husserl, E., 23, 32, 38, 42, 44, 46, 55, 57f., 61–63, 65f., 69, 73f., 76, 83, 89, 121, 123, 128, 162, 170f., 175, 221– 223, 225. Hylton, P., 200 Ingarden, R., 29, 72, 74, 77, 160, 167, 173–177, 201, 213 Jacquette, D., 29, 30, 32 Jadacki, J. J., 72, 77, 209, 212 Jadczak, R., 74f. Jespersen, B., 72, 77 Johnson, W. E., 152f. Kant, I., 113f., 206f., 210f. Kastil, A., 175 Katkov, G., 175 Kerry, B., 60, 61, 68, 73 Kerry, B., 10, 13, 19, 23, 27, 30, 60f., 68, 73 Koffka, K., 173 Kotarbiński, T., 56f., 59, 69, 72f., 75, 77, 83f., 91, 96, 102, 121–182, 226, 229 Köhler, W., 173 Kraus, O., 175 Küng, G., 138, 173f. Künne, W., 72f., 75, 77 Kusch, M., 221, 223 Lapointe, S., 230 Lambert, K., 32 Leblanc, A., 170 Leibniz, G. W., 64, 158, 164, 167, 175 Lejewski, C., 96f., 103, 125f., 132, 144, 152, 170–172, 174, 177 Leśniewski, S., 55–57, 59–62, 66–73, 75– 77, 83–124, 128, 133, 138, 140–157, 163, 165, 167f., 171, 173-175, 219, 221–223, 226, 228–230 Lewis, D., 93, 102 Liard, L., 66 Lipps, T., 173 Locke, J., 12, 123, 166 Lotze, H., 176 Łukasiewicz, D., 199, 212f. Łukasiewicz, E., 211

Łukasiewicz, J., 55–57, 59, 61–68, 70– 74, 75–77, 84, 201, 222f., 226 Mach, E., 167 Mahler, G., 35 Mally, E., 16, 32 Maritain, J., 51 Martijn, M., 72, 77 Marty, A., 58, 69, 76, 83, 124, 128, 162, 169, 171–173, 175f. Meinong, A., 10, 13, 16, 18, 27, 29–32, 47, 51, 57f., 64, 70, 73–75, 85f., 89, 108, 110f., 118, 128, 174, 200f., 208f. Mellor, D. H., 174 Mill, J. S., 12, 66, 86, 89, 163, 186, 189, 222 Milne, P., 103 Moore, G. E., 128 Morris, C., 217 Mourelatos, A. P. D., 175 Mulligan, K., 57, 172 Münch, D., 170, 176 Niiniluoto, I., 75 Novak, M., 173 Ockham, William, 90, 102 Pfänder, A., 69, 76 Plato, 184 Poli, R., 29 Przełęcki, M., 212 Quine, W. v. O., 95, 167, 168, 176, 218, 226, 227 Rasiowa, H., 175 Reid, T., 15 Reinach, A., 38, 39, 44 Rojszczak, A., 74 Rollinger, R., 73 Russell, B., 16, 32, 36, 67, 70, 83, 85f., 95, 102, 124, 128, 187, 200f., 205, 208– 210, 212f., 218f. Savonarola, 59f., 62, 68, 70 Schaar, M., van der, 51, 73 Schnelle, T., 170 Schröder, E., 97 Schopenhauer, A., 152 Schuhmann, K., 33, 170 Schwyzer, E., 51

INDEX OF NAMES Sebestik, J., 73 Sellars, W., 174 Sextus Empiricus, 15 Sider, T., 102, 104 Siebel, M., 72, 77 Siegwart, C., 55 Siek, J., 72, 77 Sikorski, R., 175 Simons, P., 72, 76, 77, 102f., 115, 118, 170, 172–174 Sinisi, V., 174 Smith, B., 30, 102, 171–173, 175 Smolka, F., 73, 77 Sobociński, B., 177 Srzednicki, J., 175 Stumpf, C., 39, 75 Suarez, 175 Sundholm, G., 37, 72f., 77

237

Suppes, P., 103 Szaniawski, K., 172 Tarski, A., 55, 56, 57, 72, 77, 84, 101, 103, 163, 171, 217–232 Twardowski, K. 9–81, 83, 84, 102, 107, 121–124, 128, 133, 136, 141, 155, 160, 167, 171–176, 183, 185, 194–199, 200, 204f., 207–209, 212, 214, 223–225 Whitehead, A. N., 152, 168, 170 Wiegner, A., 73, 77 Wiśniewski, R., 199 Wittgenstein, L., 35, 95, 128, 218f., 221 Woleński, J., 72, 75, 76f., 121, 143, 170– 173, 177, 212, 219, 223, 225f. Wolniewicz, B., 172 Woodger, J. H., 174, 177 Zemach, E., 151–153, 168, 177 Zermelo, E., 83, 124, 221

PHENOMENOLOGY & MIND

Edited by Arkadiusz Chrudzimski • Wolfgang Huemer Vol. 1

Vol. 4

Arkadiusz Chrudzimski Wolfgang Huemer (Hrsg.)

Ralf Busse

Phenomenology and Analysis Essays on Central European Philosophy ISBN 3-937202-36-6 Paperback, 381 Seiten, EUR 32,00

Vol. 2 Wolfgang Huemer / Marc-Oliver Schuster

Writing the Austrian Tradition Relations between Philosophy and Literature

ISBN 3-937202-44-7 Paperback, 213 pages, EUR 29,00

Vol. 3

Wahrnehmung, Indexikalität und Reflexion.

Hector-Neri Castañedas Ontologie und Wahrnehmungstheorie und die Möglichkeit einer phänomenologischen Reflexion ISBN 3-937202-76-5 Hardcover, 447 Seiten, EUR 98,00

Vol. 5 Arkadiusz Chrudzimski (Ed.)

Existence, Culture, and Persons The Ontology of Roman Ingarden ISBN 3-937202-84-6 Hardcover, ca. 226 pp, EUR 78,00

Arkadiusz Chrudzimski

Intentionalität, Zeitbewusstsein Vol. 6 Anna Sierszulska und Intersubjektivität Meinong on Meaning and Truth Studien zur Phänomenologie von Brentano bis Ingarden ISBN 3-937202-63-3 Hardcover, 211 Seiten, EUR 78,00

ISBN 3-937202-94-3 Hardcover, 262 pp., EUR 78,00

Vol. 7 Alexander Batthyany, Avshalom Elitzur

Mind and its Place in the World Non-Reductionist Approaches to the Ontology of Consciousness ISBN 3-937202-98-6 Hardcover, 323 pp., EUR 94,00

ontos verlag

Frankfurt I Paris I Ebikon I Lancaster I New Brunswick

P.O. Box 15 41 63133 Heusenstamm bei Frankfurt www.ontosverlag.com [email protected]

ontos verlag