Meinong on Meaning and Truth: A Theory of Knowledge 9783110325669, 9783110325041

Table of contents :
CONTENTS
INTRODUCTION
PART IOBJECTIVES AND OTHER OBJECTSOF INTENTIONAL REFERENCE
CHAPTER 1REAL, IDEAL AND 'NON-SUBSISTING'OBJECTS
CHAPTER 2OBJECTIVES AS OBJECTS OF HIGHERORDER
PART IIMEANING AND TRUTH
CHAPTER 3MEINONG'S THEORY OF MEANING
CHAPTER 4MEINONG'S LOGICAL REALISM
PART IIIMEINONG'S TRUTH IN THE EYES OFHIS CRITICS
CHAPTER 5RUSSELL'S RECEPTION OF MEINONG'SPHILOSOPHY
CHAPTER 6AGAINST THE POST-RUSSELLIANVIEW OF MEINONG'S PROPOSITIONS
CHAPTER 7SOME EPISTEMOLOGICALINVESTIGATIONS CONCERNINGMEINONG'S PHILOSOPHY
PART IVMEINONG'S THEORY IN THEPERSPECTIVE OF PHILOSOPHICALSEMANTICS
CHAPTER 8REFERENCE IN A MEINONGIANSEMANTICS
CHAPTER 9THE SPECIFIC FEATURES OF AMEINONGIAN SEMANTICS
CONCLUSION: THE COSTS OFPRESERVING EPISTEMOLOGICALREALISM
BIBLIOGRAPHY
INDEX OF NAMES

Citation preview

Anna Sierszulska

Meinong on Meaning an Truth

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]



2005 ontos verlag P.O. Box 15 41, D-63133 Heusenstamm nr Frankfurt www.ontosverlag.com ISBN 3-937202-94-3

2005

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

CONTENTS INTRODUCTION ...................................................................................... 7 PART I OBJECTIVES AND OTHER OBJECTS OF INTENTIONAL REFERENCE.............................................................................................. 9 1. REAL, IDEAL AND 'NON-SUBSISTING' OBJECTS........................ 9 ♦ ♦ ♦ ♦

Meinong's reasons for introducing the theory of objects................ 10 Against Metaphysics ....................................................................... 13 Intentionality and anti-psychologism.............................................. 17 Aussersein ....................................................................................... 21

2. OBJECTIVES AS OBJECTS OF HIGHER ORDER .......................... 27 ♦ ♦ ♦ ♦ ♦

Relations and complexes: the principle of coincidence.................. 27 Real and ideal particulars ................................................................ 29 Relations and complexes as founded objects.................................. 36 Differences between objectives and other objects of higher order. 40 Objectives as abstract entities ......................................................... 44

PART II MEANING AND TRUTH ....................................................................... 49 3. MEINONG'S THEORY OF MEANING ............................................. 49 ♦ ♦ ♦ ♦

Meaning as 'sense' and meaning as 'object'..................................... 50 Meinong on Sentences and Objectives ........................................... 62 Objectives in the role of propositions ............................................. 70 The function-like nature of objectives ............................................ 73

4. MEINONG'S LOGICAL REALISM ................................................... 93 ♦ ♦ ♦ ♦

Truth as identity of logical structure............................................... 94 A realistic identity theory of truth................................................. 101 Truth and the modalities of objectives.......................................... 104 Self-evidence as a substitute criterion of truth.............................. 113

♦ Surmise-probability of objectives as a remedy for subjectivity of cognition........................................................................................ 119 ♦ Truth is absolute, mind-independent and eternal.......................... 123 PART III MEINONG'S TRUTH IN THE EYES OF HIS CRITICS................. 127 5. RUSSELL'S RECEPTION OF MEINONG'S PHILOSOPHY.......... 127 ♦ A difference of opinion concerning the nature of propositions.... 129 ♦ An objection to Meinong's 'psychologistic' treatment of false propositions ................................................................................... 132 ♦ Russell's remarks about truth in Meinong's theory....................... 134 ♦ The ultimate rejection of Meinong's ontological ideas: 'On Denoting' ....................................................................................... 136 APPENDIX TO CHAPTER 5: MEINONG'S DEFENCE AGAINST RUSSELL'S OBJECTIONS. 139 ♦ Meinong on the famous paradox................................................... 139 ♦ Meinong's notes in the 2nd edition of On Assumptions ................. 141 6. AGAINST THE POST-RUSSELLIAN VIEW OF MEINONG'S PROPOSITIONS .............................................................................. 147 ♦ Objectives as objects – Findlay's exposition of Meinong's philosophy ..................................................................................... 147 ♦ Objectives as states of affairs........................................................ 155 7. SOME EPISTEMOLOGICAL INVESTIGATIONS CONCERNING MEINONG'S PHILOSOPHY........................................................... 163 ♦ ♦ ♦ ♦

The notion of a true judgment....................................................... 163 Timeless truth................................................................................ 176 Objectives in Aussersein............................................................... 180 The Objection of Conception Dependence ................................... 184

PART IV MEINONG'S THEORY IN THE PERSPECTIVE OF PHILOSOPHICAL SEMANTICS ....................................................... 191 8. REFERENCE IN A MEINONGIAN SEMANTICS ......................... 191 ♦ Intensional reference and denotational reference ......................... 193 ♦ The meanings of sentences and states of affairs ........................... 200 ♦ The reference of sentences in a Meinongian semantics................ 205 9. THE SPECIFIC FEATURES OF A MEINONGIAN SEMANTICS. 215 ♦ Two notions of reference and two notions of extension............... 215 Reference in Meinongian logics .................................................. 216 A 'broad' notion of extension ....................................................... 226 ♦ Two kinds of properties and two kinds of predication ................. 232 ♦ Two senses of 'there is'.................................................................. 239 The principle of independence of so-being from being............... 239 Quantifying over nonexistent objects .......................................... 241 CONCLUSION: THE COSTS OF PRESERVING EPISTEMOLOGICAL REALISM......................................................................................... 249 BIBLIOGRAPHY.................................................................................. 255 INDEX OF NAMES .............................................................................. 261

Acknowledgments: The present study is a shortened version of my Ph.D. dissertation entitled "Meinong's Theory of Truth", which was submitted to the University of Geneva. I would like to express my gratitude to Kevin Mulligan and Jan Wole ski, the directors of the project. I would also like to thank Jacek Pa niczek and Peter Simons for their favourable reviews, Arkadiusz Chrudzimski for his editorial comments, all my Colleagues and the Audiences of my presentations related to this work. Fragments of this study have been published in Reports On Philosophy 22, Logic and Logical Philosophy 15 and in Contributions to the Austrian Ludwig Wittgenstein Society, Kirchberg 2004.

A.S.

MEINONG ON MEANING AND TRUTH

7

INTRODUCTION Most studies of Meinong's philosophy concentrate on ontological issues and they are often accompanied by providing a logical system of so called Meinongian logic. The epistemological problems are raised rarely and primarily in the light of historical considerations. The purpose of this study is to provide a presentation of the views of Alexius Meinong upon truth and related issues, in such a way as to expose the points which may be interesting for analytic philosophers. Part I contains an outline of Meinong's theory of objects and his account of intentionality. The subjective 'contents' of mental acts are contrasted with 'objects' of different kinds. Chapter 2 focuses upon objects of higher order and the notion of an objective. Meinong's notion of Aussersein is introduced and it is claimed that objectives are abstract entities belonging to Aussersein. Part II presents Meinong's theory of meaning and his views related to truth and cognition. The conception of meaning is discussed especially in relation to the views of Husserl and Frege upon this issue. Meinong's theory of truth is shown to be a version of logical realism, where identity of logical structure between an objective intended and reality is the basic idea but no facts as entities in reality are postulated. A Fregean interpretation of Meinong's theory of objectives as function-like entities, and not as states of affairs, is proposed. Factuality of objectives is interpreted non-objectually as the 'obtaining' of objectives. The notion of self-evidence of judgments is presented in the role of Meinong's substitute criterion of truth. The problem with subjectivity of the experience of selfevidence is solved in Meinong's conception by means of probability attribution in uncertain epistemic contexts. Part III contains a discussion of the reception of Meinong's ideas related to truth, since Russell until the present time. This part contains a chapter concerning Russell's interpretation of Meinong's objectives as complexes, in agreement with Russell's early theory of singular propositions. It is shown why this is not a correct interpretation. The issues addressed are the accusations of psychologism directed at Meinong and the mutual misunderstandings about ontological questions between these

8

ANNA SIERSZULSKA

philosophers. Meinong's reaction to these criticisms is presented, mainly as expressed in the 2nd edition of On Assumptions. Next, there is a polemical discussion with these critics who regard Meinong's objectives either as complexes or as states of affairs. And finally, some objections related to Meinong's understanding of truth and cognition are attended to. These objections concern mainly the traces of Kantian idealism in Meinong's epistemological views. In the case of empirical judgments, there is no certainty whether they are true in the objective sense, but such uncertainty in the process of cognition does not imply that we have no possibility to acquire objective knowledge. Scepticism is overcome, because we know that many of our judgments are highly probable. Part IV is devoted to an analysis of some typical features of Meinongian-style semantics. Chapters 8 and 9 present Meinong's original views by way of comparing his ideas to later developments within Meinongian semantic theories. It is observed that if a semantic domain is understood in the characteristic Meinongian way, it contains both real and meaning-objects of different kinds. This feature of a Meinongian-style semantics is responsible for what is proposed to be called a 'double theoretical approach' to objects. It is shown that two senses of being, of quantification, of predication, of extension and of linguistic reference are required, in order to provide a theoretical framework which applies both to real objects and to abstract sense-entities. The main questions discussed in this part are related to the consequences of introducing 'merely semantic' objects into a semantic theory. The study ends with a conclusion which sums up the results of the discussions with respect to their relevance for the issue of epistemological realism. Meinong's suggestion for developing a probabilistic semantics for undetermined contexts is considered to be a positive way to counterweight scepticism in scientific discourse.

MEINONG ON MEANING AND TRUTH

9

PART I OBJECTIVES AND OTHER OBJECTS OF INTENTIONAL REFERENCE CHAPTER 1 REAL, IDEAL AND 'NON-SUBSISTING' OBJECTS I think that it is quite plausible to claim that the theory of objects, as conceived by Alexius Meinong, is not only an inquiry in ontology but also an inquiry into semantic issues. If we consider this interpretation of his work, the purpose of admitting nonexistent objects and other controversial entities in the theory of objects immediately becomes understandable. In particular, the idea of Aussersein, the realm of all objects regardless of their existential status, appears to be remarkably more semantic than ontological, for nonexistent objects are devoid of ontological interest. Since semantic theories as such were in an early stage of creation at the time when the theory of objects was developed, a sudden introduction of ideas like those of Meinong was bound to be confusing. Meinong's theory of objects has an ontological character and it is entangled in discussions of diverse ontological questions. Nevertheless, it has a double purpose, balancing in-between semantic and ontological areas of interest. The violent reaction to the absurd ontological ideas contained in the theory of objects is partly a result of a misinterpretation, for Meinong's concerns reach out beyond ontological issues. His Aussersein is a domain of objects constructed as all possible arbitrary combinations of properties, and only some of these objects coincide with real objects. This is hardly a regular way to embark on the task of presenting a metaphysical or even ontological theory, yet this may be a way to provide a very broad theory of intentional and linguistic reference, which goes beyond what is ontologically acceptable. And again, it is not obvious even now how ontologically

10

ANNA SIERSZULSKA

restrained a semantic theory should be. There are theories which admit semantic domains that extend beyond what is supposed to have being, actually or even ideally or possibly, while others exclude any entities which would not belong to the accepted range of existing objects, depending on the ontology that is assumed. From the more restrictive point of view, Meinong's conception is a bad case of theoretical superfluity. From the point of view of the semantic approach, which does not identify the sphere of semantic entities with that of ontological entities, Meinong's controversial ideas are fully acceptable if taken to belong to a theory of a semantic character. The approach according to which semantics need not be restrained by ontological concerns has been employed in logics of intentional objects, some theories of mathematical objects, possible worlds semantics and in some developments within the semantics of natural language. Meinong's conception of a general theory of all intentional objects, regardless of their existence, anticipates this need for an extraontological approach in semantics. The relation between metaphysics, ontology and Meinong's theory of objects can be represented as follows: ( THEORY OF OBJECTS (ONTOLOGY (METAPHYSICS))) Metaphysics is the narrowest discipline of the three, it is concerned only with what there is. Definitely broader than metaphysics is ontology, which deals not only with what there is but also with whatever could be. In comparison to these two traditional philosophical disciplines. Meinong's theory of objects has by far the most general and allinclusive character, as it occupies itself with what there is, what could be and whatever we can think or speak about. ♦ Meinong's reasons for introducing the theory of objects When we look at the theoretical concerns Meinong presents as reasons for founding a new discipline which he calls the theory of objects, in the first place we observe that the main task of this theory is to account for both linguistic and mental reference to entities that cannot be said to possess any

MEINONG ON MEANING AND TRUTH

11

actual existence. The sciences which he has particularly in mind while speaking about nonexistent objects are mathematics, logic and linguistics. In Meinong's time, mathematical objects were still mostly considered to belong to the realm of ideal entities, they were not seen, as yet, in the constructivist fashion. But even the question of reference to such ideal objects poses a certain problem from the purely extensional and existential point of view. Meinong thinks that it is necessary to give special attention to this phenomenon. (...) pure mathematical knowledge is never concerned with anything which must, in the nature of the case, be actual. The form of being (Sein) with which mathematics as such is occupied is never existence (Existenz). In this respect, mathematics never transcends subsistence (Bestand): a straight line has no more existence than a right angle; a regular polygon no more than a circle. It can be regarded only as a peculiarity of the mathematical use of language that this usage makes quite explicit existence-claims.1

In the above, Meinong remarks that mathematical language makes constant references to non-existent entities and that they are spoken about in a manner which assumes their existence. He maintains that this situation is due to the lack of recognition of the mathematical subject matter as properly belonging to the theory of objects. Therefore, the existential claims in mathematical language could only be understood figuratively. Because, in fact, the whole discourse of mathematics lies within the field of investigations of the theory of objects and, accordingly, it lies beyond any existential concerns. I have referred before to the fact that a suitable place for mathematics could never be found in the system of sciences. If I am not mistaken, the anomalous position of mathematics had its basis in the fact that the concept of a theory of objects had not yet been formed. Mathematics is in its essential features, a part of the theory of objects.2

However, mathematics as such, could only be seen as a specialized branch of the theory of objects, concentrating exclusively on its own particular subject matter. It does not approach the general theory of objects as closely 1 2

A. Meinong, "The Theory of Objects", p. 80. A. Meinong, "The Theory of Objects", p. 98.

12

ANNA SIERSZULSKA

as some of the related investigations which Meinong mentions: the theory of functions, of extension, of manifolds and meta-mathematics. He also speaks about mathematical logic in this context. All of these disciplines seem to have some bearing on what he attempts to capture in his theoretical work. What his work actually shares with these other areas of investigation is related to his interest in various abstract objects of human mental and linguistic activity. What I have called the encroachment of the mathematical approach beyond its strictest limits has an instinctive and unconscious character in comparison with the completely explicit attempts to expand that domain and to generalize to the fullest extent possible that way of framing a problem. These have probably already achieved some importance under the name of the general theory of functions; one cannot fail to see this in such designations as 'the theory of extension' and 'the theory of manifolds,' and even under the frequently misunderstood catchword 'meta-mathematics.' From the point of view we have adopted here, these strikingly significant investigations represent the transition from the specialized to the general theory of objects. A similar status can be ascribed to the endeavors and results customarily grouped under the general name of 'mathematical logic,' even though these endeavors are in many respects intended for an entirely different purpose.3

Not being himself a logician, Meinong apparently notices a close connection between his theoretical aims and some research in the field of logic, starting already from the level of its most basic principles. For example, in the very idea of inference he sees an expression of the basic claim of his theory of objects, which is to separate theoretical work from concrete questions about the status of the objects under investigation, as being there or being true about the world. ...the doctrine of inference is customarily worked out in a way that is in the first instance formalistic, in that practically nothing appears of any attention to the concrete experiences of the person who infers. One's primary interest is merely in what the premises and conclusion stand for objectively – which is incidentally very indicative as to the way in which the tenor of the theory of objects naturally predominates in the doctrine of inference...4

3 4

A. Meinong, "The Theory of Objects", p. 102. A. Meinong, On Assumptions, p. 137.

MEINONG ON MEANING AND TRUTH

13

Meinong's occupation with logic is, nevertheless, marginal. The results of Meinong's work belong to logic only to the extent to which theories of intentional reference and linguistic meaning can be regarded as related to logic. All that can be justly claimed is, perhaps, that he has made a lot of suggestions of great interest to logicians. Another discipline which is mentioned as remaining in close connection with the theory of objects is linguistics. Meinong discovers its importance when he realizes that the phenomena of intentionality and mental reference are quite accurately reflected in people's linguistic ways of expression. Contrary to the approach to which we are now accustomed, his initial point of departure is the mental sphere, and only then does he consider linguistic behaviour as a reflection of that sphere. But once that discovery is accomplished, linguistic and semantic concerns acquire great significance for him. The same applies to linguistic science, in particular to grammar, whose significance, in fact, has not been fully surveyed either by the old or by the new logic; indeed its significance could scarcely be appreciated until the distinction between Object and Objective had been recognized in the distinction between the meaning of words and the meaning of sentences.5

Meinong is aware that his theory of objects has to be occupied with the questions of linguistic meaning and that it is one of its functions to provide a theory of meaning for the expressions and sentences of language. ♦ Against Metaphysics Meinong emphasises that his investigations in the theory of objects are meant to extend far beyond what is understood as metaphysics. He does not wish to confine his theory to objects which actually possess being. But the generality of the new discipline is only one reason. What is more important is that the point of view of the theory of objects is entirely reversed in comparison to metaphysics, because objects are seen from the angle of how they appear and how they are referred to in consciousness. In other words, Meinong's objects are primarily objects of knowledge and 5

A. Meinong, "On Objects of Higher Order and Their Relationship to Internal Perception", p. 103.

14

ANNA SIERSZULSKA

objects of linguistic reference. The issue of whether they possess any metaphysically legitimate status seems to be of secondary concern. Without doubt, metaphysics has to do with everything that exists. However, the totality of what exists, including what has existed and will exist, is infinitely small in comparison with the totality of the objects of knowledge.6

Such a reversed approach in the way objects are theoretically perceived suggests a possibly idealistic outcome, as it is the case, for instance, with the philosophy of late Husserl. But Meinong's conception does not drift in this direction at all. Objects of intentional or semantic reference do not have to be real ordinary objects, on his view, but many of them are very obviously real. The real ones are ascribed existence. Those which are ideal, like most of mathematical objects, are merely subsisting objects. And there is also a whole sphere of non-ideal, yet mind-independent objects, which can be best described as non-subsisting. If metaphysics is a general science of the real, should we say that the theory of Objects is, in contrast, the general science of the non-real? This would obviously be too narrow. Why should real objects be excluded from the theory of Objects as such? Or would it be more appropriate to describe the theory of Objects as the theory of the subsistent, contrasting the words 'subsist' and 'exist' in such a way that, whereas all existing things subsist, it is not true that all subsisting entities (e.g. difference) also exist? Even in this case, the area which the theory of objects comprehends, as we have seen, would not be included in all its entirety; the nonsubsistent, the absurd, would be excluded. To be sure, the non-subsistent is of little concern to the natural interest, and it provides an even smaller point of purchase to intellectual understanding. But it does belong to the 'given' (Gegebenen), after all, so that the theory of Objects can by no means ignore it.7

Among non-subsisting objects are included not only so called impossible objects (which are self-contradictory), or fictional objects of mythology and of literature. The numerous theoretical entities of science belong here as well. The latter are often related to a posteriori knowledge and to highly empirical scientific disciplines, but not being in themselves really existing entities, they can have little claim to be the proper subject of metaphysical investigations. The theory of objects is supposed to direct its attention to 6 7

A. Meinong, "The Theory of Objects", p. 79. A. Meinong, "The Theory of Objects", p. 108.

MEINONG ON MEANING AND TRUTH

15

all these non-real objects, which can be meant and spoken about, in spite of fitting nowhere in the metaphysical frame. There are, therefore, precisely two sciences of highest generality: an a priori science which concerns everything which is given, and an a posteriori one which includes in its investigations everything which can be considered by empirical knowledge, i.e. reality in general. The latter science is metaphysics, the former is the theory of Objects.8

Metaphysics is not suitable to fulfil the tasks assigned by Meinong to the theory of objects, because metaphysics does not, and cannot, concern itself with all objects of knowledge indiscriminately. Meinong remarks that his theory of objects might perhaps instead be called 'metaphysics independent of existence'. (We would presently describe ontology in this way, since indeed Meinong's investigations partly belong to ontology and partly trespass it.) However, Meinong decides that this terminology would not remain in agreement with the tradition and it would probably cause even more confusion. The use of rational tools in metaphysics that appeared in the scholastic philosophy actually brought the result that metaphysics has become an a priori discipline to a large extent, but not in the way which Meinong has in mind. For all these rational devices are still applied with the purpose of investigating reality, if not necessarily perceptible reality, and not with the intention to provide a general theory of knowable entities. Thus, metaphysics is bound to what there is. In this sense, metaphysics is the science of reality, as Meinong points out, while the theory of objects is characterized by an a priori approach to the objects of knowledge. In the additional notes to the second edition of On Assumptions, we can find the following note: In Justification of the Theory of Objects: 1. There is a need for a science of the totality of objects. One might regard metaphysics as such a science. But in that case, metaphysics would at least have to treat existing and nonexisting things on the same footing; i.e. it would presumably have to existence as such out of consideration, which would certainly run counter to the intentions of metaphysics. One might therefore divide metaphysics into two parts, existential metaphysics [Daseinsmetaphysik] and metaphysics independent of existence [daseinsfrei 8

A. Meinong, "The Theory of Objects", p. 109.

16

ANNA SIERSZULSKA Metaphysik]. But in that case, all that it amounts to in regard to the theory of objects is a redesignation; adjustment would have been made to the notion of the theory of objects. 2. There is a need for a science that puts objects in the characteristic perspective of an a priori approach, a science whose main emphasis is on this approach, as opposed to the empirical attainment of knowledge, and which accommodates itself to the needs of such a distinctive approach – not only occasionally and according to incidental needs, as the empirical sciences have always done. 3. On the tradition. Scholastics speak of an 'ens latissimum,' ens rationis being comprehended in that. But it is asserted that there can be no science of it. This is surprising, since it is precisely scholasticism that separates essentia and existentia and prefers rational methods of investigation. But it doubtless does this only with the instinctive presupposition that what is really worth knowing is to be found in reality [Wirklichkeit]; it is at most as means to an end that other things are to be given attention. But then, this is precisely the situation is which the theory of objects is something new. – After that, one might be able to estimate how the theory of objects stands in regard to ontology. Here it is surely questionable as to whether the latter is supposed to be more of a theory of being [Sein], or a theory of entities [Seienden]. Each by itself would be too little from the standpoint of the theory of objects. Main thing, though: it has always been understood as a part of metaphysics. If the latter cannot be divested of the notion of reality without unnaturalness, then this tendency attaches also to the part of it that is taken separately as ontology.9

Ontology, though it is regarded as a theory of entities, as Meinong admits, does not seem to be able to accommodate all his theoretical concerns, either. The interests of ontology are also more on the side of reality than of the problems of meaning and intentional reference. Therefore, the objects of mental reference as such are well beyond its scope. Since there is no way, according to Meinong, to accommodate all the fields of interest of the theory of objects within the frame of metaphysics, ontology, or any other existing discipline, a new discipline must be created, which will be more general than any of the particular ones, or rather more basic from the epistemological perspective.

9

A. Meinong, On Assumptions, Meinong's additional notes to the 2nd edition, p. 275.

MEINONG ON MEANING AND TRUTH

17

From what has been said, I draw the conclusion that the theory of Objects has a claim to the status of a discipline independent even of the theory of knowledge, and, accordingly, to that of an independent science.10

In fact, what turns out to constitute the theory of objects is a mixture of what one would describe as ontology, theory of cognition and a semantic theory taken together. But this does not indicate that Meinong's research has been of no ultimate value to any of these branches of knowledge specifically, his work contributes significantly to all of these philosophical disciplines. ♦ Intentionality and anti-psychologism The main idea standing at the origins of the theory of objects is that of the intentional directedness of all mental acts. Since all mental acts are directed to something, it is necessary to devote some theoretical attention to these entities to which our mental acts are directed. And, although the subject matter belongs mainly to the psychology of cognition, it is not without bearing upon more general philosophical questions, or so Meinong seems to think. From problems that remain to some extent a precinct of formal logic, let us now pass on to the evaluation of a fact that all mental activity shares in, as something fundamental to itself – even though epistemology can in a certain sense claim a prior right to deal with this fact. There are hardly any mental events without objects, and there is certainly no representation without an object.11

At first, Meinong believes that mental acts are simply directed at mental contents, constituting certain ideas and representations. Naturally, these could be representations of external objects, but these further objects are not included in his considerations about intentionality. After Twardowski's famous work that differentiates the contents from the objects of representations, Meinong's position on this issue also alters. But while Twardowski speaks about external objects only, Meinong introduces a

10 11

A. Meinong, "The Theory of Objects", p. 98. A. Meinong, On Assumptions, p. 159.

18

ANNA SIERSZULSKA

three-level distinction, between the content of the act as such, the immediate object presented by the content and the transcendent object. In the same way it will be conceded that there is no having of ideas and no judging without contents. But the willingness with which to make this concession will, for many people, be based on the assumption that content and object are almost the same thing. Even I myself believed for a long time that the expressions could be used interchangeably, that is, that really one or the other of them could be dispensed with.12

In his later writings, contents are disregarded completely, as idiosyncratic psychic phenomena and, instead, he occupies himself with their nearest objects, which are supposed to reflect what is objective and repeatable in what the content presents. Nothing is more ordinary than to have an idea of something or judge something which does not exist. The non-existence of the something can have various reasons: there may be a contradiction as with the round square, there may be just a factual inexistence as with the case of the golden mountain. It can be a something which by its very nature cannot exist because it is not real: the equality between 3 and 3, the difference between red and green can subsist – as linguistic use already permits – but it cannot exist as e.g. a house or a tree does. Finally, something can be real but did exist or will exist in the future, but does not exist at present. The idea, thus, exists. Is there anyone who would assume, except for the sake of theoretical prejudice, that the idea exists but not its content? In connection with our differentiation we may think of the frequently used contraposition of the 'immanent' objectum and the 'transcendent' objectum, that is, of something of which merely an idea is had and reality. We may think that the principle that everything psychic, that is, primarily each idea, must have an object does not concern the transcendent, but only the immanent objectum, and that, for the latter, the just indicated difficulties do not hold.13

Meinong argues that what the content of the mental act frequently presents is not an existing real object. For this reason, the objects of mental acts cannot be identified unreflectively with transcendent objects of intention. The most appropriate candidates for objects of mental acts intending something non-existent, are the immanent objects presented by the very contents of these acts. Because in the case of non-existent objects, only the 12 13

A. Meinong, "On Objects of Higher Order", p. 141. A. Meinong, "On Objects of Higher Order", p. 141-2.

MEINONG ON MEANING AND TRUTH

19

idea actually exists, or rather its content, but the object does not exist. However, this does not mean that such mental acts have no objects at all, for they still have 'immanent' objects. Such objects obviously do not exist in the ordinary sense, at best they pseudo-exist, as much as the contents that present them are really existing psychic events. Yet the term 'immanent' object stands in absolute opposition to what Meinong has in mind speaking about objects of intentional reference. The immediate or nearest object of the act is never merely immanent. It is an abstract object, which can be identified with the meaning of an expression or a sentence that could be used to express the content of a given act. Obviously, abstract meanings can have no existence, but one would hesitate to say that they are only immanent. If we want to characterize precisely the state of affairs as given in the above cases we must say as follows: the ideas in question naturally do not have a transcendent objectum. There does not even exist for them an immanent objectum. Thus, the question may be raised if we even speak of existence at all in connection with immanent objecta. What, in fact, exists is the idea, in our case (or the judgment, not to mention feelings and desires). Of course, the content is included (in that existence). At first, it may seem quite surprising that they 'have' an (immanent) objectum. But at closer look, we find that this reveals exactly the nature of the concept of objects (Gegenstandsgedanke).14

Meinong is especially careful to make it clear that his use of the term 'immanent' has only an explanatory function, while he does not intend by it all the psychologistic burden that it carries. He states plainly that he neither assumes that the immanent object can in fact possess any being (apart from 'implexive' being, when there is an external object it may be identified with), nor that its character is merely psychic on such occasions. As a result, the immediate object of an intentional act either can be identified with an object which actually has some kind of being (exists or just subsists), or it is just a pure object belonging to the neutral sphere beyond being (Aussersein). Ideas of different objects may be congruent concerning the act, but they differ in something else which can be called 'content of ideas.' The content exists, is real and present, it is also psychic, naturally, even if the object of which an idea is had 14

A. Meinong, "On Objects of Higher Order", p. 142.

20

ANNA SIERSZULSKA by means of the content, does not exist, is not real, not present and not psychic. When we attempt, in a particular case, to neatly separate content and corresponding object we notice right away that the content moves into the background overpowered by the object. At least, in bypassing, we will mention that this has internal and, at any rate, external reasons which are apparent in the linguistic usage. In virtue of the fact that language is 'expression' it reveals the ideas of the speaking person. But it not only reveals the having of ideas in general but also its content-determinations. But what the speaking person wants to 'say', or more precisely, that about which he wants to speak is not that what the words express but is that what they denote (bedeuten). (...)There is an utter lack of natural expression for contents. Thus, when we need to communicate about a certain content there is nothing left to us but to use the corresponding object. In this way, we can speak, for example, about the 'content of the idea blue' and say that it is different from 'the content of the idea red.'15

The fact that these immediate objects are the proper objects of mental acts is reflected in language, as Meinong points out in the fragment quoted above. What the person wants to say is not what the words express (i.e. the contents which they express), but what they mean: what they are about. (The translation of 'bedeuten' by 'denote' is incorrect in my terminology from Ch.8, since Meinong's nonexistent objects cannot be regarded as denotations, even though they can belong to the extensions of predicates.) These objects that can be called the meanings of linguistic expressions are not as evasive as the contents by means of which they are presented. We are able to concentrate on them and recall them in our thoughts at will. They differ from corresponding external objects, if such there are, in that they do not include all characteristics of complete existent objects. They can be often 'implected' in several more complete, or fully complete and existent, objects. Then, all the properties of the incomplete object as presented in the mental act are included in the properties of the ultimate object of reference. In this way, one incomplete object can be used to refer to several objects, and also it is possible to intend one and the same object by means of several incomplete ones. Meinong does not call these incomplete objects 'abstract contents' or 'abstract senses', so as to avoid such an interpretation of them that would deprive them of their objective status. 15

A. Meinong, "On Objects of Higher Order", p. 143.

MEINONG ON MEANING AND TRUTH

21

Naturally, this does not at all hinder the fact that black and quadrangular are still two different objects; indeed, even 'something black' and 'something quadrangular', both taken 'closed' are different. All the same, the objective of sobeing 'something black is quadrangular' (where 'something black' must be intended 'open'), is factual; and so, under favourable circumstances, one and the same object certainly can be intended with the two contents. And it surely would not be natural to call only the blackboard an 'object' and, as against that, 'something black' and 'something quadrangular' contents. Nevertheless, this has happened (e.g., in Husserl's oft-cited example of 'the victor of Jena' and 'the vanquished of Waterloo');16

The content of a mental act is psychic and it is specific for a particular act: the content of a representation for this representation, of a feeling for this feeling, of a judgment for this judgment and of an assumption for this assumption. But what can be common to many acts are the objects of all these acts, understood as the abstract senses which are grasped in them, like in the case of 'something black'. Likewise, 'the victor of Jena' and 'the vanquished of Waterloo' are abstract senses, and therefore, objects, not contents, in Meinong's terminology. This is a solution adopted to avoid any psychologistic interpretation of objects of cognition and linguistic reference. Meinong's distinction between the content of the act and its nearest and remote object:

Act

(content)

object presented by the content

(content is merely psychic)

implected in

[transcendent object] [there may be no transcendent object]

♦ Aussersein On Meinong's theory, all objects, regardless of their existence or its lack, are non-subjective, mind-independent entities. Whenever immanent objects are discussed, they are taken to be subjective and concrete presentations of 16

A. Meinong, On Assumptions, p. 200.

22

ANNA SIERSZULSKA

mind-independent entities. The principle of independence of an object's nature (i.e. of its possession of properties) from its being, expresses the general approach of the theory of objects, according to which even selfcontradictory objects, like a round square, find their place among genuine objects of intentional reference. The idea of Aussersein, which is a spectrum of all imaginable objects of intentional and linguistic reference, is probably the most original in Meinong's conception. Basically, it is an idea of a semantic character. The intentional activity of the mind does not discriminate between existent and non-existent objects, and neither do any of the natural languages. One of the goals of the theory of objects is to account for this situation. As soon as it is recognized that, apart from special cases, both being and nonbeing are equally external to an Object, it is then understandable that nothing more, so to speak, is involved in comprehending the non-being of the Object than there is in comprehending its being. The above-mentioned principle of the independence of Sosein from Sein now presents a welcome supplement to this view. It tells us that that which is not in any way external to the Object, but constitutes its proper essence, subsists in its Sosein – the Sosein attaching to the Object whether the Object has being or not.17

From the perspective of an intentional act, being and non-being are external to an object. What is primarily represented in the act is the object's so-being, which makes it possible to make an intentional reference at all. Obviously, the notion of an object is in this case much broader and also 'weaker' than the customary one. The objects which only belong to Aussersein, and possess no being otherwise, would never be called objects according to the traditional approach. But they are not merely subjective contents of thought, because they can be intended, referred to and spoken about by many different subjects. A weaker notion of an object is, therefore, a price to be paid for rendering the intersubjective character of such objects. In all this, there is more than a mere 'weakening' of the initial definition of objectivity. For if objectivity consists in the 'having' of an object, and the 'had'

17

A. Meinong, "On Objects of Higher Order", p. 86.

MEINONG ON MEANING AND TRUTH

23

object must be one that has being, than a representation simply does not have an object in any of the instances of fiction... 18

That objects of intentional reference are not psychic objects is demonstrated by the way language is used, and even by the possibility itself to communicate with other users of a language about objects which actually do not have any being. Thus Meinong's non-subsisting objects from Aussersein can play the role of meanings in linguistic communication. What is specific for Meinong's theory of meaning is that there is no sharp borderline between the abstract object as the meaning of an expression and the object ultimately intended by means of this expression, which may well be a really existing object. The meaning-object which appears in the mind when the subject hears or pronounces a word, may be identified easily with an existent object. There is no separate sphere of immanent objects extending against another sphere of the objects of reality behind a veil. The veil of the idealistic viewpoint is removed. Meinong's immediate objects are only means to the end of grasping real things, whenever that is possible. For this reason, when he writes that a word expresses a content, but it means an object, he does not mention two objects each time, one in the mind and one outside of the mind. Of course, the notion of an object is used in the weaker sense in such contexts, to include both real and meaning-objects. ...in intellectual operations, what a word 'signifies' stands out in sharp contrast to what it 'expresses.' What it signifies is the object of the intellectual event in question, in the first instance the object of the representation that makes up this event or at least forms its basis.19

The object which is meant by an expression is the object of the representation related to the content the expression conveys, but the content of the representation itself is not the proper meaning of the expression. The content of the representation may differ from one act to another, even when what we would describe as the objective sense of the expression does not change. This is why contents cannot serve as meanings 18 19

A. Meinong, On Assumptions, p. 170. A. Meinong, On Assumptions, p. 24.

24

ANNA SIERSZULSKA

and it is necessary to introduce auxiliary objects into the theory, which are really the abstract counterparts of contents. (...) the auxiliary function should really be ascribed not to the object, but to the content to which this object stands as its nearest object. But the content is (...) hardly available to controlled reflection, while its nearest object is available easily and, therefore, it is advisable, as a kind of fiction, that we ascribe to the nearest object what in fact is the matter of content, so that we can speak about an auxiliary object in relation to which the intended object stands as the ultimate object.20

The auxiliary function consists in the mediation of the nearest object of the intentional act (which is an incomplete object) in grasping the ultimate object of the act's intention. The ultimate object may exist, subsist or be a non-subsisting entity. The auxiliary meaning object is, at least partially, identified with the ultimate object as to its properties, but it is in itself always just an object 'beyond being' that cannot possess existence.21 Meinong divides all objects into real existing objects, ideal subsisting objects and non-subsisting objects that can only be intended but they possess no being. All meaning-objects are either ideal or non-subsisting. These include entities that can appear as immediate objects of intentional acts, and exclude real existing objects (which are impossible to grasp due to their completeness), as well as relations and complexes (which are not apprehended directly, but only through objectives). Meaning-objects include: - ideal objects – the proper abstract objects (mathematical objects, figures of geometry) – although it is in order to observe that all meaning-objects are abstract entities - objectives: factual objectives (which are ideal and subsisting) and nonfactual objectives (non-subsisting) - auxiliary meaning-objects (non-subsisting, but may possess implexive being) - non-subsisting objects – also called nonexistent objects outside Meinong's theory (both inconsistent and consistent) 20

A. Meinong, Über Möglichkeit und Warscheinlichkeit, pp. 195-6. But Meinong speaks about 'implexive existence' of such auxiliary objects which serve to apprehend real objects. 21

MEINONG ON MEANING AND TRUTH

25

In the Diagram below, 'meaning-objects' are called 'semantic objects', according to the terminology introduced in Ch. 8 (which is not Meinong's terminology), and there is also a new category of 'merely semantic objects' which includes non-subsisting objects and non-factual objectives. Meaning-objects of all kinds are precisely at the centre of interest of the theory of objects, because our intellectual activity and use of language are based upon the phenomenon of intentional reference. The idea of auxiliary objects explains how reference to real objects is possible and it also provides a background for epistemological investigations. We saw that significations are objects. The interest in the former thus leads to the investigation of the latter, and in this way we are led to a science whose special character and justification as a 'theory of objects' I have explained elsewhere.22

Meinong is aware that whoever is interested in the process of cognition can find the most relevant material in linguistic usage and the investigation of issues related to meaning. But understanding meaning merely as denotation leads to disregarding a whole sphere of cognitively significant matters in favour of a simplified picture. Therefore, the theory of objects has to be engaged in the study of objects, but not just objects as they are in reality – not only the external denotations of linguistic expressions. Rather, its task is to concentrate upon the examination of objects as they are conceived by the cognizing mind. And all objects of intentional reference, despite their non-existence or non-subsistence, have to be the subject matter of these studies, just as they are exposed in Meinong's Aussersein.

22

A. Meinong, On Assumptions, p. 37.

26 Diagram 1

ANNA SIERSZULSKA

MEINONG ON MEANING AND TRUTH

27

CHAPTER 2 OBJECTIVES AS OBJECTS OF HIGHER ORDER Objects of intention in Meinong's theory are divided into objecta and objectives. Objecta are meanings of single expressions and they are objects of mental representations. Objectives are meanings of sentences and they are objects of intention for the acts of judgment or assumption. There is another classification into simple objecta and objects of higher order. Among the latter we find objecta of higher order – relations and complexes – and objectives. It is a specific feature of Meinong's theory that objects of higher order do not have to be existentially dependent on other objects, but they may have their foundation in a priori necessary relations between objects, or they may only presuppose lower order objects, in the sense that the inferiora are logically prior to their superius. As a result, some objects of higher order do not depend as to their subsistence upon the existential status of their inferiora, which is the case with ideal objects of higher order and, above all, with objectives, which are only logically 'dependent', but not existentially, and they are not ideally founded upon their inferiora, either. This stands in contrast to the traditional view, on which a dependent object, a relation for instance, can subsist only as far as the objects which are related either exist or subsist. In general, objects of higher order cannot possess existence, and if they are not ideal objects (like the relations between some properties of figures in geometry, or like true objectives), then they posses no being at all, apart from belonging to Aussersein. ♦ Relations and complexes: the principle of coincidence All objects of higher order are intrinsically dependent on other objects, in such a way that they cannot be conceived without presupposing other objects. It is important to observe that not being conceivable without presupposing other objects is a more general characteristic of all objects of higher order than being incapable of subsistence without the subsistence or

28

ANNA SIERSZULSKA

existence of the subordinate objects, because some objects of higher order, and especially objectives, subsist or do not subsist regardless of the existential status of their subordinate objects. There are also some dependent objects which are not objects of higher order, but which involve some other ideas as belonging to their nature. Meinong gives the example of the idea of colour, which never appears in the mind without the idea of extension. Yet it seems that, in a way, it is something external to the idea 'blue' that it must be extended in space and this is not what is meant by an object of higher order. Meinong's favourite example of a paradigmatic object of higher order is 'difference.' The relation of 'being different' makes no sense without the idea of objects that would be related by this relation. Therefore, we may say that an object of higher order is unfinished, or unsaturated, without the objects it presupposes. As is well known, there are objects which by their very nature reveal an intrinsic dependence. I do not mean a dependence as far as occurring at all is concerned, as we cannot have an idea of color without extension. Even this dependence may be founded in the nature of color and extension. But it can still be called external in comparison with the – I might say – unfinishedness, which is e.g. in the object 'difference' when it is isolated from that which is different. (...) The intrinsic dependence with which we will be exclusively concerned in the following, can also be described in this way: we are dealing, there, with objects which are built on other objects as their necessary presuppositions (Voraussetzung). Looking from this angle, we are justified in calling objects which are intrinsically dependent on those objects which are their presuppositions, in the sense just indicated, objects of higher order. In order to complete the requisite terminological conventions we must add that objects on which such objects of higher order seem to be resting are called inferiora. But an object which is built on another object is called the superior of the latter object.23

Meinong distinguishes two kinds of dependent objecta: relations and complexes. Relations require objects that are supposed to be related. These objects belong to the relation as its members. Complexes, on the other hand, are composed of their inferiora as the constitutive parts of the complex. This characteristic of objecta of higher order, that their inferiora belong to them, does not apply to objectives, which are only 'logically' dependent on other objects. 23

A. Meinong, "On Objects of Higher Order", p. 144.

MEINONG ON MEANING AND TRUTH

29

In a few words: objects of higher order are not only relations but also complexions. The inferiora of the latter are its constitutive parts as in an analogous manner the inferiora of a relation are its members.24

We also find out that relations and complexes are not entirely independent of each other, but, to the contrary, they are tightly bound together, so that it is impossible to think about a complex without the relevant relation. The reason is that a relation, together with its members, constitutes a complex. A relation is, therefore, one of the components of the corresponding complex. Although it is not correct to identify a relation with the complex, complexes coincide with relations due to their interdependency. ...if there is a relation between a and b then, ipso facto, relation members are given as component parts of a complexion. If we are about to have ideas of a and b as standing in the relation r we can only do it by having an idea of them in a complexion. The principle 'where there is a complexion there is a relation and vice versa' is a natural result holding for real things and those which are merely thought. (...) The relation is rather a part of the complexion. The complexion comprises, beside the relation, the so called component parts which constitute it and which are precisely the relation members. The relation is as dependent upon them as the complexion is dependent upon the relation. I call this relationship of partial identity and mutual dependency 'partial coincidence.'25

Obviously, so called real relations coincide with real complexes, while ideal relations with ideal complexes. But the meanings of 'real' and 'ideal' in this case are slightly different from the standard meanings of these notions. ♦ Real and ideal particulars Meinong begins his investigations from the perspective of how objects appear in consciousness, and in particular, how they are perceived. It turns out that even physical objects are subject to internal perception, when it comes to making comparisons between them, because consciousness has immediate access to these objects only as auxiliary objects which are 24 25

A. Meinong, "On Objects of Higher Order", p. 145. A. Meinong, "On Objects of Higher Order", p. 146.

30

ANNA SIERSZULSKA

presented in mental acts. It has no direct access to ordinary physical objects. It is recommended to consider a case where that of which the idea is head or which is cognized, that is, the object, is something physical, for example, a church tower, a mountain peak or something of the kind. Then, we must ask, at first, what in the world internal perception can have to do with such external objects. And again, we are thrown back on the already mentioned contrast between transcendent and immanent objecta. It is not the real church tower of whose existence internal perception tells us, but merely 'the church tower as in our ideas' (der vorgestellte Kirchturm).26

It is quite natural to assume that what is physical cannot be literally present in the mind, and even the most adequate representations of transcendent objects must be incomplete. This does not amount to saying that we always have incorrect representations, for they may be correct in what they are able to represent. We would never be able to represent physical objects if we were incapable of external perception, which is responsible for the fact that an object can be presented in a mental act in the first place. External perception also justifies our belief that it is possible for the immediate object presented in the act to be an adequate representation of the external one. If the perceived object in question shows itself to be something totally or partially physical, not an immanent objectum, but a physical reality, then we are dealing not with internal but with external perception (...)27

So, it is not denied that physical objects are subject to external perception, but when we compare them or make other mental operations involving these objects, we deal with these objects as they are represented in the mind, not with the physical objects. Thus, speaking about objects of higher order, we may leave out the problems concerning external perception and concentrate on internal perception entirely. Our demand that perception and the perceived must be simultaneous has not been made completely precise. The star, at the time that we perceive it, may have ceased to shine long before our perception of it. But the inaccuracies which come 26 27

A. Meinong, "On Objects of Higher Order", p. 156. A. Meinong, "On Objects of Higher Order", p. 160.

MEINONG ON MEANING AND TRUTH

31

to the surface here specifically concern external perception. We can leave them out of consideration when we are dealing essentially with internal perception.28

Such an approach will be legitimate in the case of all objects of higher order, with the single exception of some objectives. Objectives of 'sobeing' will still involve internal perception, but objectives of being have a different character, they apprehend a moment of factuality and, consequently, they depend upon purely external perception. The proper objects of internal perception are, of course, 'ideas', since they really exist in the mind at the time of the occurrence of the mental act. No wonder that relations between ideas will end up having more reality for Meinong than any relations between external objects. But the crucial point is that we must always be primarily talking about relations between objects of internal perception when dealing with comparisons of similarities and differences between objects, as well as with other relations between them. The perception of a transcendent object is external, and so is the perception of such inferiora of objects of higher order which possess existence as physical objects, but objects of higher order themselves belong almost exclusively to internal perception. Surprisingly, this dominant role of internal perception does not lead Meinong any further in the idealistic direction. Such is the starting point, from which Meinong arrives gradually at a fairly realistic epistemological view, and leaves behind him the view adopted after Locke and Hume: that objects are just bundles of 'ideas' and that all relations are created by mental activity. Accordingly, such qualities as 'green' and 'rectangular', understood as abstract 'ideas', turn out to be insufficient to identify a particular table. An object is not conceived as a bundle of simple ideas, because, essentially, it is not merely a collection of ideas. The qualities must be particularized as belonging to this and not some other object. They are understood as moments of a certain substantial bearer. Everyone knows that there can be no colour without extension, and that there cannot be any colour together with extension otherwise than with all the tactile and other qualities. This lack of independence of single qualities means, as I have discovered, that the natural independence and essence of 'substance' or 'thing' can 28

A. Meinong, "On Objects of Higher Order", p. 159.

32

ANNA SIERSZULSKA be found in the fact that a complex consists of qualities mutually dependent upon each other. Only when object theoretical considerations directed my attention to this, which happened quite late, did it become clear to me that through such a conception the proper idea of a 'thing' is truly lost. This idea becomes transparent, disregarding other deficiencies of independence, in the comparison between the meanings of the expressions 'green' and 'something green'.29

We may have an idea of 'green', and it may even really exist in the mind as we think about it, but in reality there are only particular moments of 'green' existing in 'something green,' to the extent to which their bearer also exists. And so, the greenness of one object is never identical to the greenness of another one, since all instances of greenness exist only as distinct moments of different things. Our general idea of 'green' is obtained by way of abstraction from these different green things. The general idea is the outcome of a mental operation, which explains the presence of general terms in language, but Meinong's view does not allow for any subsistence of universals. Nominalism binds his theory closer to reality, which consists solely of ordinary objects as concrete pieces of substance. The next step is the rejection of the idealistic conception that all relations are produced by the cognizing mind. He discovers that there are real relations, as well as real complexes. In the same way it is clear that in our example of the four nuts, the fourness (Vierheit) cannot exist as a special item of reality beside the nuts. But if we connected correctly it cannot be denied that it subsists. The number four and naturally every other number represent a complexion which is as ideal as the relation of similarity. That there are real complexions beside ideal complexions 29

A. Meinong, Über die Erfahrungsgrundlagen unseres Wissens, p. 27: Jedermann weiß, daß es keine Farbe geben kann ohne Ausdehnung, und daß es Farbe mit Ausdehnung zusammen mindestens nicht gibt ohne allerlei taktile und andere Qualitäten. Die hierin zutage tretende Unselbständigkeit einzelner Qualitäten legt es, wie ich an mir selbst erfahren habe, sehr nahe, die natürliche Selbständigkeit und damit das Wesen der Substanz oder des Dinges darin zu suchen, daß es eben den Komplex der gegenseitig sozusagen aufeinander angewiesen Eigenschaften darstelle. Erst gegenstandstheoretische Erwägungen haben mich, und zwar recht spät, darauf aufmerksam gemacht, daß durch solche Auffassung der eigentliche charakteristische Dinggedanke in Wahrheit verloren geht. Denn dieser Gedanke kommt bereits ohne Rücksicht auf anderweitige Unselbständigkeiten im Gegensatz der Bedeutungen von "Grün" und "Grünes" zur Geltung.

MEINONG ON MEANING AND TRUTH

33

does not need to be substantiated by examples. Already prescientific experience yields abundant insight into the complexity of reality (...)30

Often the inferiora of a complex are real objects, as in this example with nuts, but neither their 'fourness', nor the set of four nuts, nor number four as such, are real complexes. From this it does not follow that there are no real complexes at all. Experience tells us that there are real complexes as well as ideal ones. Of course, real complexes, like all objects of higher order, cannot exist, they can at most subsist. What is the difference, then, between ideal and real complexes and relations? It seems that the main difference is that the latter are simply not produced in the mind, but they are objectively there and can be perceived. Real relations and complexes are perceived in internal perception. They depend upon existent objects of internal perception, such as mental presentations and ideas in the mind, but they are not related to existent objects of external perception. They are perceived due to what is apparent about the relations between the immediate objects of mental acts, which are the only objects directly accessible to consciousness. So with objects of higher order it is not the case that they are real when they exist in the world and ideal when they only subsist. It is rather a division between what is objectively there to be perceived and what is a result of an intellectual activity. The contrast between real and ideal could have been characterized in a different manner which must still be considered at length. Everything real, if it exists, must, under favorable conditions, be able to be perceived by the cognizing and sufficiently capable subject. (...) However, things are different when we realize that there is a whole class of objects which essentially are not perceptible. They are the ideal objects.31

Ideal objects of higher order cannot be subjects to passive internal perception, because they have to be actively produced. Although ideal relations are not perceptible, they can, nevertheless, subsist. Our conviction about their subsistence is confirmed by self-evident a priori judgments. Real relations, on the other hand, must be perceived, they are not created in the mind. Internal perception is fully reliable only when it concerns mentally existing objects, but not so much when it concerns objects 30 31

A. Meinong, "On Objects of Higher Order", p. 150. A. Meinong, "On Objects of Higher Order", p. 151.

34

ANNA SIERSZULSKA

existing in the world, as their auxiliaries are only pseudo-perceived internally. Therefore, the only real relations which Meinong considers at first are the following real relations: between a presentation and its object, between ideas that form a complex idea, and between an ideal relation and the foundations of it. All of these depend upon objects proper to internal perception, about the existence and nature of which there can be no doubt. Later, Meinong articulates the position that some of the objects of higher order which he was taking to belong to ideal relations before, must be in agreement with the way things are related in reality. Everyone feels justified to believe – put differently: everyone has evidence – that what looks different, similar, equal also is different, similar, equal. Insofar as this is correct, it is clear first of all, that what appears to us as o'1, o'2 etc. is not just the substantial constituent o of our earlier symbolic notation; for the various appearing things all agree in that they are things. Rather, there correspond to the phenomenal determinations o'1, o'2 etc. certain noumenal determinations o''1, o''2 etc. for which it is simply evident that the same relations of comparison hold between them as between the o'; this implies at the same time the rather obvious assertion that the substantial moment o, which is secured by the already claimed good evidence, does not exist in an unnatural or, really, impossible isolation, but that what exists are in any case things which have properties.32

Even the relations of comparison, which he was using as typical examples of ideal objects, and which seem to be solely the products of the mental activity of comparing, must be somehow justified by the fact that such relations of comparison actually hold about existing objects. The qualities of objects as they are presented in the mind, called here 'phenomenal determinations', correspond, according to Meinong, to the actual properties of the external objects, called above 'noumenal determinations'. He is realistically minded enough to believe that such a correspondence is the case whenever the representation of the auxiliary object is adequate as confirmed by external perception. Then, the similarities and differences between internally perceived qualities of auxiliary objects correspond to those that hold for the qualities of external objects. Both real and ideal relations and complexes are particulars on Meinong's view. They subsist as individual objects of higher order. There 32

A. Meinong, Über die Erfahrungsgrundlagen, p. 94.

MEINONG ON MEANING AND TRUTH

35

are, in other words, as many relations of difference as there are cases that some objects are different from each other. And there is no 'difference' as such, which could help to build an idealistic world of Platonic ideas. For example, let us compare two colors A and B, say red and green. What happens there undoubtedly cannot be made clear by mere description without recourse to experience, that is, internal perception. But some important statements can be made. The A and the B idea participate at any rate, in the whole process. This can only mean that both ideas enter into a certain real relation with each other. The operation aimed at producing the relation brings about, under sufficiently favorable conditions, a new idea, namely the idea of the difference, naturally not of difference in general but of the specific difference between A and B.33

The only 'difference' there is, is the one actually subsisting between objects in each case, and it is neither a universal nor an existing entity. In Meinong's example quoted here, the difference holds for ideas, and so it is a real relation. The difference in this case is a real particular. However, when relations of comparison hold for external objects, they tend to be seen as ideal relations. They are, therefore, ideal particulars. They are ideal objects of higher order in the sense that they are not directly perceived, but produced by mental activity. And they are particulars, since they are built upon particular objects. Ideal objects of higher order are also ideal in the customary sense, because they subsist timelessly, even when their objects cease to exist. We can speak about the differences which used to hold for concrete objects in the past. So in considering some objects of higher order to be ideal particulars there is no contradiction. Apart from particular individuals, which are either ordinary objects or objects of higher order, Meinong's spectrum of objects-beyond-being includes general objects. Some of them subsist, like numbers and other mathematical objects. The rest are just general meaning-objects (Begriffsgegenstände), without any kind of being, but they may be used as auxiliary objects to refer to existing or subsisting entities. Meinong argues that it is a fact that we have general ideas obtained by abstraction and any ultra-nominalistic reduction in this respect is not justified. For years and even decades I have been opposed to a nominalistic solution of the problem of abstraction. I really don't think that I have to essentially correct today 33

A.Meinong, "On Objects of Higher Order", p. 152.

36

ANNA SIERSZULSKA what I said then(...) If I have no ideas at all of relations and complexions, how do I know, then, or how can I even think it out that they are relations and complexions, according to whose similarity the 'collectively apprehended' objects are grouped. That of which I have no idea I can even less know. (...) Instead of saying: similar complexions and relations are joined to the same terms, we could try out the assumption that the inferiora, as I call them, together with the word which is associated in virtue of the inferiora make up the complexion or relation, or, in short, object of higher order. The demand of 'lex parsimoniae' would then be especially satisfied. But can we hope to get by with such simple means?34

He thinks that similar relations and complexes are described using a single general idea, like 'difference', because of some similarity between them which we understand to be the relation of difference in each case, and not some other kind of connection. It would be perhaps possible to claim that there is no such general idea available and we only associate the inferiora of a relation with a certain word. He does not believe, however, that this solution is good enough. And he may be right in that, as it would be extremely difficult to explain why we associate given inferiora which we perceive as different with the word 'difference' if we have no understanding altogether of the general idea behind this word. It is probably correct to say, then, that Meinong's theory is not maximally nominalistic. ♦ Relations and complexes as founded objects Objects of higher order are always dependent objects, but the dependence is not always of the same kind. Real relations and complexes depend for their subsistence upon the existence of their inferiora. Ideal relations do not depend upon the real existence of their inferiora but they are founded upon them, which involves logical necessity. Objectives, on the other hand, are neither dependent upon the existence of the objects they are concerned with, nor founded in the sense of implying any necessary connection. Objectives merely presuppose other objects, which themselves may be non-subsisting, in spite of the subsistence of the objective. We can observe that the dependence which Meinong calls 'founding' concerns only ideal objects of higher order, and only those which involve a 34

A. Meinong, "On Objects of Higher Order", p. 175.

MEINONG ON MEANING AND TRUTH

37

necessary relation between the objects which are supposed to be the foundations. Thus, his understanding of founded objects is rather narrow. Depending upon the character of said real relations and under favorable conditions, ideas can be obtained of superiora of those objects which are connected with their inferiora by logical necessity. Keeping these matters in mind, I call the just sketched process founding (Fundierung), or more precise, founding of the superiora in question by their inferiora (...) Thus, founding, or the foundation process achieves the same for ideas of ideal objects as perception does for ideas of real objects. (...) Most of that which forces itself as 'relation' upon the pre-psychologic thinking is of ideal nature and is, therefore, founded.35

One might wonder why these should be necessary relations that two objects are different or that they are two, as in Meinong's example. Such relations between objects seem to be entirely contingent. However, as it was mentioned earlier, ideal relations are produced in the mind and they are relations between internally presented objects. So they do not concern any longer the contingencies of the external world, because comparisons between presented objects take place in the a priori sphere, where whatever holds is also necessary. If the objects in question are presented adequately, then the relations between the presented objects will correspond to those between the external objects. Yet the judgments apprehending these relations, are, quite strangely, a priori judgments, because there is no possibility to apprehend directly any relations in the external world. The only relations which can be perceived directly are these between mental objects as such, and they are called real relations. All relations between objects which do not primarily belong to internal perception are regarded as ideal relations, although they may be correct about the way objects are related in external reality. Concerning the same objects I can observe once that they are different, and at another time, that they are two. The first time I compared, the second time, I combined. Otherwise, however, in the given situation, that is, concerning the inferiora, both are equally necessary, namely to be different and to be two. (...) At first glance, it might not be so apparent that so called higher intellectual operations, especially judgments, can induce foundation processes from which

35

A. Meinong, "On Objects of Higher Order", p. 153.

38

ANNA SIERSZULSKA concepts of certain higher logical dignities result as, for example, possibility, necessity including disposition, causality and other derivatives.36

In fact, all relations and complexes concerning objects of the external world are ideal and founded with necessity by the qualities of the auxiliary objects presented in the mind. One can see immediately that this entire conception can only remain realistic due to Meinong's strong faith in our representing ability. Another difficulty for developing a realistic epistemology on the grounds of Meinong's theory of objects of higher order arises from the fact that reality is basically not divided into separate units, but rather forms a continuum. Meinong discusses the extreme cases of this lack of differentiation, though he is aware that the objection poses a threat for his theory. According to the general discussions of the first section, inferiora of the same superius can be, but do not have to be mutually discrete, and there are inferiora which are continuously connected. The objection denies the latter case: in reality, the continuum does not have infinitely many parts, but no part at all, it is an undivided unit. (...) If that is so we cannot speak of objects of higher order in those cases because there are no inferiora. If the objection is justified within the limits of its claim then it also poses a threat to the whole conception of objects of higher order because it is essentially by the same deliberation that the proponents of founded objects argue for a special superius, once proceeding from discrete notes to a melody and once proceeding from continuous local determinations which, in the sense of the objection, are only fictitious parts, to a shape.37

If the parts of continuum which we take to be the inferiora of objects of higher order were only fictitious, then there would be no such objects. Or they might at most belong to the 'non-subsisting' objects, but it would be by no means correct to claim that they can subsist. I think that the example with a shape explains the problem well. Space is a continuum and when we pick out some spatial determinations which form a certain shape, we divide the continuum into parts which are in a way fictitious. Yet anyone can apprehend a shape and tell what kind of shape it is. The shape is, therefore, something objective, not subjective. It may be apprehended by many 36 37

A. Meinong, "On Objects of Higher Order", p. 153-4. A. Meinong, "On Objects of Higher Order", p. 167.

MEINONG ON MEANING AND TRUTH

39

subjects of cognition. The situation is the same with all objects of higher order: even though we pick out their inferiora out of the continuum of reality, the relations and complexes at which we eventually arrive may be subsisting ones. Whatever is not simple but complex should seem to have parts. In the meantime the difficulty has become a terminological one. Whatever is divisible must contain material that can be differentiated: but the different material (das Verschiedene) which it contains does not have to differentiate itself into natural units.38

It is not required that for any complexes to subsist, reality must be naturally divided into pieces that correspond to these complexes. It does not even have to be divided into the objects that constitute their inferiora. This seems to be parallel to saying, in a different terminology, that there are no states of affairs reality would be divided into. We divide reality, since it is divisible. Meinong says that a continuum has undetermined component parts, and the actual divisions have to be fictitious in this sense. But once they are made, it is an objective matter which relations hold between the inferiora so differentiated, and which complexes subsist as the superiora containing these components. Reality can be perceived from different points of view, according to Meinong's theory, but it cannot be changed. It is clear from what we said above, that the many or even infinitely many divisions which can be undergone by continua are always brought into the continua and they, by their very nature, have merely undetermined component parts. In so far, Schumann is correct in calling those parts fictitious. But is he also correct in stating that that sort of fictitiousness also constitutes a difficulty for the foundation theory?39

Real and ideal complexes are not themselves existent entities and they do not result from a natural order of reality. Nevertheless, they may subsist, and then the objectives by means of which they are apprehended also subsist about reality. And so they are factual and they are true.

38 39

A. Meinong, "On Objects of Higher Order", p. 169. A. Meinong, "On Objects of Higher Order", p. 171.

40

ANNA SIERSZULSKA ♦ Differences between objectives and other objects of higher order

Objectives, which assume the role of propositions in Meinong's conception, belong to objects of higher order because they are dependent upon other objects. However, one would hesitate to call them objects at all, and certainly not in the customary sense. The broad use of the term 'object' in the theory of objects allows to apply this term to whatever is the object of intention of a mental act. Since Meinong is convinced that judgments and assumptions, like all other mental acts, possess objects specific to them, he arrives at the theory of objectives. Since objectives are the proper objects intended in the acts of judgment and assumption, they are called objects as well, even though they are neither real existing objects, nor do they belong to the general category of objecta. Objectives are entities of a distinct character. They serve as means to apprehend objecta of higher order. For it is difficult to expect e.g. that a complex together with its spatiotemporal components will appear in the mind. It can only be apprehended by means of an objective. This objective becomes then the immediate object of the act apprehending the complex. So, in any case, an objective can always be regarded as a meaning-object, belonging to Meinong's spectrum of pure objects. And when it happens to be factual, we may say that it also subsists. If in view of the facts, we must decide to admit that every knowing, no matter whether it is affirmative or negative, knows 'something', that every judgment judges 'something,' which is not an objectum but which does stand over against the judgment in question in the way in which the objectum stands over against its representation – if we admit this, then the ability to apprehend this 'something' has therewith already been granted the judgment. It is the business of each judgment to secure its immediate objective on its own authority, as it were. (...) Objectives can doubtless be apprehended by means of assumptions, too, as well as by means of judgments; but not also by means of representations, and for the apprehension of objectives there are no further means available to us.40

It is obvious from the beginning that objectives differ very much from ordinary objects of representation, because they cannot be mentally represented, they can only be judged or assumed. Objectives involve 40

A. Meinong, On Assumptions, p. 106.

MEINONG ON MEANING AND TRUTH

41

representations of the objecta judged about, but they can never be themselves objects of representation. Nevertheless, objectives share with other meaning-objects the status of mind-independent entities. They are no more mere psychic contents than any other objects in Meinong's Aussersein. Under such circumstances, it deserves to be explicitly emphasized above all, that the objective, like any other object, generally admits of the apsychological approach and therefore demands it. More precisely, this emphasis is directed against what appear to me to be two thoroughly erroneous views. First and foremost, it is directed against the opinion, in conformity with the 'prejudice in favour of the actual,' that our knowing, or at least our interest can only have to do with what is actual. Thus, if the objective does not belong to actuality, then it would be at most the experience apprehending it that could concern science. But the emphasis is also directed against the relativistic interpretation of the concept of an object (...) This interpretation strips the objective of that independence vis-à-vis the apprehending experience which, in my belief, belongs to the objective no less essentially that to any bit of actuality 'outside of us.' A point we shall have to bring up for separate discussion in just a moment is that neither existence nor being in any sense whatsoever is ascribable to an objective on the basis of that independence.41

The 'prejudice in favour of the actual,' mentioned so often by Meinong, applies to objectives even more accurately than to any other objects he has in mind. In the case of typical nonexistent objects, this reductionist attitude may seem justified to some extent. But objectives are 'nonexistent' objects even when they are true. Does this mean that the only genuine scientific interest should be directed to concrete 'judgings' and their truth, or possibly falsehood? Meinong does not think that it is necessary to disregard the nonexistent, but objective meaning-objects in the name of such reductionism. He emphasizes their independence from mental acts. But this is not a basis, according to his theory, to ascribe to false objectives any kind of being they would not possess otherwise. Belonging to Aussersein has nothing to do with the existence of an object. Objectives belong to objects of higher order because they are 'logically' dependent on other objects. As Meinong says, an objective must always have something 'under itself', whether it should be an objectum or another 41

A. Meinong, On Assumptions, p. 49.

42

ANNA SIERSZULSKA

objective. Therefore, objectives assume the role of objects of higher order in two alternative ways, either with respect to objecta, or to other objectives. From this we can see that objectives must occur very often as inferiora, not only as superiora. What is judged about can just as well be an objectum. Accordingly, one cannot maintain that every objective must likewise have objectives of lower order under it, as well as objectives of higher order over it. Certainly it must have an object 'under itself', in the sense just indicated, but this object can also be an objectum. (...) So to that extent every objective occupies the position of an object of higher order. Borrowing a term proposed elsewhere for use in connection with a judgment, one might designate the object or objects on which an objective is positioned as the material of the objective.42

At first, objectives are not mentioned among objects of higher order, because it is not even clear that they could be regarded as objects at all. When Meinong becomes inclined to think that what is directly apprehended in acts of judgment and assumption are actually objects of some kind, he admits that not only objecta, i.e. complexes and relations, but also objectives are objects of higher order, in spite of their different character as entities. Each objective is dependent on one or more objecta, and when there is no objectum under it, then there is an objective occupying the position of an objectum. And further, an objective which has another one under itself, may have an objective of even higher order above itself. Thus, there are objectives of higher and still higher order, which could theoretically lead to an infinite regress of different levels of objectives. The notion of objects of higher order was initially formulated in connection with relata and complexes. It was not known at the time, at least not in a strict theoretical sense, that objectives were to be counted as objects as well as objecta. Only objecta, therefore, were counted as objects of higher order. When objectives had also gained consideration in the theory of objects, it became apparent that there were different levels among them. In comparison with the objective 'A is B' the objectives 'It is a fact that A is B' or 'It is true that A is B' must without doubt be considered as objectives of higher order; and this is in general true for each objective that itself includes an objective in its 'material.' We can also express this 42

A. Meinong, On Assumptions, pp. 50-1.

MEINONG ON MEANING AND TRUTH

43

as follows: each objective is an objective of higher order with respect to another objective if the latter occupies the place of an objectum in the former.43

It may be interesting to notice that the objective 'It is true that A is B' is considered by Meinong to be an objective of higher order with respect to the objective 'that A is B'. So it seems that the paradox of the liar is ruled out by Meinong's conception quite a while before Tarski's solution to this paradox which makes use of a language of higher order. Moreover, discovering different levels between objectives convinced Meinong completely that not only relations and complexes are objects of higher order. Another reason to consider objectives together with relations and complexes is their similar existential status. Namely, it is true about objectives to the same degree as about other objects of higher order that they are never existent entities and at most can possess subsistence. In their place within the opposition of existence and subsistence which comprehends all that is, objectives are analogous to the objects of higher order hitherto considered – not all of them, but the ones most frequently considered. 'That A exists,' or again, 'that it does not exist,' is something that 'subsists' if the judgment immediately apprehending it might have been correctly made; but it does not exist once over, as it were.44

If we take an objective of being, such as 'that A does not exist', then nobody will think that 'A's non-existence' is another existent entity in the real world. But since the objective is true and factual, Meinong claims that it subsists. On the other hand, even if we take an objective of so-being, like 'that A is green,' then 'A's being green' is not an existent object but a subsisting complex, consisting of the existent A, its particular greenness, and the relevant relation, while the objective 'that A is green' also subsists. Naturally, the objective can even less be an existent entity than the complex can, because it does not have any existing components. We can say that an objective subsists only in the sense that it obtains and it is true.

43 44

A. Meinong, On Emotional Presentation, pp. 93-4. A. Meinong, On Assumptions, p. 51.

44

ANNA SIERSZULSKA

Objects of Higher Order: similarities and differences Objecta: relations and complexes

Objectives

Real

subsisting

existentially dependent

apprehended indirectly

Ideal

subsisting

founded with necessity

apprehended indirectly

Factual

subsisting

nonfactual

nonsubsisting

logically presupposing subordinate objects

- meanings of sentences; - means to apprehend relations and complexes

♦ Objectives as abstract entities Objectives differ from the other objects of higher order in many ways. To begin with, relations and complexes are divided into real and ideal, and there is no such opposition among objectives. There are no real objectives, simply because real relations between existent (mental) objects are there to be perceived and apprehended by means of objectives, which are not themselves objecta to be perceived. Also, in contrast to real objects of higher order, objectives are timeless. Moreover, objectives include no components at all, they do not contain or consist of any other objects, in which they differ from complexes. Their way of dependence on other objects amounts to presupposing other objects, without necessarily presupposing their being. (...) Our picture of various levels of order is only an indication of the fact that objects of higher order are based or built upon objects of lower order and that the former could not be there if the latter were not there first. The 'being there' is to be understood, as we have already seen, not only in the sense of existence or subsistence but also in the sense of Aussersein; and the 'at first' is meant in the peculiar, timeless sense of the logically prior. All this applies, without the slightest difficulty, to objectives.45

45

A. Meinong, On Assumptions, p. 94.

MEINONG ON MEANING AND TRUTH

45

Logical priority of the objects upon which objectives are dependent does not imply, of course, that the subsistence of objectives depends upon the existence or subsistence of their objects. There are true, and so subsisting objectives, about non-subsisting objects, e.g. 'that Pegasus has wings,' and there are true negative objectives, e.g. 'that Pegasus does not exist.' This lack of existential dependence holds also for ideal relations and complexes. But objectives do not resemble ideal complexes, since they do not share the necessity of being founded upon their inferiora that is typical for ideal objects of higher order. It can, by the way, also be seen that the characteristic dependence of the superius upon its inferiora is by no means always an a priori or necessary dependence. This can readily be seen in connection with real relata (Realrelate) and real complexes (Realkomplexe), which were counted by me from the very beginning among objects of higher order. The same holds of objectives and is substantiated by the very simple fact that there is other knowledge besides a priori knowledge and that the lack of necessity in the former is not to be explained by our insufficient intelligence. There is no object whose existence follows a priori from its nature. Existence is, nonetheless, as much an object of higher order as subsistence.46

The absence of any necessary dependence on their subordinate objects is quite understandable in the case of objectives, as such necessity would allow only for a priori judgments to take place. Meinong observes that such a situation would lead to an unacceptable consequence that we would have a priori knowledge exclusively, which is not the case. So, beside real relations and complexes, which are not based upon necessary connections, also objectives can be free from this requirement. Meinong indicates here especially the objectives of being, since they rely directly on external perception and they are not related to a priori judgments, as the objectives of 'so-being' are. It also seems to apply to negations of objectives that they do not involve any necessary dependence. Founded objects are connected with their fundamenta by necessity. Red and green are not just different; they must be different. Likewise, 3 is not just as a matter of fact greater than 2, but also necessarily thus; and so on. Now, there are certainly also negations that hold true necessarily. It is certainly every bit as necessary to negate sameness between red and green as it is to affirm differentness between 46

A. Meinong, On Assumptions, p. 95.

46

ANNA SIERSZULSKA them. But it is every bit as certain that there are negations in which necessity has no part. That a released stone goes in the direction of the earth and not away from it – that is a piece of information which, at least before the forum of human knowledge, could make no more claim to necessarily holding true on its negative side than it could on its positive side. So to the non-A as such, there still does not attach any necessity; and in that case, what we have here is not a founded object, either.47

Apparently, an objective is involved in such cases, not a negative objectum. And we would not call such an objective a founded object, based upon a necessary dependence. Besides, it is often a question of a posteriori knowledge that something is not this or that way. Objectives can apprehend these negative circumstances, because they are not founded objects and their dependence on any inferiora they may have is very loose. I think that it should be especially emphasised that objectives are not, strictly speaking, objects of any kind. Characterizing objects of higher order Meinong explains that what binds the components of a complex is a relation, which itself is its component. An objective is not really any object but the means to apprehend the complex together with the relation, which might roughly mean that an objective enables us to apprehend a complex as a certain unity of its components. Generally speaking, then, where there is a complex, there is also an objective as integrating factor in it, and one who wants to apprehend the complex cannot do it otherwise than by apprehending the objective, too. From that it now becomes directly understandable, in the first place, why representing fails in the presence of complexes in the broader sense just employed – objects of higher order, as one can also say – when these objects cannot somehow be divested of their complex character by means of abstraction.48

It follows that an objective, being an 'integrating factor' or a certain logical structure, does not, above all, literally contain any objects as its components. It is not a complex and not a relation either, but a coordination that obtains between the components of a complex, and a meaning-object that makes it possible to apprehend them as a unity. Thus, we can say that objectives are abstract entities. Just as the content of an act 47 48

A. Meinong, On Assumptions, p. 18. A. Meinong, On Assumptions, pp. 201-2.

MEINONG ON MEANING AND TRUTH

47

of representation is something particular and the presented meaning-object is something abstract, common to many presentations even if a particular object is presented, so the meaning-object of a judgment is also not a particular entity in itself. It may be objected that if many representations ultimately refer to particulars in the external world, and many judgments apprehend particular relations and complexes, then the objects immediately intended in such mental acts must be particulars as well. In reply to this one can say that the auxiliary object of a representation need not be a particular itself, although it is implected as to its properties in a particular object. In fact, being every time an incomplete object, the immediate object of a presentation has no chance to be a particular. An objective that is immediately intended is also incomplete in the representations of the objects involved, so it could not be a particular. But when we take the factual objective as such, it will not turn out to be a particular either, for it does not contain its subordinate objects as components. In the case of objectives it is even more natural to claim that there is a difference between being a particular and being about particulars. Complexes are particulars for they include particular entities, relations have particulars as members and even ideal relations can be timeless, but particular. In opposition to this, objectives may involve particular objects, but the dependence is understood only as logical priority. It must be remembered that objectives are very different from other objects: they are not objecta – so it might be even doubted if such classifications as belonging to particulars, or not, can apply to them. They are abstract entities, even when taken as structures that obtain between real objects in the world.

MEINONG ON MEANING AND TRUTH

49

PART II MEANING AND TRUTH CHAPTER 3 MEINONG'S THEORY OF MEANING Meinong's theory of meaning has often been perceived as based upon a simple 'word – object' relation. In his earlier writings it seems to amount to this two-element pattern, which is a reflection of the intentional relation between a mental act and its object (see Ch. 1). In the later period, the notion of the object of a mental act becomes more complex, as it splits into the immediate auxiliary object and the ultimate object which is intended. Accordingly, only the immediate objects turn out to be the proper meaning entities in his conception, while the more remote ultimate objects are apprehended by means of them. In this way Meinong introduces something like a distinction between senses and denotations of linguistic expressions, except that the entities playing each of these roles are called 'objects'. This conception of sense-objects is compared below to Husserl's theory of noematic sense and Frege's 'sense – reference' distinction. It is argued that Husserl's senses are more on the side of what is conceived by the cognizing mind and they make Husserl's conception gravitate towards idealism. However, Frege's distinction is not burdened by any such connotations. Frege's senses mediate reference to objects and they lie more on the objective side of what can be derived from linguistic practice. For this reason they show more similarity to what Meinong wants to convey by means of his theory of auxiliary objects. Further, Meinong's objectives are presented in the role of the meaningentities typically expressed by sentences, or sentence equivalents. On the basis of the theory outlined in Ch. 2, they are treated as objects of higher order of an abstract character. Both in their status as mind-independent meaning-entities and in the tasks performed, objectives resemble propositions, such as Bolzano's Satz an sich or Frege's thought. It will later

50

ANNA SIERSZULSKA

be argued that objectives resemble Bolzano's and Frege's propositions also in their function-like nature. ♦ Meaning as 'sense' and meaning as 'object' On Meinong's view, a distinct meaning element is present, but it does not possess the character of an abstract 'content', which would be more typical for what we traditionally call meaning-entities. The content is specific for each mental act and it is subjective. For reasons explained in Ch.1, I propose to call Meinong's content 'a subjective sense entity', in contrast to his very broad notion of the object of a mental act, which I call 'an objective sense entity'. In particular, auxiliary objects of different kinds which serve to apprehend other objects, are not at all 'contents', but 'objects' in Meinong's terminology. We will adopt the definitions: The content of a mental act – a subjective sense entity The immediate object of a mental act – an objective sense entity The above is the notion of an object in the broadest sense and it is also 'weaker' than the ordinary notion of an object. The object is considered here as any sense-correlate belonging to Aussersein, regardless of whether it possesses any sort of being. The only requirement is that it should be conceivable. Such entities would not normally be called objects. Whatever is not merely a subjective content of thought, because it can be intended, referred to and spoken about by many different subjects, is called an object on this account. In all this, there is more than a mere 'weakening' of the initial definition of objectivity. For if objectivity consists in the 'having' of an object, and the 'had' object must be one that has being, than a representation simply does not have an object in any of the instances of fiction...49

The weakened definition of an object in Meinong's theory is a means to reflect the object-directedness of all intentional and linguistic phenomena. Aussersein provides non-subjective objects of reference for all potentially referential elements of language. 49

A. Meinong, On Assumptions, p. 170.

MEINONG ON MEANING AND TRUTH

51

Meinong's two major works: Über Annahmen and Über Möglichkeit und Wahrscheinlichkeit, contain an outline of his conception of meaning, which has to be reconstructed from various passages because presenting a theory of linguistic meaning is not the main concern of his theory. It seems that already in On Assumptions, which is the earlier work, he treats only the objects immediately presented by contents of intentional acts as meanings, even though he does not state clearly that these immediate objects must be distinguished from the ultimate existent objects of reference if such there are. The reason for this is that for Meinong the ultimate object of reference contains within itself the object immediately presented in an intentional act (the auxiliary object) as a 'subset' of its properties. He never makes a sharp distinction between the auxiliary and the ultimate object. He claims that the former can be identified with the latter in an intentional act, because the auxiliary object is just a means to intend an existent thing. In On Assumptions, he apparently proposes a typical extensional two-element 'word – object' relation in place of a theory of meaning, and many philosophers still understand his conception of meaning in this way, but I think that, in fact, he is writing in this work about the relation between the presenting experience and the object immediately presented by its content as the meaning of an expression. And so it will probably be permissible to put the matter quite generally and say that a word always 'signifies' the object of the representation it 'expresses' and, conversely, that it expresses the representation of the object it signifies. (...) More precisely, a word signifies only to the extent that it expresses an intellectual experience, the object of the experience in that case constituting the signification of the word.50

This is not a direct extensional word-object relation, because the object is not understood as an ordinary existent thing. There is no mention here about any objects except for those presented by the contents of mental experiences. And such objects may very well be non-existent entities. These are not simply the objects denoted by linguistic expressions. From his later book we learn that the objects presented by the contents are really just the object-like (reified) senses of the intentional acts, not at all the objects denoted. 50

A.Meinong, On Assumptions, p. 25.

52

ANNA SIERSZULSKA (...) the auxiliary function should really be ascribed not to the object, but to the content in relation to which this object stands as its nearest object. But the content is (...) hardly available to controlled reflection, while its nearest object is available easily and, therefore, it is advisable, as a kind of fiction, that we ascribe to the nearest object what in fact is the matter of content, so that we can speak about an auxiliary object in relation to which the intended object stands as the ultimate object.51

In Über Möglichkeit und Wahrscheinlichkeit, he presents the final version of his theory of meaning. He speaks about: (...) complete objects, which can never be presented, because of the infinite number of their qualities. In order to grasp them we must rely, always and without exception, upon auxiliary objects. (...) After the previous remarks it is obvious that very different auxiliary objects can be used to refer to the same ultimate object (...) Word meanings are very often auxiliary objects. (...) What relates certain objects to certain words is that, as I tried to show in another place, these words 'express' experiences, in particular they express presentations, the immediate objects of which are meaning objects.52

51

A. Meinong, Über Moglichkeit, pp. 195-6: "(...) die Hilfe leistet ja strenggenommen nicht der Gegenstand, sondern der ihm als seinem nächsten Gegenstande zugeordnete Inhalt. Aber der Inhalt ist, worin immer das seinen Grund haben mag, der kontrollierenden Selbstwahrnehmung ebenso schwer zugänglich, als sein nächster Gegenstand ihr leicht zugänglich ist; es empfiehlt sich daher eine Art Fiktion, die gestattet, am nächsten Gegenstande zu betrachten, was eigentlich Sache des Inhaltes ist, so daß von einem Hilfsgegenstande geredet werden kann, dem der gemeinte Gegenstand ganz wohl als Zielgegenstand gegenüberzustellen ist." 52 A.Meinong, Über Moglichkeit, pp.197-198: "(...) vollständige Gegenstände der unendlich großen Anzahl ihrer Bestimmungen wegen niemals präsent sein können. Sie zu erfassen, dazu sind wir also jederzeit und ausnahmslos auf Hilfsgegenstände angewiesen. (…) Auch daß demselben Zielgegenstande eventuell sehr verschiedene Hilfsgegenstände gleichsam dienstbar gemacht werden können, ist nach früherem selbstverständlich. (…) Wortbedeutungen sind sehr häufig Hilfsgegenstände. (…) Was gewisse Gegenstände als 'Bedeutungen' an gewisse Wörter knüpft, ist, wie ich an anderer Stelle darzutun versucht habe, daß diese Wörter Erlebnisse, zunächst insbesondere Vorstellungsinhalte 'ausdrücken', denen die Bedeutungsgegenstände als nächste Gegenstände zugeordnet sind." (My translation, I translate 'Hilfsgegenstand' as 'auxiliary object' and 'Zielgegenstand' as 'ultimate object' following John Findlay, Meinong's Theory of Objects and Values, p. 177.)

MEINONG ON MEANING AND TRUTH

53

In this work Meinong sets out a very refined theory of meaning. His meaning objects or auxiliary objects, are distinguished from both the presenting contents and from the denotations of expressions. The meaning of an expression is the object as it is given to the mind, or grasped. At this point, it will be useful to define some terms. Most of them have already been introduced in the discussion of the theory of objects and objects of higher order. I will begin with some general notions, concerning all mental acts. The content of a mental act (Inhalt) – a subjective sense entity, related to a particular act and non-repeatable. The immediate object of a mental act (unmittelbares Objekt) – an objective sense entity, repeatable (in an intentional relation) as the object of many different acts of different subjects. The ultimate object of a mental act (Zielgegenstand) – the object which is the ultimate target of intention: it may be, but need not be, merely a sense entity. It is the object of mental and linguistic reference. If external and existent, it is the denotation of a word or expression. The auxiliary object of a mental act (Hilfsgegenstand) – the immediate object in its relation to the ultimate object, i.e. playing the role of a meaning entity. It is the meaning of a word or expression. The nearest object (nächster Gegenstand) and the remote object (entfernter Gegenstand) of a mental act – another way to refer to the auxiliary and ultimate object. The incomplete object (unvollständig) – an expression used to describe the immediate object, in contrast to an existing ultimate object, which is complete in all its properties. Meinong's structure of an intentional act can be represented in this way: Scheme 1: content

Inhalt

- presents -

an immediate (nearest, auxiliary) object

- implected in -

[the ultimate object]

unmittelbares [der - presentiert - implektiert in Objekt Zielgegenstand] (Hilfsgegenstand)

54

ANNA SIERSZULSKA

The relation of linguistic reference may be represented by Scheme 2: E(1) – (C(1)) – AxO(1).... X..... [O] E(2) – (C(2)) – AxO(2).... In the scheme: X is the subject, subject X intends O, E is an expression, Es denote O (If existent), C is the content, the round brackets mean that the content is subjective and fleeting, AxO is an auxiliary object, O contains both AxO's as implected in it, O is the ultimate object. The square brackets mean that there may be no transcendent ultimate object. Since AxO1 and AxO2 are auxiliary objects used to grasp the ultimate object O, they are two different subsets of O's properties. There may be as many such auxiliary objects as there are possible subsets of O's properties. It is also important to note that every auxiliary object is an incomplete object, which can potentially become a means to intend not just one object, but each of several particular complete objects. In this sense, abstract incomplete objects play the role of meanings of general terms, e.g. 'a dog' which is not a particular one. In the cases when a particular object is intended, the auxiliary object can be identified with the ultimate object as 'the same' object on the basis of their ontic identity, even though they are not identical in all their properties (the auxiliary object is less rich). In this sense, the basic extensional word-object pattern of signification is preserved. Scheme 3: The extensional pattern of signification E(1)............... O E(2)................

Meinong's pattern of signification E(1).................AxO(1) ≈ O E(2)..................AxO(2) ≈ O

And for general meanings: E .............{O(1), O(2), O(3)}

E .............AxO ≈ O(1) v O(2) v O(3)

MEINONG ON MEANING AND TRUTH

55

Meinong's theory of meaning lies on the border between the extensional and the intensional semantic traditions, because it preserves the extensional pattern of signification in extensional contexts, but at the same time it provides means to account for the meaning of expressions in nonextensional contexts. Meinong's auxiliary objects can play the role of senses and simultaneously they can be identified with the signified objects. This is possible due to his conviction that there is no divide or chasm between the object as it is presented and as it is in reality, their ontic identity is natural to assume. Meinong's attitude to logic is practical and maybe this is why he does not try to develop any logical system that would be an idealized pattern for natural languages, a kind of ideal language, the idea of which was starting to become popular at his times. For Meinong, logical rules are applicable in natural language contexts and formalizations may be helpful, but only in an instrumental sense, which is well characterized by David Lindenfeld: While Meinong does not explicitly refer to logic as a type of language, it is clear from his discussion of it that he considers it to be analogous to particular use of language rather than to an ideal language. He viewed the traditional rules of inference, as given in syllogisms, as applicable to certain natural contexts only.53

However, Meinong's theory of meaning is universal with respect to particular natural languages, because it is an approach from the perspective of intentional phenomena, which are treated as more basic than their linguistic expression. The introduction of auxiliary objects as intermediate meaning entities between words and their denotations, allows to account for the differences in meaning of co-referential expressions, i.e. for their different senses. The auxiliary object is what words mean, the ultimate object is what they name. What does this theory contribute to the notion of an ideal language? It gives a satisfactory description of meaning that can be applied to all languages, and it elaborates some rudimentary aspects of structure: languages are intertranslatable insofar as the similarities and differences between their objects of reference can be analyzed in terms of objectives of so-being.54

53 54

D.Lindenfeld, The Transformation of Positivism, p. 146. D.Lindenfeld, The Transformation of Positivism, p. 163.

56

ANNA SIERSZULSKA

The meanings of expressions can be analyzed not just in terms of their objects of denotation, but also in terms of the properties ascribed to these objects of denotation by each expression, as shown by the examination of the auxiliary object related to each expression respectively. And it is probably right to maintain, as Lindenfeld does, that with the help of these abstract auxiliary objects it is easier to explain intertranslatability of languages. Meinong's meaning theory resembles certain ideas of Husserl, as both philosophers belong to the Brentano school. Nevertheless, their work develops in different directions, which will be visible if we compare Meinong's theory with Husserl's mature conception from Ideas II. While Husserl puts strong emphasis upon the noematic sense of an intentional act, Meinong merely mentions the contents related to acts, or to words and sentences. Thus the meaning element in Husserl's theory does posses the character of an abstract mental content, whereas Meinong's meaning element is close enough to Husserl's intentional object, because both are objects presented immediately by intentional contents. The important difference is that for Husserl an expression can be meaningful without a fulfilment of its noematic sense in the form of an intentional object, as it is the case with non-existent objects, while in Meinong's conception, the object immediately presented by the content must be there in each intentional act. Scheme 4: For Husserl, the intentional act can be represented in this way: X ––––––– E ––––––– NS ––––––– > [IO] ≈ [O] By comparison, for Meinong: X ––––––– E ––––––– (C) ––––––– AxO… implected in... [O] X is the subject, E is an expression, NS is the noematic sense, C is the content: for Husserl NS is a part of the overall content of the act; for Meinong C is present but it is subjective and fleeting, IO is the intentional object, AxO is the auxiliary object and it is implected in O, O is the ultimate object, the square brackets mean that IO and O may be absent, the

MEINONG ON MEANING AND TRUTH

57

relational axis for Husserl ends with an arrow to indicate directedness of the intentional relation towards an object that can remain unfulfilled. It is plain from the above scheme that these accounts of meaning differ. They differ in a single detail, whether each intentional act actually does possess an object of intention, or not. The advocates of intentional logics find it more convenient when the objective senses of intentional acts are reified in the way proposed by Meinong to provide an object for each act. It is easier then to treat such sense-objects as mind independent objective entities, than in the case of Husserlian senses as abstract contents. But almost everyone agrees that there is essentially no great discrepancy between the two conceptions, e.g. Zalta says: This intermediate view also incorporates the Husserlian analysis of intentionality to explain the directedness of mental states in general. In particular, the mental states these sentences describe have a "direction." The direction is represented by an A-object in its role as a noematic Sinn. The A-object attaches the properties involved in the cognitive state to a single focal point. Though Husserl would disagree with the view that thinking about Grendel has both a content and an object, we have offered reasons why it is important to think that one A-object serves as content, while another serves as object. Other than this discrepancy, we see no reason to think that there is any deep inconsistency between Husserl's views and Meinong's.55

Zalta's abstract meaning objects (A-objects) play the role of senses and also of 'denotations' in these cases when there are no existent objects of denotation available. Actually, not only Husserl would disagree that the name 'Grendel' has a denotation, but it is quite plausible to maintain that such terms have objects of intentional reference. Yet Husserl would disagree even with this claim. As Jacek Pa niczek points out,56 Husserl's and Meinong's conception represent two different approaches to intentionality, the directional one (intentional acts have contents directed at objects if there are such objects) and the relational one (all intentional acts, and all meaningful words, have objects of intentional reference). The noematic sense is an intentional medium and it can hardly be identified 55

Edward Zalta, Intensional Logic and the Metaphysics of Intentionality, p. 114. Jacek Pa niczek, The Logic of Intentional Objects. A Meinongian Version of Classical Logic, pp.116-18. 56

58

ANNA SIERSZULSKA

with the auxiliary object, which is an intentional correlate. However, Pa niczek says: But maybe they are, in a sense, complementary concepts referring to two different aspects of intentionality? Maybe the intentional correlate and the intentional medium are just two parts or aspects of the same entity? Notice in particular that even within the framework of the directional conception we used to talk about acts directed to non-existent objects, although the stipulation of intentional objects is theoretically superfluous in such cases. The affirmative answer to the question would make the two conceptions complementary.57

According to this position, the contents of acts appear to be almost indistinguishable from their intentional objects. This is very much in agreement with Meinong's original idea of the intentional relation, except that he would rather identify the sense with the content and disregard it, instead putting all his emphasis upon the auxiliary object. If we reverse his strategy and identify Meinong's auxiliary object with what may be called the objective part of the content of the act, the discrepancy between the two approaches will vanish, but at the same time the meaning element will be drawn closer to the mental or subjective and further from the objective side of the intentional relation. Something like this is characteristic of the late phase of Husserl's philosophy and it always indicates a possibility of an idealistic outcome. This is not a welcome result, since for Meinong the meaning object is either implected in or identical with the object of reference. The natural tendency, therefore, would be to keep the meaning element as close to the object of reference as possible, without denying certain analogies between these two theories of meaning. Therefore, let us say that in both theories there is a distinct meaning element which possesses the character of an objective sense entity, but in Husserl's theory from Ideas II the meaning element is an abstract content, called noematic sense, while in Meinong's theory from Über Möglichkeit it is an abstract object, called the auxiliary object. Of course, it does not follow that there is any reason to oppose the idea that auxiliary objects play the role of senses in Meinong's conception. Quite to the contrary, one might hope that this approach should become the 57

J.Pa niczek, The Logic of Intentional Objects. A Meinongian Version of Classical Logic, p. 118.

MEINONG ON MEANING AND TRUTH

59

standard interpretation of Meinong's theory of meaning. However, I would object a little to identifying all sense entities with the Husserlian conception of noematic sense as the abstract intentional content. This has become a popular view after Follesdal's demonstration of the analogy between Frege's and Husserl's notions of sense. But perhaps it would be better to keep Frege's , and also Meinong's , notion of sense apart from Husserl's . One may really not be sure whether Husserl's rich conception of the intentional phenomena is not too much to be considered while analysing the relation of linguistic reference in the Fregean fashion. Perhaps such an identification with Husserl's phenomenological notion of a noema could lead to a more subjective version of Frege's sense, or Meinong's auxiliary object, than one would wish to incorporate into a theory of meaning in any case. Frege's theory of meaning seems to be more suitable for drawing an analogy with Meinong's conception, because what Frege calls senses of expressions, as in the case of the 'Morning Star' and the 'Evening Star', can be identified, easily and without wrong connotations, with what Meinong calls 'the auxiliary objects' related to these expressions. In particular, Frege's theory does not carry with it the whole range of connotations related to Husserl's theory of noematic sense. I would say that Fregean senses resemble Meinong's auxiliary objects much more than they resemble Husserl's noemas. They are more on the objective side of linguistic reference, than on the side of the intentional content of what is meant. The relation of linguistic reference for Meinong (based on Scheme 2), and for Frege, looks like this: Scheme 5: Meinong's relation of reference

Frege's relation of reference

E(1).....................AxO(1)...

E(1)...................S(1)... O

E(2).....................AxO(2)...

O E(2)...................S(2)...

The expressions E(1) and E(2) express for Meinong different subjective intentional contents, which are omitted in this scheme as irrelevant [C(1) and C(2)], because they are subjective and it is by means of the presented auxiliary objects AxO(1) and AxO(2) that we determine the ultimate

60

ANNA SIERSZULSKA

reference to O. There are no 'senses' in Meinong's theory, the auxiliary objects play the role of senses. In Frege's theory the intentional content is not even mentioned, and the senses S1 and S2 determine the reference as the entities identified with O. For instance, the objects of reference of the expressions 'the Morning Star' and 'the Evening Star' are identified with the ultimate object of reference 'Venus.' This takes place in exactly the same way as in Meinong's theory, where the auxiliary objects determine the reference to the ultimate object O by being identified with O. Therefore, we can say that Frege's senses correspond to Meinong's auxiliary objects very closely, while their correspondence to Husserl's noemas is a little more loose. Yet, all the three meaning conceptions are based on the assumption that meaning consists in 'a way something is given', which is Frege's definition of 'sense', and which applies equally well to noemas and to auxiliary objects. Already Findlay mentions58 that the distinction between the auxiliary and the ultimate object shows some similarity to Frege's distinction between the sense and the reference of an expression. And Meinong's conception is presented as definitely resembling Frege's 'sense – reference' distinction in a paper by Peter Simons, entitled "Meinong's Theory of Sense and Reference." On the interpretation of Meinong's views provided by Simons, Meinong's incomplete auxiliary objects can be identified with Frege's senses. What is important about Simons' paper is that it is one of the interpretations according to which Meinong's semantics is not based on a simple word – object relation, and the only one on which the analogy between Frege's senses and Meinong's auxiliary objects receives proper attention. Simons writes: Of the various objects involved in this more complex analysis, the incomplete object quasi-presented by the content is the one fit to serve as sense of an expression, and is, if not in so many words, so selected by Meinong .59

There is a passage which suggests even greater similarity with Frege's sense-reference distinction: 58 59

J.Findlay, Meinong's Theory of Objects and Values, p. 184. Peter Simons, "Meinong's Theory of Sense and Reference", p. 179.

MEINONG ON MEANING AND TRUTH

61

The sense of a word or phrase is the open incomplete auxiliary object quasipresented by an experience aimed at grasping some intended target object, which in the context is that expression's referent.60

Unfortunately, in a more recent paper (with E. Morscher) "Meinong's Theory of Meaning", Simons partly withdraws this claim. By introducing the intermediate realm of senses between signs and their objects they [Frege and Husserl] can make ontological savings among the objects, which Meinong could not. For Meinong the choice was between a subjectivistic, psychologistic theory, or a full blown referential objectivistic theory, which is the one he develops. (...) His wide realm of objects is as wide as it is partly because it plays together the roles which Frege and Husserl divided between senses and referents. (...)Between the mental act and its object is inserted a quasi-intermediate object called the auxiliary object, and Meinong renames the object proper the target object. The auxiliary object is incomplete and its properties are, in favourable cases, all properties of the target object. Meinong enlists the target object associated with an act prompted by hearing or thinking of a word as that word's general or usual meaning, thus introducing something akin to Frege's familiar distinction between sense and reference. As something objective, such an auxiliary object can ensure the objectivity of the meaning of a word or phrase. The distinction is, however, not generalized beyond the case of nominal expressions and so does not become part of a systematically three-levelled semantics as in Frege, but it does show Meinong's flexibility and intellectual vitality even in his last years.61

The objection against Meinong's semantics that it is not systematically a three-levelled theory is not quite justified. Apart from nominal expressions, also whole sentences have meaning entities ascribed to them, since objectives are the meanings of sentences. We would have a two-levelled semantics for sentences, if sentences signified directly complexes in reality. In Meinong's theory, sentences do not signify complexes directly. Objectives expressed by sentences are the means to apprehend complexes and they are parallel to Frege's senses of sentences. The parallel can even be drawn so far as to assume a function-like character of objectives. The subject-predicate structure of Meinong's objectives is not an obstacle here, 60 61

Peter Simons, "Meinong's Theory of Sense and Reference", p. 184. E.Morscher, p. Simons, "Meinong's Theory of Meaning".

62

ANNA SIERSZULSKA

as it is easily convertible to the propositional function form as in Frege's notation, e.g. instead of saying that 'x is P' we can always write 'Px'. Meinong himself speaks about unsaturated objects of higher order which are in need of being saturated by the right kind of entity. The special feature of objectives is that their 'arguments' (entities by which they are saturated) are external and do not belong to the objective. This justifies the interpretation of objectives as purely abstract entities, like Frege's senses of sentences. Such saturation in the case of an objective results in its factuality. Being factual implies being true if grasped in an act of judgment. So we have a three-levelled theory after all: sentences express objectives as meanings, and their 'sentential functions' can be said to yield the appropriate truth value as a reflection of factuality or non-factuality of objectives. For this reason, I think that Frege's and Meinong's conceptions are extremely close, since, on one hand, auxiliary objects are indeed objective sense-entities resembling the senses proposed by Frege, while on the other hand, objectives are better explained as function-like entities than as states of affairs. ♦ Meinong on Sentences and Objectives Currently it is often observed that Meinong's conception is in fact one of the early attempts to develop a theory of meaning. His theory of objectives is based upon and supported by investigations concerning the nature of linguistic meaning. In particular, the results obtained in Meinong's famous On Assumptions are derived from the analysis of the meanings of sentences and the interdependencies between meanings of sentences. An expression of such a view can be found in the following quotation from a paper by Evelyn Dölling. Although he understood the theory of Annahmen as a contribution to the further development of the theory of objects (Gegenstandstheorie) conceived by him, he holds the analysis of language to be one of the crucial presuppositions to drafting this theory. Language forms the framework within which different types of Annahmen can be studied. Starting from Martinak's Psychologische Untersuchungen zur Bedeutungslehre he is interested in language, above all, as a system of signs. Presupposing this understanding of language Meinong inquires

MEINONG ON MEANING AND TRUTH

63

about the relationship of sign and meaning (Bedeutung) in general, and especially that of word and meaning, as well as that of sentence and meaning.62

It is emphasised in Dölling's paper that Meinong was especially interested in other early attempts to develop a theory of meaning. His theory of objectives belongs to this line of investigations. In On Assumptions it is explained that sentences express judgments and assumptions, but their meanings are objectives. Of course, viewed from the perspective of the objects of consciousness, objectives are primarily what is intended by judgments and assumptions, and in principle they do not have to be expressed in language. We know such objects as significations, and right away a conjecture suggests itself: just as the property of signifying objects normally belongs to words, there will be linguistic forms that have objectives as their significations.63

Meinong observes that the situation with the meanings of linguistic units is not obvious in various contexts. Therefore, one cannot conclude that an objective is always expressed as the meaning of a sentence. It is often so that a single word has an objective as its meaning, rather than an objectum. In the end, he assumes that it is always the case that a sentence is the specific linguistic unit the meaning of which is constituted by an objective, except that sometimes we deal with equivalents of sentences. The signification of a word does not always consist in an objectum; often enough it consists in an objective. But if, further, words in the latter case make their appearance as the equivalents of sentences – as we have seen them do – then in the event that such a word has a signification, the equivalent sentence can scarcely be denied signification; and thus we come back to the theme of sentential signification put aside at the end of the preceding chapter. Sentences, too, we can now say, have signification in the narrower sense that we precised on the relevant occasion; moreover, these sentential significations are objectives.64

Apart from sentences, also some expressions of non-sentential form may express acts of judgment or assumption, and therefore they may have objectives as their meanings. 62

Evelyn Dölling, "On Alexius Meinong's Theory of Signs" p. 206. A. Meinong, On Assumptions, p. 45. 64 A. Meinong, On Assumptions, p. 48. 63

64

ANNA SIERSZULSKA As to signification, the matter works out quite simply along the lines of what has been said. All sentences signify objectives; this we know. But does this sort of signification, objective-signification, belong only to sentences? Obviously not, as anything that expresses an assumption will thus far have to signify an objective, too.65

The meaning of a sentence or an equivalent of a sentence is composed, according to Meinong, of the meanings of words united in a certain way. We may notice that the meaning of a sentence, and so the structure of an objective, is compositional. In the end, what is obtained is an object of higher order based upon the meanings of the words which form a given sentence. The meanings of these words are regarded as the 'inferiora' of the objective. To be sure, there is normally never a lack of significations where there is a sentence. These significations seem to consist in the significations of words, the words united into a complex in the sentence, and to consist in the objects of higher order that are based on these word-significations and which have them as their inferiora.66

On Meinong's theory, the abstract auxiliary meanings of words can be always identified with existent particular objects, if such there are. This might lead to a conclusion that objectives have real objects as their inferiora – that they are complexes of real objects, instead of being merely 'about' particular objects. Such a view cannot be right, because also nonexistent and abstract general objects play the role of the meanings of linguistic expressions, and so they become the inferiora of objectives. The basic pattern of signification for sentences in Meinong's theory: A sentence –––– expresses ––– a judgment ––– intends ––– an objective (or its equivalent) (or assumption) Linguistic unit –– (an act with its content) –– object of the act = meaning If we try to represent the structure of an act of judgment, we can see that in these cases when a factual objective is identified with the intended one, the apprehended complex still stands outside of the act itself and it is not 65 66

Alexius Meinong, On Assumptions, p. 261. Alexius Meinong, On Assumptions, p. 29.

MEINONG ON MEANING AND TRUTH

65

signified by the sentence. There are also many cases when no factual objective can be identified with the intended one, or when the objective intended is true but involves non-subsisting entities. In such cases the sentence still has meaning, even though no complex or state of affairs is apprehended. It is obvious, therefore, that objectives are not to be identified with complexes or states of affairs, and so the meanings of sentences are not complexes or states of affairs. Meinong's structure of an act of judgment:

Act

(content)

objective as intended

[objective as factual]

[apprehended complex]

[objective as non-factual]

[∅]

(content is psychic)

Objectives depend logically, but not existentially upon their objects of lower order. This means that objectives merely presuppose their objects of lower order as entities belonging to the overall domain of pure objects. It is discussed at more length in Ch. 2 why objectives are not identified with the complexes which they serve to apprehend. The meanings of sentences are not complexes in reality, such as would be the counterpart of the expression 'the glass table'. They are propositions, like 'that the table is made of glass". On Meinong's theory, the latter cannot be understood as a state of affairs, because states of affairs are usually considered to consist of real objects and universal properties as their constituents. Objectives not only do not consist of real objects, but there are also no universal properties according to Meinong. There are only general incomplete objects by means of which we can refer to particular instances of a property. Meinong calls them meaning-objects (Begriffsgegenstände). A meaning-object (Begriffsgegenstand) is the nearest object of intention expressed by a given word. All linguistic meanings are in fact Begriffsgegenstände. They can be used to refer to something abstract, like the 'triangle' or 'number', something existing, like a 'mountain', and also to properties and relations. But in themselves, such meaning-objects are neither particular nor existent. Real objects and particular properties are

66

ANNA SIERSZULSKA

existing entities and the constituents of complexes. However, they can never be constituents of abstract and timeless objectives. Can we say, therefore, that objectives consist of concepts? It is possible to say, I guess, that Meinong's meaning-objects have a conceptual character, since an objective – as intended – is a sense-entity in itself that consists only of sense-entities: auxiliary meaning-objects that convey the representational contents used to intend ultimate objects. These meaning-objects are mainly general objects which are used to intend particulars. For example, 'green' is a general object that may stay general in some contexts, but it may also be used to intend particular moments of greenness. This is the way properties appear in intended objectives, in contrast to complexes, in which they always appear as particulars. I would not like to say, however, that objectives consist of concepts. I think that it is better to treat intended objectives as simply consisting of auxiliary meaning-objects, with a direct referential access to particular entities in the world. As to factual objectives, the structure of the objective is identical to the intended one, but external existing entities are inserted into its structure instead of mere meaning-objects. So if the apple is green, its particular greenness will be involved in the factual objective (as intended by means of the general meaning-object 'green'). But the particular existing greenness is not really a constituent of the factual objective. It is only inserted into its structure as an external object appropriate for it to obtain. Far from treating objectives as pieces of reality, Meinong initially has the problem whether objectives are objects of any kind at all, even if they are no ordinary objects of mental representation. ...can we simply say that sentences are word-complexes characterized by the fact that they signify objectives? But that would not be conformable to the strict concept of signification just now brought back into consideration. True, we said that insofar as an experience presents its object to thought, signification belongs to the expression of that experience – the object then constituting the signification. Now, however, we know that this in no wise holds true of every sentence. There is indeed an objective over against every sentence, but from what we have so far seen of the facts, the objective is a presented one only when it is judged about, and not when it is simply judged.67

67

Alexius Meinong, On Assumptions, p. 48.

MEINONG ON MEANING AND TRUTH

67

An objective which is judged does not appear to us as an object, not in any similar way as the meaning of a term appears to be an object. This is because an objective is not an ordinary object that could be directly represented in the mind, it is an object of higher order that carries in itself certain complexity and, what is more, an abstract one, i.e. a sense-object. (Which is also the reason why the reference of sentences to propositions does not resemble in the least the singular reference of terms to objects – propositions are not objects in the ordinary sense. See also the analysis of sentence reference in Ch. 9.) An objective 'becomes' an object to us when we concentrate on its truth value by means of a higher order objective. Meinong's judged-about objective may be regarded as either a state of affairs which is or is not the case, or as an abstract proposition which is either true or false. Yet it would not be natural to predicate truth about complexes or states of affairs, while it is natural to predicate truth about objectives. Contrary to the standard approach in Meinongian studies, which favours the option with states of affairs, it is proposed here to regard as basic the interpretation of an objective as an abstract proposition. For the interpretation of Meinong's objectives as states of affairs certain similarities with Husserl's conception are responsible. Technically an objective is the intentional object of an act of judgment or assumption. The corresponding entity in Husserl's conception is a state of affairs. This is why it seems obvious to identify an objective with a state of affairs. The full and entire object corresponding to the whole judgement is the state of affairs judged: the same state of affairs is presented in a mere presentation, wished in a wish, asked after in a question, doubted in a doubt etc.68

However, as we remember from the comparison of Husserl's and Meinong's theory of meaning for expressions, there is a certain difference in their views, which also finds its specific expression in the signification pattern for sentences, even in Husserl's early conception from Logical Investigations. Let us have a closer look at Husserl's structure of an act of judgment and his pattern of signification for sentences in Logical Investigations, where Husserl's views might be considered to be closest to Meinong's. 68

Edmund Husserl, Logical Investigations, p. 579.

68

ANNA SIERSZULSKA What plays the part of object to judgement and opinion we call the state of affairs judged: we distinguish this in reflex knowledge from the judging itself, the act in which this or that appears thus or thus, just as in the case of perception we distinguish the perceived object from the perception as act.69

Thus in an act of judgment we have to distinguish the act itself and the matter of the act, which presents a state of affairs as its immediate intentional object (as intended in the act). This intentional state of affairs if intended correctly may be identified as the same state of affairs that is the case in reality. Of course, several different intentional states of affairs may be identified with the same state of affairs in reality as the different ways it is intended. But the intentional state of affairs as presented by the act matter is not the meaning element in Husserl's conception. The act matter – the overall presenting content of the act – involves an instantiation of ideal content, the ideal sense of an act of judgment, which is the same in all acts such that they present an exactly identical state of affairs as immediately intended. This ideal content is the proper meaning element. Husserl's structure of an act of judgment: Act – act matter (ideal content)

[state of affairs as intended]

≈

[state of affairs in the world]

The basic pattern of signification for sentences in Husserl's theory: Sentence –– expresses –– ideal content –– intends –– state of affairs Meaning = the ideal species of the act matter, not the object of the act Comparing Meinong's and Husserl's patterns of signification for sentences, we may be inclined to suppose that since Meinong's content is merely psychic and the sentence signifies the object of intention directly, i. e. the objective, then the meaning of the sentence is a state of affairs. Yet there is a difference in the way the object of intention of an act of judgment is understood in both conceptions. Husserl's state of affairs is an ideal entity which actually consists of the objects that are its components. As such it can be identified with a fragment of reality, a state of affairs in the world. 69

Edmund Husserl, Logical Investigations, p. 611.

MEINONG ON MEANING AND TRUTH

69

The presentation of a state of affairs is the basis for a judgment, which possesses its proper assertive character only if it reaches its fulfilment and otherwise the act is just a presentation of a state of affairs. Instead of Meinong's assumptions, Husserl has mere presentations of states of affairs. Meinong's objectives are integrating structures that can obtain between objects (see Ch.2), but can never contain any objects. They are simple entities and do not possess components. So when Meinong says that an objective intended can be identified with the factual objective that obtains in reality, he does not mean that it can be identified with a state of affairs in the world. Whereas according to Husserl's idea of act fulfilment, and so of truth, the object of intention of an act of judgment can be identified with a state of affairs in the world as being given in the strictest sense. The object presented is literally a certain fragment of reality. But the epistemologically pregnant sense of self-evidence is exclusively concerned with this last unsurpassable goal, the act of this most perfect synthesis of fulfilment, which gives to an intention, e.g. the intention of judgement, the absolute fullness of content, the fullness of the object itself. The object is not merely meant, but in the strictest sense given, and given as it is meant, and made one with our meaningreference. It does not matter, for the rest, whether one is dealing with an individual or a universal object, with an object in the narrower sense or with a state of affairs, the correlate of an identifying or distinguishing synthesis. (…) If we at first keep to the notion of truth just suggested, truth as the correlate of an identifying act is a state of affairs (Sachverhalf), as the correlate of a coincident identity it is an identity, the full agreement of what is meant with what is given as such.70

It may be quite misleading to interpret Meinong's conception through Husserl's , because the intentional state of affairs in Husserl's identity scheme for true judgments turns out to be an actual state of affairs. Husserl's truth: the state of affairs as intended = the state of affairs in reality One might conclude, by analogy, that also Meinong's objective, which is the intentional object of the judgment and the meaning of the related sentence, is a state of affairs as it is in the world. One might think that for 70

Edmund Husserl, Logical Investigations, p. 765.

70

ANNA SIERSZULSKA

Meinong sentences would signify directly fragments of the world. But they do not. For Husserl, the act of judgment, if true, intends a part of reality consisting of real objects, a state of affairs. And indeed, later Husserl identifies the meaning of a sentence with the intended state of affairs. The closest Meinong comes to the idea of such an entity that consists of real objects in the world is when he speaks about complexes. But an objective can never be identified with a complex. It is an abstract intelligible structure that obtains, and therefore an objective sense-entity, different from a complex or a state of affairs. ♦ Objectives in the role of propositions In Meinong's theory, truth is not ascribed to 'objecta', such as complexes, but to sense-entities of an abstract character, that is, to objectives. Objectives belong to objects of higher order, and they could be best characterized as logical structures which may obtain in reality. Meinong's true objectives subsist and they constitute knowledge. This provides him with one more argument against the 'prejudice in favour of the actual' attitude, because objectives are nonexistent entities without which, as he seems to suppose, knowledge would be impossible. If I say, 'It is true that the antipodes exist,' truth is ascribed not to the antipodes, but to the Objective 'that the antipodes exist.' But this existence of the antipodes is a fact which, as everyone sees immediately, can very well have a subsistent status, but cannot be still another existent entity in its own turn, as it were. This holds, likewise, for all other Objectives, so that every cognitive act which has an Objective as its object represents thereby a case of knowing something which does not exist.71

Objectives are not entities belonging to reality. How to explain, therefore, that if they are true, Meinong calls them facts? Meinong is not the only philosopher to endorse the conviction that a true proposition is a fact. And it is often so that each theory making a claim such as this one, does it for different reasons. The most common associations would be with ontological theories of truth, idealistic identity theories, deflationistic theories, or antirealistic ones. But it will be argued that Meinong's theory 71

A. Meinong, "The Theory of Objects", p. 80.

MEINONG ON MEANING AND TRUTH

71

does not belong to any of these. It is a realistic identity theory of truth, based upon the identity of logical structure holding between Meinong's propositions and fragments of reality. Objectives, like propositions, are meaning entities and they are related to language, in the sense that they can be expressed in language. Whenever they are expressed, they function as meanings of declarative sentences. Objectives constitute knowledge and they can be subject to deductive operations. Objectives can appear as premises in reasonings based on inference and help to confirm the truth of other objectives. There is really no such function which would belong to propositions and which could not be performed by objectives. For this reason, it is not surprising that Meinong identifies his objectives with Bolzano's propositions, it may be even supposed that he was influenced by Bolzano's theory. In opposition to Husserl, who criticizes Bolzano's notion of a proposition (being rather in favour of judgment species or types), Meinong likes this notion. However, he does not adopt Bolzano's terminology, which was not generally recognized at the time, Bolzano being a neglected figure in the philosophical tradition. The main thing which he mentions as weighing against Bolzano's term, is that it puts the linguistic aspect of a proposition (Satz an sich) in the foreground, while Meinong wants to stress more strongly that an objective is neither bound to any particular sentence, nor to any particular language; that it is an objectively independent entity. Concerning Bolzano's 'proposition in itself', it seems, on the whole, beyond doubt that this above all approximates the objective very closely, if it is not out-and-out identical with the objective. That is corroborated in the latest accounts of Bolzano's theory. Had he received the share of attention and recognition that he so deserved, then there could have arisen a tradition in favour of the word 'proposition' for what I mean by 'objective,' in which case I should be among those not wishing to oppose this tradition. Such recognition, however, seems only just now on the point of breaking through.72

Bolzano's strategy of beginning investigations from the linguistic level rather than from the level of objects of consciousness is admittedly less old fashioned than Meinong's reverse way of proceeding. But proceeding in this reverse way, Meinong obtains a uniform theory of all objects of 72

A. Meinong, On Assumptions, p. 75.

72

ANNA SIERSZULSKA

cognition and a theory of meaning at the same time. Objectives, like all the other meaning entities, belong to his spectrum of objects beyond being, regardless of whether they are true or false. Aussersein is all-inclusive and undiscriminating with respect to whatever can become an object of cognition and whatever can be meant in language. So negative and false objectives have an entirely independent status as objects in their own right. They are certainly not less mind-independent than Bolzano's propositions. Another proposition-like entity similar to an objective is Frege's 'thought', the main difference lying in the fact that Fregean thoughts belong to the ideal sphere of the third world, while Meinong's objectives are not ideal except for the true ones. It is not necessary, for the fictional and, especially, for the false objectives, to be ideal in order to retain their objective status – being objects of Aussersein suffices for this purpose. Apart from this single difference, Meinong's objectives resemble Fregean senses of sentences very closely. On the basis of what has been said about objectives so far, we may assume that objectives are abstract objects of higher order and that they are what we would call 'propositions' in a more familiar terminology. If we compare the properties traditionally ascribed to propositions and those of objectives, we will find no striking differences. Both propositions and objectives are subject-independent, in the sense that they can be potentially thought by different subjects. They both possess a mind-independent way of being. Both objectives and propositions can be true without being actually expressed in any language. They are also both independent of any particular language and can be expressed in many different languages. But propositions are the abstract contents of declarative sentences, while objectives are not 'contents,' either of sentences or mental acts, they can be regarded as the meaning-objects that sentences and mental acts of judgment can potentially intend. This solution is employed to avoid the psychologistic connotations related to speaking about 'contents' of mental acts. Yet, the result is not remote from what we understand by a proposition, and, in any case, closer to this notion than to the notion of a complex or a state of affairs, as these are usually understood.

MEINONG ON MEANING AND TRUTH

73

♦ The function-like nature of objectives It will be argued in more detail now that Meinong's notion of an objective resembles propositions possessing a function-like character, such as Bolzano's Satz an sich and Frege's prototype of a propositional function. The truth of such propositions results from their 'saturation' with the right kind of entity, whereupon the factuality of the proposition follows. This happens in the same way as a function yields the desired value only if provided with the proper kind of argument and not some other one. Obviously, there are differences and the first thing that may be mentioned is that Meinong's sense entities are treated as objects, which might discourage attempts to identify them with Frege's senses. We have seen that in the case of Meinong's auxiliary objects the theory of meaning obtained is practically indistinguishable from Frege's theory. In the case of objectives, their reification is merely a superficial, anti-psychologistic device as well, for otherwise they retain all features of abstract sense entities. In a mental act, Meinong distinguishes the act itself, the content belonging to the act and specific to it, the object immediately presented by the content of the act, i.e. the internal or intentional object (an objective in the case of judgments and assumptions), and the ultimate object of reference, which may be identical to the immediate object. The content of the act is psychologically dependent on the mind of the subject. It is the specific and subjective way in which the immediate object of the act is presented.73 Therefore, the content of a judgment cannot play the role of a truth bearer. There is no way to think about it as something abstract to be grasped in different acts, like a proposition or a Fregean thought. Instead, Meinong ascribes this role to the immediate object of a judgment, namely to the objective intended by its content. An objective is not an abstract content, but it is a sense-entity and for this reason it can be considered to be very close to Frege's Gedanke or Bolzano's Satz an sich. The content of the act presents the immediate object and we realize that this is its essential 73

On contents as psychologically dependent see Ch.1, Über Annahmen, p. 277, Über Möglichkeit, p. 163, note 3. In particular he claims that what is sometimes called the 'logical content' is not a content for him, but the immediate object of the act. His 'content' is always 'psychological content'.

74

ANNA SIERSZULSKA

role, but we do not "think contents", so to say, – we "think about" objects. In reflection, contents cannot be focused upon, they are fleeting psychic phenomena for Meinong. We are only able to concentrate upon the objects that the contents present as their immediate objects. The problems with the ontological status of contents, as well as with their subjective character, often objected to by the advocates of the extensional semantic tradition, are circumnavigated in this way. The distinction between the content and the nearer and further objects of an act is important for redefining the notions inherited from the Brentanian tradition and it also allows one to avoid assuming psychologically dependent truth bearers (such as the mental contents of judgments would be). As a rule, the objective given by means of the content of a judgment or assumption, may be considered to be the meaning conveyed by the act and expressed by a sentence of a given language, and this objective plays the role of the proper truth bearer in the process of cognition. The role of the truth maker can be ascribed to the factual objective, the structure of which obtains in reality. Since a factual objective, being identical to the true objective, is a sense-entity and does not consist of objects, it can only be a truth maker as the 'integrating factor' in the complex which involves the logical structure of this factual objective. In this way, facts belong to reality as the ways it is objectively ordered, but not as pieces of reality. We may sum up what has been said in Ch.2 about the nature of objectives, in comparison to other objects of higher order, as follows: a. objectives are classified as objects of higher order, but of a special kind – their dependence upon other objects has the character of logical priority, there is no dependence in any existential sense b. unlike other objects of higher order (which are complexes and relations), objectives do not belong to 'objecta', they can only be called objects in a very broad sense, and they are not particulars – they are abstract entities – they concern particular objects, but they do not contain them c. in contrast to objecta of higher order, there is no distinction between real and ideal objectives, because they are not particulars d. objectives as semantic objects play two roles: 1.the role of apprehending complexes (objectives of so-being), and also the

MEINONG ON MEANING AND TRUTH

75

existence of all objecta (objectives of being); 2. the role of constituting the meanings of sentences e. there are no predetermined objectives, their inferiora can be picked out arbitrarily from the continuum of reality (there are also no predetermined complexes) The most explicit statement which Meinong makes about the nature of an objective, or rather what kind of entity it is that he calls an objective, is the following: Generally speaking, then, where there is a complex, there is also an objective as integrating factor in it, and one who wants to apprehend the complex cannot do it otherwise than by apprehending the objective, too.74

An objective is described as the integrating factor of a complex. Of course, what is meant in this context is not just that an objective is an object of intention, but that it is a kind of structure, which obtains in reality between the components of a complex. An objective is not the complex itself, for the components of the complex are not components of the objective. A complex is what is apprehended, it consists of its basic objects and the relation, but what makes a unified structure of it is the objective, the integrating logical structure of the complex. The English translation of this fragment is quite faithful, as in the original text an objective is called the 'integrierendes Moment' of a complex. Allgemein also: wo ein Komplex vorliegt, dort liegt auch ein Objectiv als integrierendes Moment daran vor, und wer den Komplex erfassen will, kann das nicht anders, als indem er auch das Objectiv erfa t.75

On the basis of the above, we may assume that an objective has the character of a logical structure or a certain coordination of objects. In this sense, it may be permissible to regard it as a function-like entity, a prototype of a sentential function, in opposition to a complex-like objectum, or a state of affairs. There are several features of objectives which make this interpretation legitimate. In the first place, the subordinate objects stand to the objective 74 75

A.Meinong, On Assumptions, p. 201. A.Meinong, Über Annahmen, p. 280.

76

ANNA SIERSZULSKA

as arguments to a function: they do not belong to it. And therefore we can have true (and subsisting) objectives concerning impossible entities. From this we may further infer that there is no existential dependence between an objective and its inferiora, for it would be difficult to imagine a subsistent state of affairs depending upon impossible entities. But the inferiora do not even have to subsist: the round square is surely different from the oval triangle, though none of the objects here functioning as inferiora subsists. One may therefore be inclined to surmise that being does not matter in this respect, but only being thus-and-so. This means that if there is to be any meaningful talk about difference at all, we must presuppose beings which are thusand-so with a 'thus-and-so' such that 'difference' can be applied to them.76

It is also questionable whether there can be any negative states of affairs, but this objection does not apply to objectives. On Meinong's conception, negation may be attached to an objective either externally or internally, and transposition of the negation is allowed as well. This is a feature typical for abstract propositions. Above all, it is not true that there are no negative objectives, but only positive ones which are rejected, as some commentators have claimed (cf. Ch.7). Meinong often mentions negative objectives as objects of negative judgments and assumptions. As early as the opening statements about assumptions, we had occasion to insist on their determinacy within the antithesis of yes and no as a characteristic factor, a factor that separates assumptions from representations and connects them with judgments. It is now easy to see that this factor may also be identified by reference to the character of the objectives apprehended by means of assumptions: like judgments, assumptions are affirmative or negative according to whether they apprehend positive or negative objectives.77

In addition, he openly emphasizes in the notes to the second edition of On Assumptions, that it is a mistake to suppose that there are no negative objectives according to his theory.

76 77

A.Meinong, On Emotional Presentation, p. 61. A.Meinong, On Assumptions, p. 98.

MEINONG ON MEANING AND TRUTH

77

In particular, the contention that I do not allow for negative objectives is erroneous even in connection with the first edition. It was just that I had by no means yet taken full recourse to the objective in that edition.78

This would support a propositional interpretation of objectives. And so would the way objectives are treated as premises in formal inferences. Meinong is entirely in favour of using a symbolic language and other formal devices in such contexts. He even maintains that a body of knowledge should be formed by a logically consistent system of objectives. Such an approach to objectives that allows, as a matter of course, for diverse logical operations upon them, confirms their status as abstract sense entities. For one thing, the capacity of normal intelligence in this regard oughtn't to be given too little credit. For another, it will have to be taken into consideration that aids of the most diverse sorts partly reduce and partly lighten the work of thinking. Such aids include symbols, formulas, and operations on same; they include auxiliary and mediating notions which on occasion enable us to at least indirectly bring in judgments that no longer have been able to remain present to the mind. An inference of the form "A is B, B is C, C is D..., M is N; hence A is N" might represent an example of the simplest sort; here certainly not all of the premises can remain present to the mind at the same time.79

Finally, if we pay attention to the form of some of Meinong's argumentation, we will notice that he treats objectives as incomplete structures, which can take different arguments. This is quite apparent in situations where he substitutes one of an objective's 'arguments' for another. In such cases, the unchanged part of the objective plays the role of a propositional function with a slot to be filled by a variable argument. From there, we immediately come closer to our above hypothetical judgmentformula if we temporarily put B in the antecedent in the place of A, which gives us "If B exists, then B exists"; for a proposition like "An existing B exists" is hardly very different from this. And if I take A back into the antecedent, the tautological character disappears again.80

78

A.Meinong, On Assumptions, p. 303. A.Meinong, On Assumptions, p. 135. 80 A.Meinong, On Assumptions, p. 148. 79

78

ANNA SIERSZULSKA

This is hardly a way to speak about states of affairs, to which Meinong's objectives are often taken to be quite similar. States of affairs could not be truth bearers in the way objectives are. Meinong points out that the application of the latter term is much broader, since there are negative and unfactual objectives, while the idea of negative, or unfactual states of affairs is puzzling. As objectives are truth bearers and falsehood bearers as well, there must be also unfactual and impossible objectives. The former term did not appear to me to be suitable as a substitute, because its natural domain of application is too narrow. It is difficult enough even to call this a state of affairs, that it is not snowing now; but the applicability of the term to unfactual objectives seems to break down altogether.81

Meinong understands states of affairs in the way which is typical for his time, namely as fragments of reality, actually containing objects. On a different understanding of this notion, we often speak about possible and impossible states of affairs, without implying thereby that there exist any objects with which they are concerned. This notion of a state of affairs is already close enough to a mere proposition, and perhaps it could be accepted as a counterpart of an objective. But two further objections rule this out: first, even on this view there are no negative states of affairs, while it is plainly acknowledged by Meinong that there are negative objectives. Secondly, the idea of a factual state of affairs always implies that it literally consists of objects. Meinong remarks that the idea of a state of affairs as the 'objectivity' which is intended, would enclose in it also the objects proper that exist in reality.82 Basically, it seems that states of affairs can belong to reality like its constituting parts, while objectives naturally cannot. It has been observed by several commentators of Meinong's philosophy that objectives cannot contain objects as their constituents. In this respect they differ from states of affairs and complexes.83 The 81

A.Meinong, On Assumptions, p. 76. Meinong states an objection concerning the comparison between his objectives and states of affairs in Über Möglichkeit, p. 157. 83 It has been observed by G.Bergmann (Realism…, p. 349), M-L.Schubert Kalsi (Introduction to Meinong's Objects of Higher Order, p. 21), p. Simons ("On What There Isn't", p. 171). Compare also the discussion below in this chapter. 82

MEINONG ON MEANING AND TRUTH

79

interpretation of objectives as complexes was adopted after Russell and it is not really correct.84 What is understood under the name 'objective', Meinong claims, is much closer, if not almost identical to Bolzano's 'proposition in itself' (Satz an sich). The only difference that he himself can see between these two terms, is that propositions are always related to sentences of a language, which is not necessarily the case with objectives (Meinong thought that as the term 'Satz an sich' indicates, Bolzano's propositions were necessarily language related). Objectives may be only intended, without being expressed in a language, as Meinong says. However, it is possible to express objectives in a language, in which case they become the meanings of sentences.85 It is an entirely different issue, though, whether objectives can also carry some extra-linguistic sense, or whether we always think using the conceptual framework of a certain language. A 'proposition' is an objective that is apprehended and perhaps also expressed; it is at least present and formulated in words, as we might say. This factor is even more important than truth or factuality – propositions can also be false. On this interpretation, according to which we are dealing with the formulated objective, the theory of objectives might quite well avail itself of the term 'proposition' in its extralinguistic employment, as is indeed done in science and everyday life.86

There are many similarities between Meinong's and Bolzano's conceptions of a proposition. The most important one is that the propositions in both conceptions are non-subjective entities. Another thing which may be interesting to notice is that, contrary to what Meinong thought and to the connotations of the expression Satz an sich, Bolzano admits that there may be propositions which are not and have never been expressed in language. I should like to note that there are also propositions which are not presented in words but which somebody merely thinks, and these I call mental propositions.87

84

More about the relation between an objective and Russell's complex is to be found in Part IV, in the account of Meinong-Russell dispute. 85 Cf. On Assumptions, p 25 and Über Möglichkeit, p. 198. 86 A.Meinong, On Assumptions, p. 75-6. 87 Bernard Bolzano, Theory of Science, p. 20.

80

ANNA SIERSZULSKA

Objectives are primarily regarded as thinkable and only secondarily as expressible, but the linguistic expression of objectives is very important also in Meinong's conception. In fact, on any conception of propositions those propositions that have never been expressed in language need to be included, if they are potentially expressible. This is acknowledged by Bolzano when he states that there are truths in themselves which have never been thought or spoken as yet. Instead of saying, like Meinong, that such truths possess subsistence, Bolzano says that they are always present in the mind of God. I shall mean by a truth in itself any proposition which states something as it is, where I leave it undetermined whether or not this proposition has in fact been thought or spoken by anybody. It follows indeed from God's omniscience that each truth is known to him and is continually represented in his understanding, even if no other being is acquainted with it or thinks it.88

The mind of God does not contain only truths, but all ideas and false propositions as well. In this way, the mind of God is a counterpart of Meinong's Aussersein, because from being present in God's mind, it does not follow that false propositions have to obtain. In God's infinite understanding every true proposition is present as an actual judgment. False propositions are also present in God's understanding, not, however, as judgments, but merely as ideas of objects about which he judges.89

Bolzano's proposition must be either true or false, but need not be judged. He makes a difference between judged and merely thought propositions, parallel to Meinong's judged and assumed objectives. Every judgment contains a proposition which is either true or false. (...) There is an essential difference between a judgment and the mere thinking or representing of a proposition.90

If Meinong's true judgments are judged with evidence, the intensity of evidence determines the degree of probability which we can ascribe to a 88

B.Bolzano, Theory of Science, p. 32. B.Bolzano, Theory of Science, p. 43. 90 B.Bolzano, Theory of Science, p. 43. 89

MEINONG ON MEANING AND TRUTH

81

given judgment. In Bolzano's conception we find a very similar idea, under the name of the confidence of a judgment. On this basis, we judge a proposition with greater confidence if it appears to be more probable than its counterpart. Each of our judgments, depending on the considerations that precede it, is carried out with greater or less force. I call this the confidence of the judgment. If a proposition appears just as probable as its contradictory then we can judge neither that it is true, nor that it is false, rather, we doubt.91

There is also a certain analogy with the way Bolzano treats the question of truth and correspondence, which is related at the same time to perhaps the strangest idea in Meinong's conception, namely that judgments of so-being always have an a priori character. Something similar is the case with Bolzano's theory and a clear explanation is offered. You have probably been told that truth (i.e. transcendental truth) consists in a certain agreement between our ideas and the objects to which they refer, but this expression is not quite correct. We would rather say that our judgments are true if our idea of a certain object is connected with the idea of an attribute which this object really has. Hence if you want to persuade yourself of the truth of your judgments it is not necessary to compare your ideas with their objects (...) For if one of your ideas does not agree with a certain object then it is, for that very reason, not an idea of that object (...) this agreement between an idea and one or several objects to which it refers or whose idea it is should not be thought to be a kind of resemblance. The idea 'something' and the things to which it refers (i.e. all the things that there are) have no similarity whatsoever with each other. For it is not by virtue of similarity that an idea represents this or that or no object at all.92

Thus Bolzano explains that it is not necessary to compare our judgments with the objects themselves for resemblance, because we judge on the ground of the ideas of objects as they are represented in our mind. So in order for a judgment of so-being to be true, the idea of a certain object should be connected correctly with the idea of an attribute possessed by this object. A judgment at this level is indeed done without any direct recourse to experience, but it is related to experience indirectly via the notion of representation. The same representational account of thinking is 91 92

B.Bolzano, Theory of Science, p. 44. B.Bolzano, Theory of Science, p. 50.

82

ANNA SIERSZULSKA

responsible for Meinong's classification of a whole group of judgments of so-being as a priori, which we would rather regard as empirical. However, in both conceptions we have the absolute notion of truth as unquestionably basic. Bolzano says that if a proposition changes its truth value, it is only because, consciously or not, some part of it has been changed, e.g. a part defining time or place or object to which it applies. It cannot be denied that every proposition is either true or false and never changes: either it is true forever, or false forever, except if we change some part of it, and hence consider no longer the same but some other proposition.93

The truth of objectives is dependent upon the interpretation of demonstrative pronouns and time specifications in the same way. Analysing such cases, Meinong explains that the truth of each objective is eternal94, because the circumstances to which the objective applies always remain unchanged for a given objective. And if they change, we have another objective then, which applies to different circumstances. Therefore, all truths are eternal, not just those that concern what does not exist in time. Objectives resemble Bolzano's propositions in this, that truths concerning spatiotemporal objects are eternally true. They do not cease to be true when their objects disappear from the realm of existents. It has been indicated above that Meinong's objective is a function-like entity, and the same feature is characteristic of Bolzano's proposition, but in a more explicit manner. The objects in Meinong's objective are related to it in a way similar to how arguments are related to a function, but no extensive use is made of the idea of exchangeability of these arguments, they are not treated as variables, apart from a few contexts. Bolzano's notion of satisfiability of a proposition is an application of this idea. We can see that his proposition, open for substitution, turns into a propositional function, which is satisfied by some arguments from a certain domain and not satisfied by others. I therefore want to give a special name to the concept of the relation of all true propositions to the total of all propositions which can be generated by treating certain ideas in a proposition as variables and replacing them with others 93 94

B.Bolzano, Theory of Science, p. 194. A.Meinong, On Assumptions, p. 53-4.

MEINONG ON MEANING AND TRUTH

83

according to a certain rule. I wish to call it the satisfiability [Gultigkeit] of the proposition.95

As a result, Bolzano obtains some true propositions and some false ones, unless when a given propositional function is universally satisfiable. Meinong uses the concept of a tautology, but not of a universally satisfiable proposition. Still, his objective displays the relevant features which qualify it, together with Bolzano's proposition, as possessing the character of a function. It will be best seen in comparison to Frege's original idea of a function, that there are important similarities between the way an objective and a function are conceived. The first condition for some abstract entity to be called a function is that it is incomplete without its arguments, which are in a sense external in relation to it. The arguments do not belong to the function, but when the function is satisfied they form a complete unity together with it. I am concerned to show that the argument does not belong with the function, but goes together with the function to make up a complete whole; for the function by itself must be called incomplete, in need of supplementation, or 'unsaturated.'96

All Meinong's objects of higher order are said to possess this feature that they are incomplete, or unfinished, without the objects upon which they depend.97 However, only in the case of objectives are the objects presupposed external in relation to the objective and, for this reason, they can be compared to the arguments of a function, for the objective coordinates these objects without containing them. The external objects of an objective are 'taken' as arguments but they remain external, without constituting a whole in any more substantial sense, such as a state of affairs. Meinong's argument against a part-whole relation between the inferiora and the objective itself is that from the being of a factual objective of non-being we cannot infer the being of the subordinate object, as we normally would if the part-whole relation were to obtain.98 95

B.Bolzano, Theory of Science, p. 196. Gottlob Frege, "Function and Concept", in: Peter Geach and Max Black, Translations from the Philosophical Writings of Gottlob Frege, p. 24. 97 A.Meinong, "On Objects of Higher Order", p. 144. 98 A.Meinong, "On Objects of Higher Order", pp. 84-85. 96

84

ANNA SIERSZULSKA

The concept of a function is a mathematical concept, but in the context in which it was introduced by Frege, it was to be understood in a much broader sense, for Frege assumed function-like nature of all concepts, linguistic, logical and mathematical. Any ordinary linguistic unit could be considered to be a kind of function if one part of it were taken to be 'unsaturated' and in need of completion by the other part as its 'argument'. The notions of an 'argument' and of the 'value' of the function were understood very broadly as well. Not only numbers were allowed as the former, and truth values as the latter, but different objects in general could appear in these roles. In Frege's analysis of the function 'the capital of x', he takes 'the German Empire' to be a possible argument, and the value of the function for this argument turns out to be 'Berlin'. In a similar way we can say that the 'arguments' of an objective, e.g.' The dog is red', are coordinated by the objective so that as the value of the 'function,' we obtain the value 'true'. An objective, like a function, is an abstract entity, but the dog and its redness are concrete entities in the world. Statements in general, just like equations or inequalities or expressions in Analysis, can be imagined to be split up into two parts; one complete in itself, and the other in need of supplementation, or 'unsaturated.' Thus, e.g., we split up the sentence 'Caesar conquered Gaul' into 'Caesar' and 'conquered Gaul.' The second part is 'unsaturated' – it contains an empty place; only when this place is filled up with a proper name, or with an expression that replaces a proper name, does a complete sense appear. Here too I give the name 'function' to what this 'unsaturated' part stands for. In this case the argument is Caesar. We see that here we have undertaken to extend [the application of the term] in the other direction, viz. as regards what can occur as an argument. Not merely numbers, but objects in general, are now admissible; and here persons must assuredly be counted as objects. The two truth-values have already been introduced as possible values of a function; we must go further and admit objects without restriction as values of functions. To get an example of this, let us start, e.g., with the expression 'the capital of the German Empire.'

MEINONG ON MEANING AND TRUTH

85

This obviously takes the place of a proper name, and stands for an object, if we now split it up into the parts 'the capital of and 'the German Empire' where I count the [German] genitive form as going with the first part, then this part is 'unsaturated,' whereas the other is complete in itself. So in accordance with what I said before, I call 'the capital of x' the expression of a function. If we take the German Empire as the argument, we get Berlin as the value of the function. When we have thus admitted objects without restriction as arguments and values of functions, the question arises what it is that we are here calling an object. I regard a regular definition as impossible, since we have here something too simple to admit of logical analysis. It is only possible to indicate what is meant. Here I can only say briefly: An object is anything that is not a function, so that an expression for it does not contain any empty place. (...) Value-ranges of functions are objects, whereas functions themselves are not.99

One might perhaps doubt whether an objective in Meinong's theory is what would be accepted as a function in Frege's sense. Yet, we read in Frege that anything which is not a function and does not contain any empty place (even potentially) is an object. So this definition of an object would include all and only Meinong's objecta (for it is no less general than Meinong's own definition) but not objectives. And of course objecta cannot, even potentially, be function-like structures containing empty places, in contrast to objectives. Objectives are the only entities on Meinong's theory which are not classified as 'objecta'. What is not an 'objectum' for Meinong, is an 'objective'. What is not an 'object' for Frege, is a 'function'. The analogy can be even maintained in this respect that just as there are first-level and second-level functions in Frege's conception, there are also first-order and second-order objectives on Meinong's theory. Second-order objectives have other objectives, instead of objecta, as their 'arguments.' This feature is also a strong point in favour of their obviously propositional character.

99

G. Frege, "Function and Concept", p. 31-2.

86

ANNA SIERSZULSKA Now just as functions are fundamentally different from objects, so also functions whose arguments are and must be functions are fundamentally different from functions whose arguments are objects and cannot be anything else. I call the latter first-level, the former second-level, functions.100

Still, there must be a limit to this analogy between a function and an objective, since the mathematical notion requires that the law of correlation between the potential arguments be precisely and explicitly specified. Certainly, such a correlation takes place and can be seen in the structure of an objective, but the nature of this correlation is usually determined by means of the rules of natural language rather than in any more precise manner. The expression 'y is a function of x' has no sense, unless it is completed by mentioning the law of correlation. This is a mistake in the definition. And surely the law, which this definition treats as not being given, is really the main thing.101

Thus, the similarity will remain on the general conceptual level, when we take Frege's original broad concept of a function as the basis for comparison with Meinong's objective. Then we can say that objectives are functions taking arguments from the Meinongian domain of Aussersein and yielding either the value true or the value false. Some very interesting remarks in relation to the issue what sort of entities are Meinong's objectives have been made by Gustav Bergmann. Namely, he explicitly maintains that objectives have the general character of functions. The relation which holds between the objects involved and the objective, is function-like. The objects are the arguments of the function. If you keep in mind that every particular is 'simple' and 'independent' and consider that c , e and (c ; e ) are three particulars, you should find it easy to assay the 'connection' between the first two and the third. If you still hesitate, weigh the expressions Meinong uses. c and e are said to 'determine' (c ; e ). The latter is also called the Superius; the two former its Inferiora. The 'connection' is a function. The Inferiora are its arguments; the Superius its value. As for connections, so for objectives. About all this there is no doubt. Meinong's is a function ontology. The only question left is whether it is an explicit or an implicit 100 101

G.Frege, "Function and Concept", p. 38. G.Frege, "What is a Function?", p. 112.

MEINONG ON MEANING AND TRUTH

87

ontology of this type. Does he, or does he not give ontological status to these functions? The answer is that he doesn't. Meinong's is an implicit function ontology.102

But Bergmann thinks that both the objects and the objective itself have to be particulars. In consequence, he identifies the objective with the value of the function, and not with the function itself. The value of the function would be in this case the apprehended complex, whereas Meinong writes plainly that an objective is not identifiable with the apprehended complex. For Meinong, an objective is the 'connection', the logical structure binding the objects.103 Therefore, it should be identified with the function itself, which by taking certain objects for its arguments makes it possible to intend a complex. The way Meinong characterizes an objective is very similar to what Bergmann understands as a 'nexus'. Even though Bergmann does not realize it. He thinks that a nexus is necessary to combine the constituents of a complex together, because a relation alone is not sufficient to perform this task, being really the product of such a combining. On Bergmann's view, a relation, exactly as a complex, needs a nexus to bind its members. Meinong's objective plays the same role as a nexus, but with a single and important difference: an objective is not a constituent of a complex the way a nexus is considered to be its constituent. In my style, then, the question has only one answer. An ordinary thing is a complex; its qualities are among its constituents; but they do not exhaust it. In the paradigm, , in addition to the qualities which ground red and round, respectively, must have at least one further constituent 'connecting' them, as blue and oval are not connected. This third constituent I call a nexus.104

Further, we can notice that there is a difference in how Bergmann and Meinong understand a fact. For Meinong, an objective is a fact if its structure subsists in reality, that is, if objects are connected in reality in this specific way. For Bergmann, a complex is a fact, in which he follows the Russellian pattern. Neither a nexus, nor its successful binding of the components of a complex, is regarded as a fact on its own, so to say. And 102

Gustav Bergmann, Realism: A Critique of Brentano and Meinong, p. 349. A.Meinong, On Assumptions, p. 202. 104 Gustav Bergmann, Realism..., p. 9. 103

88

ANNA SIERSZULSKA

this is because a nexus is always taken to be another component of the complex, together with the other components, and only all of them bound in a complex can be called a fact. So the complex as a whole is a fact. In the first triple, though not in the second, the nexus actually connects the two qualities into a complex. This complex, or fact and the circumstance that one of the three entities which are its constituents actually connects the other two is one and the same. In the case of the first triple, there is thus a complex whose constituents it exhausts. In the case of the second, there is no such complex.105

According to Bergmann, there are serious consequences of allowing the connecting structure of a complex not to belong to the complex as its proper constituent. In this situation, it is difficult to deal with such a connection within the limits of what is acceptable from an ontological point of view. He calls ontologies making use of unspecified connections of this kind 'function ontologies', since they coordinate entities to others in a way resembling mathematical functions. In function ontologies, as I call them, some entities are, as one says, 'coordinated' to some others, without any connotation whatsoever of the one being 'in' the other, being either a constituent or a part, or a component of it. A function is such a coordination.106

Therefore, he claims that Meinong's is a function ontology. As he actually identifies Meinong's objective with a complex, he thinks that Meinong's explicit statement that objectives have no genuine constituents decides the question. Taking the inferiora of an objective merely as arguments, the value of the function is a simple entity, a 'thing' in his terminology. It is not really a complex object. And so it cannot be a fact. He seems to be partly right in this analysis: an objective, conceived as a logical structure, is not a complex object consisting of other objects. Still, an objective is not the apprehended complex and the value of the sentential function of an objective is not the complex, but 'factuality' or 'non-factuality'. Of course, it is wrong to say that if objectives do not have genuine constituents, Meinong's complexes do not have them either. This, again, Meinong states quite plainly (compare Ch.2). So Meinong's complex is indeed a complex 105 106

Gustav Bergmann, Realism..., p. 10. Gustav Bergmann, Realism..., p. 7.

MEINONG ON MEANING AND TRUTH

89

object of higher order, while his objective is neither a complex nor an object in the same sense. Bergmann is mistaken that what is not a complex object in his sense cannot be a fact, on a completely different interpretation of a fact than his. His world has to be divided into pieces, because he finds it incomprehensible otherwise. His complex ontology is an expression of that. Mine is a complex ontology. Yet my choice, if it be one, does not depend on the adequacy of Russell's or any other proof. What compels me in a manner that leaves me no real choice, lies deeper than argument can reach. I simply do not understand how any coordination among entities can be its own ontological ground. Who or what does the coordinating?107

Another option is to say that it is a fact that a certain structure, which may be called an objective, obtains in reality. Not being an ordinary object, only a sense entity, it need not be grounded ontologically in anything. An insightful analysis of Bergmann's ideas, combined with an expression of a point of view opposed to both Bergmann's and Meinong's treatment of facts, is provided by Erwin Tegtmeier. Tegtmeier admits that Bergmann's classification of Meinong's theory as a function ontology is justified. Instead of a function ontology, where the connection between the arguments and the value of the function is not grounded ontologically, and instead of an ontology of complexes, Tegtmeier proposes an ontology of facts. What are, for the complex ontologist, the constituents of the complex, are, for the function ontologist, the arguments of a function. To each collection of arguments is co-ordinated a function value without any suggestion that the former are in the latter. Hence, a function value can be simple although it has several arguments. Bergmann rejects function ontologies because no ontological ground of the connection between function arguments and function values is offered. (The talk of co-ordination which is usual in set theory suggests a mental operation. However, such an operation could not produce a connection between non-mental objects.)108

It does not seem to be entirely convincing, however, to say that the kind of coordination associated with a function cannot result in any connection 107 108

Gustav Bergmann, Realism..., p. 8. Erwin Tegtmeier, "Meinong's Complexes", p. 90.

90

ANNA SIERSZULSKA

between non-mental entities, for logical structures and linguistic meanings, as sense entities, need not be taken to be merely mental. On the other hand, Tegtmeier is right that no connection between external objects can be involved in a relation between the function arguments and values. But in the case of Meinong's objectives we may assume that we speak about agreement with what can be perceived as obtaining with respect to external objects. So in Meinong's theory of objectives we remain within the limits of what can be intended. A complex subsisting in reality is not what can be directly intended. In opposition to a complex, an objective as involving only the representations of objects, can be intended directly. And this judged objective, if true, can be called factual. Bergmann is both right and mistaken to understand it as a simple entity. It is a complex structure, but a simple entity as an intentional object. Thus, an objective conveys enough complexity to reflect the structure of a complex, not being a complex itself. This allows Meinong's universe not to be divided into complexes or facts. Bergmann's view implies such a division, endowing complexes with full ontological legitimacy. For Bergmann, a complex is the basic entity, and the nexus is only an explanatory device, it never plays any role on its own, apart from being a constituent of a complex. Tegtmaier's claim that Bergmann's nexus is the fact seems to be a metaphor for 'making the complexity of a fact possible'. Bergmann is convinced that Meinong's analysis of complexes fails because he has no adequate notion of a fact. The necessary condition for an adequate notion of a fact is, according to Bergmann, in addition to recognizing a nexus, a clear distinction between things and facts. By 'things' Bergmann does not mean physical objects such as tables, trees and clouds, but the simple entities of which complexes ultimately consist. Bergmann wants to show that only facts are complex. His main argument seems to be that a nexus actually connecting things into a complex is a fact.109

Bergmann's fact must be complex and the nexus is not itself complex, though it is the binding element making complexity possible. The nexus would be close enough to Meinong's objective, if it were indeed the fact on Bergmann's conception. But it is not. The complex as a whole is the fact.

109

Erwin Tegtmeier, "Meinong's Complexes", pp. 89-90.

MEINONG ON MEANING AND TRUTH

91

Tegtmeier agrees that a fact must be complex, but he rejects the idea of a connecting element, like the nexus. I agree with Bergmann that Meinong's objectives are not facts, also with his central thesis that only facts are complex, but not with his analysis of facts. In Bergmann's analysis of facts the nexus which connects the constituents of a fact is essential. This seems to be an original Meinongian conception developed in discussing Locke and Hume but not to be found in them. It was adopted by Russell and Grossmann, too. I have pointed out elsewhere that Bergmann's arguments for the necessity of a nexus are not sound. My view is that no connector is necessary if facts are granted ontological status, since then the fact itself can ground the connection between its constituents.110

Tegtmeier confirms in this way Meinong's conception of objectives as 'integrating' structures. However, he understands facts as states of affairs in reality and does not assume any need for a binding structure as an objective entity to be taken into account. Meinong's objectives are not suitable entities to play the role of such facts. One can only say that a very significant part of Meinong's conception of truth would be lost if an objective were not regarded as a structure shared by human thought and reality. To summarise: in order to explain complexity ontologically, one has to explain facts. Meinong is on the right track. His objectives resemble facts in many respects. However, Meinong's influential conception of complexes as unified by a connector is not adequate, because it presupposes the finished complexity of facts. It is not the connector that creates complexity, but the fact of its holding between the constituents. Thus, the connector cannot ground complexity.111

Meinong would certainly accept that a certain structure, like an objective, is not a fact in itself, but it is a fact that it obtains in reality. For this reason, I do not think that Meinong's conception of objectives presupposes 'the finished complexity of facts.' Such a view is the consequence of treating objectives on the parallel basis with complexes: as objecta, rather than abstract objects possessing the character of functions. That abstract objects of all kinds are taken to belong to the spectrum of objects beyond being is not a reflection of the way reality as such is conceived. I think that, in line 110 111

Erwin Tegtmeier, "Meinong's Complexes", p. 98. Erwin Tegtmeier, "Meinong's Complexes", pp. 99-100.

92

ANNA SIERSZULSKA

with Meinong's theory, it is possible to say that an objective, as a kind of function, is satisfied when, given certain objects for its arguments, it turns out to be factual. Which only assumes that an objective is a proposition, and not any object in reality among a preset realm of facts. A proposition as a sense entity perhaps does not have to possess any more ontological justification other than the status of a 'semantic' object. What functions and Meinong's objectives also have in common is that they cannot be represented in the mind in the way ordinary objects can. They are intelligible, but mentally non-representable structures, which may obtain for objects in the world, but need not obtain. This is how I would like to see Meinong's objective: not as a complex, not as a state of affairs, not as a fragment of reality, but as a function-like proposition, an entity related to human thought as perceiving the logical structure of reality.

MEINONG ON MEANING AND TRUTH

93

CHAPTER 4 MEINONG'S LOGICAL REALISM Meinong's theory is described by Findlay, and then by Lindenfeld, as an expression of the doctrine of logical realism. I think that this is indeed the essence of Meinong's entire approach to the matters of truth and meaning. In order to believe that a realistic view about cognition and knowledge is justified, he has to assume that the structures grasped in mental acts and expressed in language have their counterparts in the structuring of reality. After the age of idealism, when much emphasis was laid upon the role of human rationality, it was necessary to find an explanation of how the mental sphere and the rationality of thought could be related to something in the world, so that we would be able to say that any cognition actually takes place. There has to be an analogy between these two realms, which finds its theoretical realisation in the different variants of logical realism. Lindenfeld counts Meinong among the representatives of this doctrine. The third major form of synthesis was the doctrine of realism, or logical realism, as it is sometimes called. Its principal adherents were Frege, Russell, G.E.Moore, Meinong and Husserl.(...) For the logical realists, the linguistic analogy held out the promise of transcending the conflicting claims of empiricism and idealism – that is of doing justice both to the concreteness of experience and to the role of abstract reasoning, without lapsing into relativism on the one hand, or a teleological view of history (the frequent consequence of idealism) on the other.112

The starting point of the analogy is the basic linguistic unit, which is a sentence. There have to be counterparts in the world of the propositions expressed by sentences. Such counterparts are usually considered to be facts, understood either as complexes or actual states of affairs. Lindenfeld maintains that they have to be built of things as their constituents, and that, in analogy to sentences, they are the basic units of the world. To see what philosophical analogies could be drawn from this principle, one has only to consider that the basic unit of syntax is commonly considered to be the sentence, the meaning of which is qualitatively different from the words which 112

D.Lindenfeld, The Transformation of Positivism, p. 131-2.

94

ANNA SIERSZULSKA constitute it. Sentences have Gestalt-qualities, so to speak: If the structure of the world resembles the structure of language, then, the basic units of the world must be more complex than 'things', which are the proper analogues of words. There must also be 'states of affairs' of which things are constituents.113

Neither Meinong's objectives nor his complexes fulfil this condition, in the sense that they are not the basic units of the world. There are no such units of the world on Meinong's theory, since even his complexes are arbitrary, because of possessing constituents differentiable, yet essentially undifferentiated, from the continuum of reality. Obviously, objectives as means to apprehend complexes share this feature. We have a different version of logical realism in Meinong's theory. There are no such units of the world which would be made of things, but there are facts of a different kind. These facts are abstract logical structures that obtain. They belong to reality as 'ways it is organized'. Truth depends upon an agreement of the logical structure which is inherent to reality and the one which is apprehended. ♦ Truth as identity of logical structure Epistemological realism involves aspiring to obtain cognitive results that reflect the actual state of reality. The notion of truth related to this way of thinking must be based upon a certain kind of agreement between a proposition, or a sentence, and the world. This is the objective notion of truth. Such understanding of truth is criticized from the epistemic perspective. Sceptical arguments make us believe that the reality we know, mediated by mental contents and systems of representation, never shows its real face to us. In other words, it is claimed that no comparisons with the unmediated world are possible to verify which judgments are correct. For this reason, diverse epistemic notions of truth are adopted instead of the objective one, e.g. based upon coherence, usefulness or instrumental value of theories. Most truth theories which introduce an epistemic notion of truth reject its objective notion entirely. However, in the case of Brentano and also Meinong, we have an example of simultaneously maintaining both the epistemic and the objective notions of truth. They 113

D.Lindenfeld, The Transformation of Positivism, p. 139.

MEINONG ON MEANING AND TRUTH

95

want to preserve the objective sense of truth, but they redefine it from the epistemic point of view, with self-evidence as the epistemic truth criterion. It is possible because the notion of truth based upon self-evidence does not exclude acquiring objective knowledge. It does not guarantee such a result, but in fact it makes some sort of agreement with the world possible, for self-evidence is, after all, a direct cognitive experience. In spite of criticizing the idea of correspondence, Brentano accepts the objective status of truth. His understanding of this idea is close to one of Aristotle's formulations of the definition of truth, slightly adjusted to comply with his own theory of judgments. Aristotle's definition states: 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, and of what is not that it is not, is true. (Metaphysics 1011b26)

Brentano's version of this definition in one of his later essays: Anyone who judges that a certain thing exists, or that it does not exist, or that it is possible, or impossible, or that it is thought by someone, or that it is believed, or loved, or hated, or that it has existed, or will exist, judges truly provided that the thing in question does exist, or does not exist, or is possible, or is impossible, or is thought of...etc.114

This definition expresses beyond any doubt the idea of a certain kind of correctness related to the objective notion of truth. Nevertheless, it is immediately followed by Brentano's epistemic definition of truth, with a comment that one definition comes to the same thing as the other, which can only mean that his epistemic notion of truth is also understood as leading to the objective truth in the outcome. The epistemic definition says: Truth pertains to the judgment of the person who judges correctly – to the judgment of the person who judges about a thing in the way in which anyone whose judgments were evident would judge about the thing; hence it pertains to the judgment of one who asserts what the person whose judgments are evident would also assert.115

Among other things, this implies that a true judgment, in order to be true, does not have to be judged with self-evidence always and by everyone, 114 115

Franz Brentano, The True and the Evident, p. 121. F.Brentano, The True and the Evident, p. 122.

96

ANNA SIERSZULSKA

because the experience of self-evidence does not make it true. The epistemic notion of truth does not replace completely the notion of correctness in the sense of the former definition. Brentano defines truth by means of self-evidence, because correctness of judgments cannot be perceived as a result of comparison between what is in the mind and reality. Therefore, he supposes that it has to be perceived directly, namely as self-evidence of judgments. But how does one compare something that is in the mind with something that is not in the mind? Our knowledge of what is not in the mind is certainly not acquired as a result of making a comparison. Where such knowledge is immediately given, it is a matter of a simple and evident apprehension.116

Brentano claims that there cannot be any possibility of comparing a presentation, which is mental, with its real object, because either the object would have to be known beforehand, otherwise than by way of presentation, and this is impossible, or such comparisons would lead to an infinite regress. This is why he adopts a notion of truth which is just a matter of agreement between what a judgment asserts and reality. To say that a person judges correctly and to say that what he judges is true, are one and the same.117

The notion of truth proposed by Meinong seems to be much closer to correspondence. There should be an identity – which is, at first sight, just the maximum degree of correspondence – between the objective intended in a judgment and the actually subsisting one. But no mention is made of comparing objectives as mental representations to any states of affairs in reality. Like Brentano (and like Frege), Meinong criticizes the traditional notion of correspondence as not suitable for investigations concerning truth. Instead of following the traditional formula that the content of the intellect should correspond to the things, there is an identity between the logical structure which is grasped in a judgment and the one which obtains

116 117

F.Brentano, The True and the Evident, p. 111. F.Brentano, The True and the Evident, p. 115.

MEINONG ON MEANING AND TRUTH

97

in reality. In this sense, his notion of truth is very Brentanian. One should assert what is correct about the world.118 But if these are the circumstances under which one might employ the epithet 'true', what does it mean? It would be in close accordance with common sense, as well as with time-honored tradition to answer this way: What one asserts is true when it agrees with what is – or with what is factual (...) So in its essentials, that situation is one in which an objective somehow presupposed as pseudo-existing is confronted with the pure objective, as it were, on the strength of the latter's factuality.119

The subsisting objective, when it becomes apprehended in a judgment, acquires the property of being true. Meinong speaks about identity here, and it is the identity of an objective seen from two different angles. This is not to say that there is any identity between the true objective as a mental object and the same objective as something existing in the world, which would resemble a state of affairs or a fact on the post-Russellian view.120 The two objectives are entities of the same kind, as it seems, they are logical structures or coordinations that obtain or do not obtain in reality. One is grasped in a mental act and the other is a pure objective that subsists. None of them is a piece of the real world. Objectives do not exist in the world. An objective may present it correctly that a certain object exists in the world, but this does not make the objective a part of the world as well. There exists only the object the existence of which has been stated. Meinong writes that the notion of 'adequacy of representation' was mistakenly understood as a relation between a mental content and a real object. However, it is not the content of the act that should agree with a real object, but the object of the act. If the real object is square, for example, the content of the act need not agree with it and be square as well. It is only supposed to present a square object. By analogy, if a judgment apprehends the objective 'that the table is square', then it 118

Which can be traced back to the definition by Aristotle in the "weaker" version. Compare also Jan Wole ski, Metamatematyka a epistemologia and Jan Wole ski, "Brentano's Criticism of Correspondence Conception of Truth and Tarski's Semantic Theory". 119 A.Meinong, On Assumptions, p. 71. 120 This will be the main difference from an identity theory of truth of the Russellian type.

98

ANNA SIERSZULSKA

apprehends a subsisting factual objective if the objective obtains, yet the objective need not literally contain a square table. This means that the content of the judgment should intend something that is correct about reality in order for the judgment to be true. To wit, it is by all means correct to the point of triviality that if I want to apprehend a quadrangular table by my thinking, I should conceive of the table, e.g., neither as round nor as oval, but simply as quadrangular. But does this somehow mean that for this purpose my representation itself, or more precisely its content, would have to be quadrangular? Or is not the requirement limited, rather to this: that my representation if it is not to lead to any error, must simply be the representation of something quadrangular, i.e., of a quadrangular object? On closer inspection, therefore, the agreement required is by no means an agreement between my representation and the actuality in question, but an agreement between the object of my representation and this actuality.121

The object or objective intended in a judgment is the internal object of the act, the object of its immediate intention, and therefore it belongs to the intellect, but in a different sense from the content of the act. An objective never has a merely mental character. Even the objective of a counterfactual judgment is not merely a mental object. Apart from being the object of a judgment (i.e. pseudo-existing in a judgment), an objective is always identical to a subject-independent subsisting or non-subsisting objective in Aussersein. When it is identical to a subsisting objective, the judgment is correct about the world. Truth is not understood as a similarity of mental content and any actually existing thing. As far as a representation can serve as the basis for an evident affirmative judgment, it offers me a means for getting hold of an actuality, as it were, or intellectually apprehending it – which of course might be a quasi-actuality, something 'subsisting' rather than 'existing'. But we cannot then place this known actuality or quasi-actuality back alongside our representation of it, on a par with the representation, so as to compare these two matters of fact for similarity or dissimilarity. (...) there is no basis here for any claim that the content of the representation has any definite similarity or dissimilarity to the actuality that is coordinated with it.122

121 122

A.Meinong, On Assumptions, p. 190. A.Meinong, On Assumptions, p. 190.

MEINONG ON MEANING AND TRUTH

99

By means of the content of a judgment one can 'get hold of' something actual. However, the representational part of the objective is only coordinated with something actual, it cannot be compared to it for similarity. In general, Meinong's identity means intending correctly the same objective that subsists in reality, while his adequacy, not unlike Brentano's correctness, is a kind of agreement between the objective intended and what is in the world, which does not imply any comparisons. Instead, there is the concept of a judgment hitting the target objective. (...) the truth no longer presents itself as simply a case of grasping a factual objective, in other words, as a case of total identity between the pseudo-existing objective and the factual objective hit upon by the judgment, but as an equality, an agreement of one objective with the other. In this sense truth falls under the famous 'adaequatio intellectus et rei', which has been mistakenly understood as a relation between a content and an object.123

When Meinong uses the word 'identity', he does not express by it the idea of identity in all details, a total identity between the objective intended in an act and the factual one. The factual objective is only hit upon by the judgment, as the one that obtains. The factual objective is identical to the immediate objective as far as its 'identity', or sameness, is concerned, so to say. But the immediate objective is incomplete in its representation of its subordinate objects whenever it concerns real objects, while the factual objective obtains between existing individuals. This kind of cognition involving direct apprehension of factuality is called penetrative. The experience of the act's self-evidence allows us to gain a direct recognition of what is the case. If penetrative apprehension reaches its natural goal, then the objective in question becomes not only penetratively grasped, but it will be also penetratively hit upon 123

A.Meinong, Über Möglichkeit, p. 42: "(...) sich die Wahrheit nicht mehr einfach als der Fall des Erfassens eines tatsächlichen Objektivs, anders ausgedrückt, als der Fall der Identität des pseudoexistierenden Objektivs mit dem betreffenden tatsächlichen Objektiv schlechthin, sondern als die Gleichheit, das Übereinstimmen des einen und des anderen Objektivs darstellt. In diesem Sinne fällt die Wahrheit dann zusammen mit der vielberufenen "adaequatio rei et intellectus", die man nur ganz mißverständlich auf das Verhältnis von Inhalt und Gegenstand bezogen hat."

100

ANNA SIERSZULSKA

as factual, as we may say, in an apprehending experience, which is entirely different in principle from mere contemplative apprehension and the contemplative hitting upon inescapably accompanying it.124

The penetrative apprehending experience, which is different from contemplative apprehension (see the section on self-evidence), is due to the experience of self-evidence. The moment of factuality of an objective can only be perceived through self-evident acts of judgment. Otherwise, the judgment may still apprehend a factual objective, but the factuality of its objective cannot be apprehended. The judgment without evidence, although true (in the outer sense), grasps and hits a factuality penetratively, but in a way only by accident, whereas penetrative grasping belongs naturally to an evident and therefore (inherently) true judgment. Obviously, a factual objective can also be apprehended accidentally by means of an assumption: but this remains contemplative apprehension that in no way can become penetrative because of the accidental factuality of the apprehended objective.125

Factual objectives may also be the objects of evidenceless judgments or assumptions and it is possible to speak about penetration here in an objective sense. There is still a penetration of reality in such cases, since there is still an identity between the objective intended and the obtaining one, but the objective of the act is not true from the epistemic point of view.

124

A.Meinong, Über Möglichkeit, p. 255-6: "Erreicht das penetrative Verhalten sein natürliches Ziel, so wird das Objektiv, um das es sich handelt, nicht nur penetrativ erfaßt, sondern es wird, wie man passend sagen kann, als ein dem Erfassungserlebnis tatsächlich gegenüberstehendes auch penetrativ getroffen, was vom bloß kontemplativen Erfassen und dem damit unvermeidlich zusammengehenden kontemplativen treffen ganz prinzipiell verschieden ist." 125 A.Meinong, Über Möglichkeit, p. 256: "Das evidenzlos gefällte, aber (äußerlich) wahre Urteil erfaßt und trifft eine Tatsache penetrativ, aber nur gewissermaßen per accidens, in des dem evidenten und daher (innerlich) wahren Urteil das penetrative Treffen von Natur zukommt. Natürlich kann ein tatsächliches Objektiv per accidens auch durch eine Annahme erfaßt werden: das bleibt aber stets ein kontemplatives Erfassen, das durch den Zufall der Tatsächlichkeit des so erfaßten Objektivs in keiner Weise penetrativ gemacht werden kann."

MEINONG ON MEANING AND TRUTH

101

Meinong's conception of truth as identity does not resemble the hegelian-idealistic identity truth theories, or deflationary identity theories, because of the penetration of reality and because true judgments are taken to be correct about something which is not subject-dependent. ♦ A realistic identity theory of truth Meinong's theory of truth is an identity theory, on which facts are nothing else than true propositions. But is it indeed the case that Fregean-type accounts of truth have to give up the idea that truth is a notion of any substance? Meinong's theory certainly does not abandon this idea, and arguably neither does Frege's . On both of these theories a proposition is true only if it robustly obtains, i.e. if it is 'correct about reality' as we might say. But both these theories do not specify what this 'correctness about reality' consists in, claiming that it is a primitive notion and impossible to define. And both do not find room for facts as entities in the world, suggesting at the same time a realistic reading of what they mean by the obtaining of a true proposition. In On Assumptions, Meinong explains what it means for an objective to be true and what it means to be false. He speaks about an objective from two different angles. On one hand, it is the proposition which is actually present (pseudo-exists) in the mind of the judging person. On the other hand, it is the abstract proposition (the pure objective) which can potentially be judged by different subjects, but need not actually be judged by anyone. If the pure objective is correct about reality, Meinong calls it a factual objective, and when it is incorrect, it is a non-factual objective. The identity claim concerns the proposition in the mind and the abstract proposition. Judged objectives are true if they are identical to factual objectives. Therefore, a true objective is a factual one. What one asserts is true when it agrees with what is – or with what is factual, which is the clearer expression in view of the foregoing. So in its essentials, the situation is one in which an objective somehow presupposed as pseudo-existing [in the mind] is confronted with the pure objective, on the strength of the latter's factuality.126 126

Alexius Meinong, On Assumptions, p. 71.

102

ANNA SIERSZULSKA

By analogy, judged objectives are false if they are not identical to factual objectives. Then, it is also the case that a falsely judged objective is identical to one of possible abstract non-factual objectives belonging to Meinong's Aussersein. But Aussersein lies beyond all metaphysical concerns and, accordingly, the non-factual objectives possess no being of any kind, they are abstract, 'merely semantic' entities. So we can say that a false proposition is a pure meaning entity. The natural negativum of 'true' is 'false'. At the simplest, an objective is called false in the event that it is not true. And so it is called false in the event that it does not answer to any factual objective. Or, finally, it is called false in the event that it is not factual.127

Meinong's objectives may be factual or not, but they can never be facts in the customary sense, as entities belonging to reality. Why is it so? In the first place, all of them are purely abstract. Factual objectives can be regarded just as abstract function-like coordinations, and the objects are inserted into the basic structure as external arguments about which it is correct. The intended objectives as they appear in the mind consist of representations of objects and also of some non-representational propositional content. The non-representational part of the sense of an objective states the coordination between objects that must be correct with respect to reality and the representations of objects refer to existent or nonexistent entities. As a result, an objective is not a likely candidate for an entity which could be actually found in the world. If A and B are representational objecta, then we can surely say the following: If someone believes at one time that A is, or that A is B, and at another time believes that A is not, or that A is not B, then with this change nothing has changed in the representational objects, and therefore presumably nothing in the representations, either. If one follows Russell's way of thinking, one might at first want to contest this, since presumably even the opposition of 'is' and 'is not' could be included among the things represented.128

Since Meinong affirms a version of logical realism in his theory, he assumes that the logical structure of a true objective, even though not 127 128

Alexius Meinong, On Assumptions, p. 72. Alexius Meinong, On Assumptions, p. 99.

MEINONG ON MEANING AND TRUTH

103

straightforwardly representational, must be identical to the logical structure obtaining in reality. In this sense, the true objective in the mind and the factual objective are the same. If there is any difference between the factual objective and the judged ones, the difference concerns the representational part of their content. A factual objective is taken to concern objects as such, while the objective which is actually judged can involve only aspectual representations of the objects. This does not mean that “I have seen the Morning Star" and “I have seen Venus" express identical true objectives, but they can be both eventually identified with the same pure factual objective which includes all possible ways to represent the object in question. The identity between the objective judged and the factual one still holds as long as the reference to the same objects takes place, and as long as the basic logical structure, which obtains with respect to reality, is still identical. Thus the objective which is judged can be identified with the pure factual objective, so that we can say that a true objective is factual. Factuality is a property which objectives may possess or not. It is a primitive property, and it is indefinable. An objective must bear factuality in itself; and as far as I can see, its factuality is a basic property for which there is no definition and, at least for the time being, also no description.129

Inasmuch as a true objective is the same as a factual one, and under the assumption that factuality is an indefinable property of objectives, we may conclude that truth is as indefinable as factuality on Meinong's conception. That truth is indefinable is one of the claims of a deflationary theory of truth and a reason for identifying facts merely with true propositions. But the message behind this claim may just as well be inflationary. There is another inadequacy in the typical criticism of the phrase "A fact is a true proposition"- a fact is turned into an object. For Meinong, an objective is a kind of an abstract object, but strictly speaking there are no facts, it is only the case that objectives possess the modal property of being factual. We may understand the cutting of reality by objectives to be arbitrary (cf. the discussion concerning the undetermined but determinable 129

Alexius Meinong, On Assumptions, p. 56

104

ANNA SIERSZULSKA

constituents of the continuum of reality in Ch. 2). Objectives can cut reality in many different ways or pick out its different fragments. Since such fragments of reality could hardly be called objectives, but either complexes or simple objects, we may conclude that there is no need to consider objectives ontologically as entities of the world. All we have to assert is their status as subject-independent meaning-objects, members of the realm of pure objects, to be grasped by an intention of the mind and to be meant by the sentences of a language. In this way, objectives retain a nonsubjective character, they can become the objects of the acts of many subjects, and at the same time the ontological status of factual objectives (as logical structures that obtain in reality) does not posit problems, similar to the problems with facts as constituents of reality. The identity theory, if it is taken to be realistic, suggests that a factual objective is independent from a true objective, that it is something independently and objectively there in the world... This is right, but it is not something material in the world, it is something FORMAL in it. Every objective, factual or not – true or false, is an abstract object. It is not a fragment of reality, like a complex can be, because it includes nothing of the substance of reality, nothing of the world's matter, nothing of its physical energy. An objective is a purely formal entity. Meinong's true objective in a judgment reflects the actual arrangement or coordination of things in the world, and it is identified with this arrangement. Not on the basis of correspondence, though. Objectives obtain or they don't. The relation is function-like and the function is 'satisfied' as concerning a certain fragment of reality. There is no room here for making comparisons. In opposition to a conception of this kind, the correspondence theorists are bound to make comparisons between sentences or propositions and what they call facts. ♦ Truth and the modalities of objectives Objectives do not possess existence. The ways of being typical for objectives are either subsistence or the neutral status of pure objectives in Aussersein. True objectives subsist, which is an ideal way of being. This allows Meinong to introduce an absolute notion of truth, as a consequence of the idea that true objectives subsist timelessly. Thus it is also possible to

MEINONG ON MEANING AND TRUTH

105

say that their capability to be true is eternal, that it characterizes them absolutely, in the present, past and future. Among the other ways in which subsistences differ from existences is by their not being tied to any determinations of time, and in this sense being eternal – or rather timeless. Naturally this holds for the objective too. My desk is a thing existing at a certain time; but that it exists at this time is something which subsists now as well as in all the past and future, although it was inaccessible to knowledge in past times and it will have disappeared from knowledge in future times.130

To be more precise, all objectives are timeless abstract entities. Those objectives that are not true in judgments and, accordingly, they do not subsist, possess an abstract way of being as well. All objectives are timeless as pure objects. But only some of them, the subsisting ones, can be true in judgments, regardless of the point in time at which the act of judgment takes place. Objectives grasped in judgments are true if they are identical to factual objectives131, but it should be added that an objective identical to a factual one is never true when there is no judgment it would be the object of. On Meinong's conception, truth is inseparable from cognition. If on one occasion it is called factual that A is B, and on another occasion it is called true, what distinction is really operating here? Apparently the distinction is this, that while the reference to an apprehending judgment only appears to play a role in the case of 'factual', this reference has an essential meaning in the case of 'true'.132

For an objective to be called true, there must exist a mental act apprehending this objective. Objectives as considered independently of cognition and any mental acts, belong, according to Meinong's theory of objects, exclusively to the sphere 'beyond being' (Aussersein), which contains all possible objects of consciousness. (...)when I think of unclouded human happiness or of the perpetual motion machine, my thoughts are directed to "something", i.e. to an object, just as surely as if it were a matter of the most everyday piece of actuality. (...) If the object as 130

A.Meinong, On Assumptions, p. 52. Cf. Über Möglichkeit, pp. 40-42. 132 A.Meinong, On Assumptions, p. 71. 131

106

ANNA SIERSZULSKA

such is absistent and we recognize this, then right there we have an objectapprehension that is unconfined by the limits of being.133 [absistent = beyond being]

So called pure objectives belonging to this sphere are possible intentional objects which our consciousness can be directed at. They possess a certain existential status (subsistence or its lack), depending on the state of the world, but they are neither true nor false. (...) although factuality is nothing subjective, it is by no means free from subjectivity: and actually it is many times stressed here, that truth and falsehood do not lie in things, but in the grasping of things.134

Truth has an epistemological character on Meinong's view. Truth, as a property not of any objects in the world, but primarily of judgments, is tightly bound to cognition. However, in consequence of the limitations of cognition, the apprehension of truth is not free from subjectivity. In our apprehension of an objective, we cannot always judge with certainty about its truth, and therefore subsistence, or its lack. In this way, commencing from the question of the subsistence of an objective, and from the modal sphere of the necessity, factuality, possibility and impossibility of its subsistence, we move into the parallel realm of modalities such as are related to cognition: truth, probability, falsehood. Truth, probability and falsehood are understood in Meinong's conception as parallel to modal properties of objectives. Nevertheless, they are not modal properties in their own right. Strictly speaking, truth by no means constitutes a new and peculiar modal property of the objective (...) For the concept of truth has arrived at that fundamental position in epistemology only insofar as it has been thought permissible to lay aside the restriction to pseudo-existing objectives, a move that probably just brings

133

A.Meinong, On Assumptions, p. 170. A.Meinong, Über Möglichkeit, p. 41: "(...) obwohl die Tatsächlichkeit nichts Subjektives ist, doch keineswegs frei von Subjektivität: und in der Tat ist von alters her ja immer wieder betont worden, daß Wahr und Falsch nicht in den Dingen liege, sondern im Erfassen der Dinge."

134

MEINONG ON MEANING AND TRUTH

107

one unawares into the domain of a genuinely fundamental modal property, namely factuality.135 At all events, it seems that there are not, strictly speaking, any distinctive modal properties in objectives beyond factuality, possibility, necessity, and their opposites.136

The problem with Meinong's truth and the whole group of parallel modalities is that they are not 'permanent' properties of objectives, and for this reason they cannot be conceived as genuine modal properties. They are predicated about objectives only when these are considered as apprehended by mental acts and pseudo-existing in the mind of the subject (a pseudoexisting objective is the one which is the immediate object of the act of judgment, it pseudo-exists in this particular act; in opposition to a pure objective, it is a concrete mental entity137). Thus necessity, factuality, possibility and non-factuality are genuine modal properties of objectives as abstract pure objects, while their function of truth bearers is only temporary or occasional, even though it is one of the only two functions they serve, next to being objects of intention. This hyper-epistemological point of view becomes more plausible when we regard truth as a potential property, or a disposition, of all subsisting objectives. Modal properties, in the sense in which they apply to objectives, could not be properties of objecta. We certainly say that a given object is a possible object. By this we mean that the nature of the possible object is such that the properties assigned to it do not contradict each other, but not that it could or could not be the case, as we say about objectives. Of course, possibility of objectives depends upon the nature of their subordinate objects, in such a way that an objective, whose subordinate object as apprehended by the objective is incomplete with respect to the property predicated in this objective, can only be called possible if the property in question can be predicated about this object without contradiction. In other words, possibility of objectives is related to incomplete apprehension of their subordinate objects. Factual objectives are flatly and uninterestingly possible on Meinong's view, for obviously 135

A.Meinong, On Assumptions, p. 72. A.Meinong, On Assumptions, p. 73. 137 See On Assumptions, 49 ff. 136

108

ANNA SIERSZULSKA

they cannot imply any contradiction. The logical notion of possibility in the most general sense, as a lack of contradiction, applies to objectives in the context of a priori reasonings, when we can say that an objective is possible provided that it stands in no contradiction to any true objectives. Something completely different is meant when the possibility of an objective in an empirical context is considered. The factuality and possibility of an objective in an empirical context, depend on the state of reality. In particular, on the scale between factuality and impossibility, there are different degrees of empirical possibility which can characterize objectives. So that I can make this clear as briefly as possible, it should first of all be pointed out that factuality is by its very nature located at the end of a scale of magnitude, a scale whose points can be conceived of as representing every degree of possibility. The word "possibility" is here understood in the sense in which possibility admits of gradation. In this sense, one objective is called "very possible" and another "hardly possible," and in special circumstances, e.g., such as are afforded by games of chance, numerical determinations of degrees of possibility are undertaken. So let us say that all possibility is reduced factuality.138

Factuality is the highest limit of possibility to be true for objectives. Then, we have different degrees of possibility. The degree of possibility sometimes can be calculated, as we can calculate the possibility of drawing a certain number in a lottery. What Meinong understands as degrees of possibility, may be compared to probability in its most unsubjective sense: it is the degree to which something is probably the case in consequence of how the world is itself, and regardless of the state of anyone's knowledge. In the subjective sphere of cognition, factuality of objectives is never apprehended directly, but as their truth, and the degree of possibility of an objective is apprehended as a degree of probability. I may be justified by evidence in surmising that a certain result will occur in gambling. Naturally, the factuality of the result does not then stand established; but its possibility does, and as a stronger surmise is justified, the possibility is greater. But in this surmise, one's apprehending of the possibility is no more explicit than our previous apprehending of a factual objective's factuality, when we judged that objective. And as in that case, here too a judgment-about provides 138

A.Meinong, On Assumptions, p. 68-9.

MEINONG ON MEANING AND TRUTH

109

what is required. One simply judges so: that it is in such and such a degree possible that the result will occur.139

A factual objective may be either necessary or contingent. When it is necessary, its truth is apprehended with self-evidence for certainty accompanying the act of judgment. We can say that its factuality is apprehended as truth. Logical possibility of objectives is also apprehended with evidence for certainty, as a lack of contradiction between objectives. But contingent objectives, except in introspective judgments, may only be apprehended with evidence for surmise. We are not certain, then, about their truth, and we ascribe different degrees of probability to them, according to the strength of our experience of the act's self-evidence. Thus, the supposed factuality of empirical objectives – their possibility to be true – is apprehended as probability. Thus, the impossible is that which is not possible or which stands at the zero-point of the possibility-line. In this sense, neither impossibility nor possibility is tied to apriority; by way of contrast, rational impossibility and the rational possibility derived from it by negation are customarily distinguished as logical impossibility and logical possibility, respectively. – Objectives that are by their nature inaccessible to the rational mode of knowledge are customarily opposed to necessary objectives as contingent ones.140

The objective notion of probability can be identified with the degree of empirical possibility of an objective. It is a special case of apprehending the possibility of objectives that objective probability degrees can be ascribed to them on the basis of a mathematical calculation. Meinong calls it 'surmise-free' probability, because in the cases when it is possible to apply an objective probability calculation, the degree of possibility is grasped in its actual value with evidence for certainty. Obviously, as the investigations of the introduction have shown, the term 'probability' can also be taken in the sense of 'possibility', with an eye especially upon 'surmise-free probability'. But this non-subjective use of the word 'probability', as already mentioned, remains stranger to our feeling of language than its subjective use; on the other hand, it can be hardly doubted that whoever

139 140

A.Meinong, On Assumptions, p. 69. A.Meinong, On Assumptions, p. 71.

110

ANNA SIERSZULSKA

utilizes the probability calculus aims mainly to obtain the most precise surmise and not to grasp the degree of possibility.141

The objective probability of a certain result in a game of chance is in fact also related to a kind of surmise concerning this result. Yet it differs from subjective surmise-probability, since the degree of surmise-probability is grasped with evidence for surmise only. Then, the apprehension of the degree of possibility of an objective is merely approximated and may be incorrect. Objective probability, to the contrary, reflects the degree of possibility of an objective correctly, so there is no surmise here concerning the degree of its possibility, just a surmise as to whether something will be the case. Objectives possess both subjective and non-subjective properties, so those of them that find expression under the name 'probability' may be confronted with each other as 'surmise-probability' and 'surmise-free probability'. The expressions 'probability in the narrower sense' and 'possibility' are more characteristic.142

Objective probability is contrasted with probability in a narrower sense, which is surmise-probability. We may say that two notions of probability, 'objective' and 'subjective', are employed by Meinong in relation to contingent objectives as their possibility is apprehended. Objective probability in games of chance and relative probability in inductive reasonings, can be calculated with the means developed by probability theories (Meinong uses the law of large numbers and Bayes's theorem). 141

A. Meinong, Über Möglichkeit, p. 305: "Freilich ist, wie die Untersuchungen der Einleitung ergeben haben, der Terminus "Wahrscheinlichkeit" auch im Sinne von Möglichkeit zu deuten, indem man speziell die "vermutungsfreie Wahrscheinlichkeit" ins Auge faßt. Aber bleibt dieser unsubjektive Gebrauch des Wortes "Wahrscheinlichkeit", wie schon erwähnt, dem Sprachgefühl immer fremder als der subjektive; dann aber wird es schwerlich in Zweifel zu ziehen sein, daß, wer Wahrscheinlichkeitsrechnung treibt, sein Absehen doch zumeist auf möglichst präzisiertes Vermuten und nicht auf das Erfassen von Möglichkeitsgraden gerichtet hat." 142 A.Meinong, Über Möglichkeit, p. 711: "Objektive haben sowohl subjektive als unsubjektive Eigenschaften, und was von diesen auf den Namen 'Wahrscheinlichkeit' Anspruch hat, kann einander als 'Vermutungswahrscheinlichkeit' und 'vermutungsfreie Wahrscheinlichkeit' gegenübergestellt werden. Charakteristischer sind die Bezeichnungen 'Wahrscheinlichkeit im engeren Sinne' und 'Möglichkeit'."

MEINONG ON MEANING AND TRUTH

111

But the application of such tools is limited. Most of the time, surmiseprobability cannot be calculated in the sense that its numerical value cannot be provided, as it occurs in such situations when the degree of probability of an objective can be estimated only subjectively, according to the strength of the evidence for surmise. The degree of possibility is never directly accessible to a judgment and when we cannot calculate it, we can only depend upon the subjective experience of evidence. Surmiseprobability finds application to all kinds of uncertain judgments concerning possibility of objectives. The subjective probability that an empirical objective is true, and therefore factual, is ascribed in proportion to the strength of the evidence for surmise. The evidence for surmise, if it is a justified surmise, corresponds to the degree of possibility that the objective may be true. In this way, as Meinong says, the probability of an uncertain objective grasped in a judgment aims to merge with the non-subjective possibility of this objective to be true. In other words, just as truth is the way to apprehend factuality, probability in the narrower sense is the way to apprehend the degree of possibility to be true of uncertain empirical objectives. Degrees of probability ascribed to objectives aim to reflect degrees of their possibility. That being the case, one could define probability as the possibility of truth. Naturally, we may ask ourselves what purpose there is in setting the possibility of truth alongside simple possibility, as is done in this reemployment of the notion of possibility. The legitimation lies in the function of surmise as a substitute for insufficient certainty; surmise fulfils this function in proportion to its inherent chance of hitting something true or correct. If one subsequently leaves the objective's presupposed pseudo-existence out of account here, as is done in the case of truth, then probability merges together with possibility. But one can still keep it separate from the latter by making use of the "subjective" aspect of possibility in another way. An objective can be called probable with the idea that it constitutes an object of justified surmise, and one can think of the magnitude of the possibility as accompanying the strength of the surmise that is justified.143

Meinong's notion of possibility allows for degrees. His 'empirical' possibility differs in this respect from ordinary logical, or rational, possibility, based on non-contradiction, which does not possess degrees. The degrees of possibility are associated with either 'objective' or 143

A.Meinong, On Assumptions, p. 72-3.

112

ANNA SIERSZULSKA

'subjective' probability attributions, and so they may be calculated at least in some circumstances. As factual objectives are not true permanently, but only potentially in the acts of judgment, also possible objectives can only potentially be apprehended as probable. And the possibility to be true of empirical objectives can only become probability as ascribed to empirical judgments. Still, an objective may be probable only to the extent to which it is possible. The natural relationship between probability (in the narrower sense) and possibility is clear immediately. This kind of probability does not consist in surmising the objective with justification, but in the fact that it can be surmised with justification. Therefore, it is to be expected that a possible objective will be also probable, and it will only be probable to the extent to which it can be called possible.144

It is clear, therefore, that the 'parallel' modalities: truth, probability and falsehood, are not permanent modal properties of objectives in Meinong's theory. They are related to the apprehension of the genuine modal properties: factuality, necessity, possibility and unfactuality. Diagram 2 Modalities of Objectives factuality

possibility

(apprehended as)

(apprehended as)

truth 144

1

probability

non-factuality (apprehended as)

0

falsehood

A.Meinong, Über Möglichkeit, p. 36: "Dagegen ist die natürliche Zusammengehörigkeit der Wahrscheinlichkeit (im engeren Sinne) und der Möglichkeit auf den ersten Blick klar. Da diese Wahrscheinlichkeit nicht darin besteht, daß das betreffende Objektiv mit Berechtigung vermutet wird, sondern darin, daß es mit Berechtigung vermutet werden kann, so liegt es nahe genug, zu erwarten, daß ein mögliches Objektiv auch wahrscheinlich, und zwar in dem Maße wahrscheinlich ist, in dem es möglich heißen darf."

MEINONG ON MEANING AND TRUTH

113

Diagram 3 Possibility

rational judgement

logical

empirical

consistency (has no degrees)

possibility to be true (comes in degrees)

logical possibility

empirical judgement

(apprehended as)

surmise-free probability (objective)

(evidence for surmise)

surmise probability (subjective)

♦ Self-evidence as a substitute criterion of truth Contrary to Brentano, Meinong treats the idea of self-evidence as a criterion of truth with great reserve. That is because the need to rely upon self-evidence introduces an element of subjectivity into the apprehension of truth. Brentano emphasises that we usually do not speak about being 'self-evident for somebody', since we believe that what is self-evident is self-evident regardless of a particular subject's experience and he rejects gradation of self-evidence.145 But Meinong is not convinced about the correctness of such a stand-point. Truth which is dependent upon the experience of self-evidence possesses on his view a subjective character. This concerns every concrete act of judgment when the truth of its objective is considered from the perspective of what is available to cognition. However, such a subjective apprehension of truth is contrasted with what is objectively true.146 Self-evidence cannot be regarded, for this 145 146

e.g. in The True and the Evident, p 68 and p. 144. Über Möglichkeit, p. 32-33.

114

ANNA SIERSZULSKA

reason, as the criterion of truth, because it is not the case that a true judgment must be self-evident, and the question always remains whether, apart from an inner confirmation, the judgment is objectively true, i.e. justified by the factuality of the objective it intends.147 Evidence cannot be defined as the "property of being true" (...) it remains an open matter every time, whether a given judgment, regardless of its inner grasping of truth, is also an evident judgment in the sense of the definition in question. Truth does not depend on evidence by definition (...)148

The subjectivity of the apprehension of truth results in the uncertainty of cognition mediated by self-evidence. Evident judgments do not guarantee their own truth, their self-evidence may be only apparent, while nonevident judgments may just as well be true.149 Self-evidence is not, according to Meinong, a valid criterion of truth, although it serves the purpose of recognizing truth for want of a better criterion. Evidence does not mark out truth, but it can provide a convenient criterion of truth.150

The sceptical approach is a result of Meinong's attempt to explain why it does not happen every time that a judgment intending a factual objective uncovers its factuality due to self-evidence, which would be the case if a true judgment carried self-evidence as a rule. Meinong observes, on the other hand, that if a true judgment always carried in itself an ability to confirm the factual nature of its objective, then infallibility of cognition would be a necessary consequence of a theory so formulated. Our 147

This suggests a possible influence of Bolzano's distinction between inner confirmation and objective justification of judgments. 148 A.Meinong, Über Möglichkeit, p. 463: "Evidenz dürfte nicht etwa als 'Eigenschaft wahr zu sein' definiert werden (...) es bliebe also jedes mal noch die Frage offen, ob ein gegebenes Urteil unbeschadet dessen, was die innere Wahrnehmung darüber aussagt, auch ein evidentes Urteil im Sinne der fraglichen Definition sei. Wahrheit hängt also an der Evidenz nicht ex definitione (...)" 149 Mediated evidence in case of inferences is a special case in this respect. Cf. On Assumptions for inferences about a priori matters, p. 134 ff, and Über Möglichkeit for inductive reasonings with mediated evidence, p. 675. 150 A.Meinong, Über Möglichkeit, p. 721: "Evidenz macht die Wahrheit nicht aus, kann aber Günstigenfalls ein Wahrheitskriterium abgeben."

MEINONG ON MEANING AND TRUTH

115

deficiency in perceiving the relation between reality and its apprehension, as he writes, obviously gives no ground to such a conviction. (...) I also cannot see how our so deficient knowledge of the relation between the apprehension and what is apprehended could give support to such a conviction. In fact, I am by no means in a position to understand the relation between evidence and truth more deeply on this ground (...)151

Undoubtedly, Meinong focuses rather upon the deficiencies and faults of the theory of self-evidence than upon strengthening the theory in order to stabilize the foundations of knowledge. The problems with the theory of evidence which he mentions mostly concern the impossibility of excluding an error in spite of the presence of this property of judgments, and the question of regressive reasoning at the attempts to double-confirm reflectively that judgments are self-evident. Meinong tries to face the objections, but he partly subdues to them. He provides a detailed discussion concerning the apprehension of factuality via self-evidence of judgments. He distinguishes two ways of grasping an objective in an act of judgment: contemplative apprehension and penetrative cognition. The apprehending function that the contents perform is presentation, upon which all contemplative thinking is based: evidence, on the other hand, quite certainly serves the purpose of grasping factuality (or possibility), but in an entirely penetrative way. It constitutes what was called above the fundamental act of cognition and gives a determination of a judgment to the act. If this is right, then every possibility to see evidence as belonging to the content of the act, is excluded.152 151

A.Meinong, Über Möglichkeit, p. 442: "(...) ich kann auch nicht absehen, wie unsere so mangelhafte Kenntnis der Relation des Erfassens zum Erfaßten Anhaltspunkte für einen solchen Beweis bieten könnte. Freilich bin auch ich keineswegs imstande, der Relation zwischen Evidenz und Wahrheit noch tiefer auf den Grund zu sehen (...)" 152 A.Meinong, Über Möglichkeit, p. 442: "Die Erfassungsfunktion der die Inhalte dienen, ist die Präsentation, bei der alle Kontemplation anhebt: die Evidenz dagegen dient zwar sicher dem Erfassen der Tatsächlichkeit (resp. Möglichkeit), aber durchaus dem penetrativen. Sie konstituiert ja dasjenige, was oben der Fundamentalakt des Erkennens genannt worden ist und eine Determination des Urteilens seinem Akte nach darstellt. Ist das richtig, dann ist damit jede Möglichkeit, die Evidenz dem Inhalte des Urteils zuzuschreiben, abgeschnitten."

116

ANNA SIERSZULSKA

The content of the act grasps the objective contemplatively: it presents the objects involved and apprehends the way they are coordinated. There is, however, no such moment of the content that would be related to selfevidence and cannot be, because the content of every true judgment would have to unveil the factuality of its objective, which is not the case. The presence or absence of self-evidence in a judgment has no impact upon the truth of its objective. From this, Meinong draws a conclusion that such modal properties of the objective as factuality or possibility are grasped non-contemplatively and non-presentationally, due to the self-evidence of the act itself. I cannot doubt about it that penetrative hitting upon by way of an evident judgment is the basis of all cognition: perhaps we could describe what does not belong to the content, and therefore is given outside the contemplative sphere, as a penetrative fundamental act or simply as a fundamental act.153

It is not required, as Meinong maintains, that in order to confirm the apprehension of factuality through an evident judgment, an additional evident judgment about the basic judgment being evident must also appear (even though such a judgment about the judgment is needed to become aware of its evidence).154 Such a double-confirmation he regards as unnecessary, since it has no importance for the occurrence of cognition what the subject thinks about the character of his judgment, as far as the critical judgments do not interfere with the basic one. The fact that the existence of cognition can be grasped and proved in no other way than by way of cognition itself does not ground any objection, along the principle of self-justification. If the proof uses freely a premise that assumes directly or indirectly the existence of cognition, then we have a petitio principii. However, the status of the apprehending judgment has nothing to do with the above premise, which does not prevent that this status is the same as determined by the truthcharacter, or cognitive-character, of the correct judgment. Whether and how the judging subject thinks about the judgment's character is unimportant, as long as 153

A.Meinong, Über Möglichkeit, p. 264: "Ich kann nicht daran zweifeln, daß penetratives Treffen durch evidentes Urteil die Grundleistung alles Erkennens ist: vielleicht könnte man das, was davon nicht mehr zum Inhalt gehört und dadurch außer die kontemplative Sphäre gestellt ist, als penetrativen Fundamentalakt des Erkennens oder auch wohl als Fundamentalakt schlechthin bezeichnen." 154 A.Meinong, Über Möglichkeit, p. 446 ff.

MEINONG ON MEANING AND TRUTH

117

the critical judgments do not interfere with the basic judgment itself, which may easily happen.155

The role of self-evidence is such that without the experience of selfevidence one could still be right, but there would be no point of support for the conviction that one is right. The critical doubt concerning accidentally correct judgments would always have to be very high. Were it impossible to 'recognize' whether any judgment apprehends a factuality or not, there would be completely nothing to contradict the occurrence of any case. Even then it would be possible to 'be right', but there would be no ground at all to know if and when one would be right; the critical doubt would be able to find at best only an accidentally correct solution.156

Naturally, the self-evidence of a judgment influences greatly the subject's beliefs. The apparent necessity of a confirmation by a critical judgment may have something to do with understanding self-evidence as a kind of insight into reality. Meinong points out that when we speak about insight in ordinary contexts what is meant is intellectual insight. Therefore, the notion of an insight seems not to pertain to mental activities which do not involve comprehension. But as it is possible to comprehend without possessing any insight into reality, so it is also possible to have an insight into reality without comprehending. Meinong claims that empirical 155

A.Meinong, Über Möglichkeit, p. 458: "Daß nämlich die Existenz der Erkenntnis nicht anders als selbst durch Erkenntnis erfaßt und erwiesen werden kann, begründet dem Selbstgültigkeitsprinzip zufolge keinen Einwand. Würde freilich bei dem Erweise eine Prämisse benutzt, die die Existenz der Erkenntnis direkt oder indirekt in Anspruch nähme, dann hätte man es mit einer petitio principii zu tun. Aber die Beschaffenheit des erfassenden Urteils geht nicht in dessen Prämissen ein, was nicht hindert, daß sie dasjenige ist, was den Wahrheits- resp. Erkenntnischarakter des betreffenden Urteils entscheidet. Ob und wie das urteilende Subjekt über diesen Charakter denkt, ist einerlei, solange derlei Nachurteile nicht etwa, was freilich begegnen kann, das Urteil selbst beeinträchtigen." 156 A.Meinong, Über Möglichkeit, p. 462: "Und könnte man es keinem Urteile 'ansehen', ob es eine Tatsache erfasse oder nicht, so läge darin freilich nichts, was das Eintreten jedes Zufalles verhinderte. Man könnte auch dann 'recht haben', aber man hätte schlechterdings niemals einen Anhaltspunkt zu wissen, ob und wann man recht hätte; der kritische Zweifel könnte niemals eine andere als höchstens eine nur wieder zufallich richtige Lösung finden."

118

ANNA SIERSZULSKA

judgments may possess self-evidence, which is a pure insight without comprehension. One can therefore say: in everyday life, when we speak about insight, we mean insight with comprehension, and as a result of this we deprive of evidence all judgments which lack comprehension. In reality, there is, as it happens, also insight without comprehension, although not the other way round, comprehension without insight. And as we can differentiate between immediate and mediated evidence, so we have an equal right to place immediate comprehension in opposition to mediated comprehension.157

This is connected with his idea of self-evidence as belonging not to the content, but to the act of penetrative cognition itself. The act grasps the factuality of its objective as a modal moment through the experience of self-evidence. Grasping the moment of factuality is not correlated with anything that could present itself to the mind as a content and become subject to comprehension. The self-evidence of an act provides a contentless insight into reality. An insight of this kind requires no comparison of thoughts and things, as suggested by the classical correspondence theories. The truth of an objective is a matter of correctness which is confirmed intuitively. This much can be said in favour of Brentano's and Meinong's conceptions of self-evidence, as intuitive, direct apprehension and as an epistemically valid criterion of truth. No doubt it is quite natural to accept that one can often recognize the agreement of a judgment with a fact immediately, as something self-evident and obvious. Experiencing judgments as self-evident has a role in the formation of our beliefs and we would not be inclined to assert anything in a decisive way without it. However, self-evidence as a criterion of truth allows for too much subjectivity, and in Meinong's conception it serves rather as a substitute criterion, tightly related to the process of cognition. It is the objective 157

A.Meinong, Über Möglichkeit, p. 235: "(...) das Tägliche Leben meint, wo vom Einsehen die Rede ist, sogleich das verständnisvolle Einsehen, so daß es Urteilen, denen das Verständnis fehlt, daraufhin auch die Evidenz aberkennt. In Wahrheit gibt es, wie berührt, ganz wohl Einsehen ohne Verstehen, dagegen nicht umgekehrt Verstehen ohne Einsehen, und sowie man zwischen unmittelbarer und mittelbarer Evidenz zu unterscheiden pflegt, mit ebensoviel Recht darf man dem unmittelbaren Verstehen jenes mittelbare Verstehen gegenüberstellen."

MEINONG ON MEANING AND TRUTH

119

notion of truth, based upon the identity of the objective intended in a judgment with the factual one, that determines the standard of truth, even when it turns out to be cognitively inaccessible. ♦ Surmise-probability of objectives as a remedy for subjectivity of cognition Rational knowledge can reach the goal of finding the absolute truth due to the experience of evidence for certainty accompanying rational judgments. In the foregoing we saw fit to claim without further qualification that evidence appropriate to certainty was a content factor in a judgment; now we can naturally make such a claim about evidence taken specifically as a priori or rational. And if the factuality of the objective is what evidence appropriate to certainty apprehends, then necessity is what is similarly coordinated with rational evidence, as the latter's object.158

In the case of rational judgments, the subsistence of the objective, which is grasped by means of its self-evidence, is parallel to the necessity of a given objective. Claims about the necessity of objectives can be justified on the basis of their content alone, although such a justification may require a formal proof. The correctness of rational judgments is confirmed by their self-evidence either immediately or through the evidence of the premises of the inference. In any case, evidence for certainty does not apprehend any reality transcending the judgment's objective itself, or its relations with other objectives. The resulting judgment has a good chance to be absolutely true.159 Empirical cognition is concerned with the misty regions between the obviously true and the unmistakably false. This is the sphere of the merely probable, which may be apprehended at best with evidence for surmise. This kind of cognition leaves open the question of whether the objective intended can indeed be identified with a factual one. Judgments which lack evidence for certainty, i.e. all empirical judgments except introspective ones, intend objectives that are never subjectively confirmed to be true by 158

A.Meinong, On Assumptions, p. 70. As to rational judgments and the evidence of inferences see Erfahrungsgrundlagen, p. 32.

159

120

ANNA SIERSZULSKA

the experience of their evidence, but they seem to lie on the borderline between truth and falsehood. Truth is strengthened probability. Probability has an intermediate position between the true and false, regardless whether one considers probability or what is probable.160

The inability to determine the truth value of an empirical objective must result in a truth value gap of an epistemic character. Such possibly factual objectives, when intended by judgments, can be ascribed different degrees of probability instead of truth. In this way, the gap is filled and we are not left without any clues how to make predictions and draw conclusions in situations of epistemic indeterminacy of the truth values of empirical objectives. It has already been explained above that in exactly the same way as truth can only be predicated of objectives when they are apprehended in judgments, also probability can be predicated of objectives only if they are apprehended. Like truth, probability is not a property of objectives as pure objectives. And similarly, we can speak here of potential probability of objectives as potential objects of intention. Probability is predicated of an objective primarily insofar as a surmise is directed at the objective. And while the epithet 'probable' naturally does not connote correspondence to a factual objective, it does connote the possibility of such a correspondence, the chance of being true.161

For a probable objective, there is a certain degree of probability that it hits upon a factual objective. In other words, there is a chance that an empirical judgment is correct about reality in the sense that the judgment's objective can be identified with a factual one. So there is a possibility of the surmise being correct, which Meinong calls 'possibility to be true', but no certainty

160

A.Meinong, Über Möglichkeit, p. 722: "Wahrheit ist gesteigerte Wahrscheinlichkeit. Wahrscheinlichkeit hat eine Mittelstellung zwischen Wahr und Falsch, gleichviel, ob man die Wahrscheinlichkeit oder das Wahrscheinliche in Betracht zieht." 161 A.Meinong, On Assumptions, p. 72.

MEINONG ON MEANING AND TRUTH

121

and no justification to decide about its truth from the epistemic perspective.162 If we decide about truth (subjective or absolute) with justified certainty, as it was shown above, then probability in the narrower sense, the surmise probability, stands to truth as justified surmise stands to justified certainty: here truth appears, cum grano salis, as the upper limit of probability in the same way as certainty is the upper limit of surmise.163

The highest possibility to be true amounts to factuality and so the highest probability amounts to truth. It is not surprising that truth appears in Meinong's theory as the upper limit of probability, because when it is possible to ascribe the maximum degree of probability to an objective, it will most certainly be absolutely true. Ascribing probability equal 1 requires absolute certainty, related to inherent factuality of objectives. Such a situation, however, takes place rarely and it concerns rational cognition and introspection exclusively. The surmise-probability of empirical objectives is of a lower degree, anywhere between 1 and 0 on the scale. In opposition to the objective notion of probability, Meinong calls surmise-probability 'probability in the narrower sense', which includes only 'subjective' probability attribution inherently confirmed by the strength of evidence for surmise. The apprehension of the probability of an objective through the strength of evidence is correct, if it corresponds to the degree of possibility to be true pertaining to the apprehended objective. Justified is a surmise, if the intensity of the existence and so the possibility of its objective, is in agreement with the strength of the surmise: the outer certainty justification is the limiting case here. 51. There is also inner confirmation of a 162

In the case of empirical probability, the notions of inner confirmation and outer justification are most of the time both expressed by the word "Berechtigung". 163 A.Meinong, Über Möglichkeit, p. 44: "Bestimmt man, wie oben vorübergehend geschehen ist, die Wahrheit (subjektiv und absolut) durch Heranziehung der berechtigten Gewißheit, so steht unverkennbar die Wahrscheinlichkeit im engeren Sinne, die Vermutungswahrscheinlichkeit, der Wahrheit gegenüber wie die berechtigte Vermutung der berechtigten Gewißheit: hier erscheint cum grano salis Wahrheit ebenso als die obere Grenze der Wahrscheinlichkeit, wie Gewißheit die obere Grenze der Vermutung ist."

122

ANNA SIERSZULSKA

surmise and surmise-evidence. Evident surmises may be false in spite of their evidence.164 Probable (in the narrower sense) is an objective, if it can be apprehended by means of a justified surmise. The strength of the surmise goes together with the degree of probability: certainty and truth are the upper limits.165

Subjectively confirmed surmises may turn out to be incorrect, they may not reflect the correct degree of possibility to be true pertaining to a given objective. Subjective evidence for surmise may possess an unjustified degree of intensity, as we move within the realm of subjective probabilities, with the whole complexity of the epistemic situation. Empirical cognition can never reach the highest level of certainty and its justification is mostly subjective. This is the case also with inductive reasonings, where it is possible to make some probability calculations. Empirical judgments are probable to a subjectively confirmed degree. This kind of probability depends upon initial knowledge and other subjective factors which decide that a judgment receives an inner confirmation.166 Assigning probabilities to uncertain judgments, as suggested by Meinong, can become less arbitrary with the help of some tools developed by later subjective probability theories, like betting strategies, desirability scales and conditional belief estimations. The probabilistic strategy is employed as a remedy to overcome scepticism in the sphere of empirical cognition. Due to the subjectivity of evidence for surmise, and in want of any other epistemically valid criterion of truth, we would be often at a loss when trying to draw conclusions from uncertain empirical premises. Probability attribution provides a solution in this cognitively uneasy situation. Meinong's views influenced immediately 164

A.Meinong, Über Möglichkeit, p. 721: "Berechtigt ist eine Vermutung, sofern der Vertatsächlichungsgrad, also die Möglichkeit ihres Objektivs, zur Vermutungsstärke paßt: die äußere Gewißheitsberechtigung ist der Grenzfall hierzu. 51. Es gibt auch innere Vermutungsberechtigung und Vermutungsevidenz. Evidente Vermutungen können trotzt Evidenz auch falsch sein." 165 A.Meinong, Über Möglichkeit, p. 722: "Wahrscheinlich (im engeren Sinne) ist ein Objektiv, sofern es durch eine berechtigte Vermutung erfaßt werden kann. Die Vermutungsstärke geht zusammen mit dem Grade der Wahrscheinlichkeit: Gewißheit und Wahrheit machen die obere Grenze aus." 166 Compare A.Meinong, Über Möglichkeit, p. 656 ff and p. 666.

MEINONG ON MEANING AND TRUTH

123

the development of a probabilistic approach to uncertain propositions at the beginning of the 20th century, e.g. by Łukasiewicz and Reichenbach. His views on truth have also influenced the development of many-valued logic, in particular by Jan Łukasiewicz, but this is not a part of his own conception. It should be remembered that his theory of degrees of probability attributed to uncertain objectives is related to their epistemically borderline character only. ♦ Truth is absolute, mind-independent and eternal Meinong's understanding of the notion of truth is grounded in the conviction, essential also for Brentano, that truth has an absolute character. Having adopted self-evidence as a substitute criterion of truth as a result of rejecting the traditional conception of correspondence, Meinong faces the problem of defining the relation between a true judgment and the world. His solution, along the lines of logical realism, is briefly characterized above. He also has to deal with the issue of subjectivity of evidence as a truth criterion, which underpins the crucial questions concerning empirical cognition. However, in spite of many difficulties, he formulates a mindindependent notion of truth and treats it as basic for his philosophy. In order to secure a mind-independent understanding of the notion of truth, Meinong proceeds in the opposite direction to late Brentano, that is, he aims at obtaining a non-psychologistic conception. Brentano's theory accepts the individual judgments of particular subjects as bearers of truth, provided that they are identical to the evident judgments of an ideal subject of cognition. Meinong transfers the predication of truth from individual acts of judgment to the objectives intended by these acts which are the immediate objects of their intention. Thus it is possible to speak of the truth of a judgment only with respect to the truth of its objective. Objectives are independent of mental acts of particular subjects and they can be grasped by the acts of different persons. Treating them as truthbearers does not result in predicating truth about mental acts or their subjective contents, but about the objects of their intention, which may be correct or incorrect renderings of the ways things actually are in the world. Yet truth in its absolute understanding is not grounded in the cognitive experience of any subject. The truth of a judgment is determined on

124

ANNA SIERSZULSKA

Meinong's conception by the factuality of its objective, regardless of any cognitive context or any subjective ascription of self-evidence to the judgment in question. The business of life demands of us that in a given situation we know what is and how it is – is in fact, of course. For this to occur, factuality need by no means make a proxy appearance in the apprehending experience, through its subjective correlate, evidence. There may even be something of the nature of excess baggage in evidence, as against which the requirement of truth represents the practical minimum.167

As Meinong indicates here, the inner confirmation through self-evidence of judgments is really not what one should focus upon when speaking about truth. The essential issue here is simply whether the judgment is correct or not, and most of the time no reflection about self-evidence is needed to recognize immediately the factuality of the judged objective. Taking recourse to the property of factuality of objectives allows Meinong to preserve a subject-independent notion of truth. Nevertheless, it can only be potential truth: truth as a capability of a factual objective to be grasped in acts of judgment, where all acts, regardless of the subject, the time, the place and whether they are experienced as self-evident or not, grasp a true objective if they intend an objective that is identical to the factual one. In every case when a factual objective can be identified with the objective of a judgment, this results in the absolute truth of the judgment. Due to this identity of the fact with the objective of the judgment, such truth is nonrelative. Considering this account of true thoughts we can speak about a relative notion of truth in opposition to the absolute notion characterized above.(...) It would be good to regard this sense of the word 'truth' as improper in comparison to its proper sense exposed above. In any case, against the actual practice of the word's usage, we are justified to speak beside 'subjective truth' also about 'objective or nonsubjective truth'.168 167

A.Meinong, On Assumptions, p. 72. A.Meinong, Über Möglichkeit, pp. 42, 43: "Man könnte im Hinblick auf diese Ausgestaltung des Wahrheitsgedankens von einem relativen Wahrheitsbegriff neben dem oben charakterisierten absoluten reden. (...) Man wird immerhin gut daran tun, dieses Sinn des Wortes 'Wahrheit' als uneigentlich dem oben exponierten Sinne als 168

MEINONG ON MEANING AND TRUTH

125

However, the absolute notion of truth is applicable to cognition only to the degree to which factuality is attainable to the cognizing subject. Accepting the absolute notion of truth as basic evokes the objection that this is question begging. In order to find out if a judgment is true, or rather if the objective intended by this judgment can be identified with a factual one, it would be necessary to be able to recognize its factuality independently, without any mediation of the judgment. But objectives can be grasped only as objects of judgments and assumptions. Meinong answers this objection by stating that it is directed not against his own conception in particular, but against most theories of knowledge, especially against theories disregarding immediate knowledge. A judgment which grasps whether another judgment or objective is true, or not, must itself fulfil the demand to be true: if we can only rely upon this judgment, such a grasp remains impossible. As we can see, there is no mention here of evidence; it pertains to the judgement only a potiori, only if it is present as a property of the judgment. This objection directs itself against all who think that every cognition must consist in judging, so not especially against me, perhaps not against me at all (...)169

Meinong's idea of the apprehension of factuality in penetrative self-evident judgments is precisely the opposite of the conviction that all cognition must consist in judging. The moment of factuality is grasped immediately as a kind of intuitively confirmed experience. And although this kind of cognition has a limited scope, Meinong's theory of knowledge does not seem to involve any circularity. Cognition always assumes a possibility of cognition, while the question about the relation between its result and its object is itself subject to dem eigentlichen entgegenzustellen. Jedenfalls aber ist man dem tatsächlich vorliegenden Wortgebrauche gegenüber berechtigt, neben 'subjektiver Wahrheit' auch von 'objektiver oder unsubjektiver Wahrheit' zu reden." 169 A.Meinong, Über Möglichkeit, p. 448: "Das Urteil, das feststellt, ob ein Urteil resp. Objektiv wahr ist oder nicht, muß selbst dem Anspruch genügen, wahr zu sein: ist man ausschließlich auf das Urteil angewiesen, so bleibt also diese Feststellung unmöglich. Von Evidenz ist hier, wie man sieht, noch gar nicht die Rede; sie ist nur a potiori mitbetroffen, sofern sie bloß als Eigenschaft am Urteil auftritt. Direkt richtet sich der Einwurf gegen alle, die jedes Erkennen für urteilen halten, also durchaus nicht im besonderen gegen mich, ja vielleicht überhaupt nicht mehr gegen mich."

126

ANNA SIERSZULSKA

cognition. This leads to a kind of regress. The self-reference of cognition does not allow to decide, above any doubt, that even one instance of it actually takes place. Meinong emphasises, however, that there is no serious reason to doubt a possibility of truth for observational judgments concerning reality. Even if it is a kind of miracle that cannot be explained or accounted for, cognition is a common-place phenomenon and it is a rudimentary fact of epistemology. Yet there is still something in this cognitive apprehending of an actuality that might be called the miracle of epistemology, to vary a well known saying of Schopenhauer, or the basic fact of all knowing, as it undoubtedly better could be called. For this fact there is neither description nor explanation; there is just acceptance, the attitude that is the end we are always thrown back upon, in the face of ultimate facts. I suppose I should not avoid the customary word 'transcendence' in designating this fact; following this usage, the aforementioned basic property of a justified existential affirmation can also be spoken of as that judgment's transcending toward an actuality.170

The fact of a judgment's transcendence is impossible to question. In an act of penetrative judgment, cognition hits the target of the factual objective, as the logical structure underlying a selected fragment of reality is apprehended. If the act is accompanied by the experience of evidence, the judgment also grasps the factuality of its objective, giving the subject a conviction about the truth of his judgment. An apprehension of the objective's factuality is, however, not required, and the truth of the objective is independent from it. The subjectivity of factuality apprehension via self-evidence does not affect the truth of the objective intended in an act of judgment, as Meinong's notion of truth is essentially mind-independent. Cognition is subjective, but truth is not. If a judgment is true, the objective intended in the act of judgment is identical to the factual objective, even when the subject is only able to ascribe a certain degree of probability to it, but not truth. As we gain more knowledge, our assessments of the probability of objectives based upon their subjective confirmation tend to be more accurate, and we can see that our cognition approximates truth in this way.

170

A.Meinong, On Assumptions, p. 161.

MEINONG ON MEANING AND TRUTH

127

PART III MEINONG'S TRUTH IN THE EYES OF HIS CRITICS CHAPTER 5 RUSSELL'S RECEPTION OF MEINONG'S PHILOSOPHY Alexius Meinong, undoubtedly a distinguished philosopher of his time, has been mostly perceived by analytic philosophers as the negative figure of Bertrand Russell's criticism, and only recently has his thought started to evoke some genuine interest, mainly on the part of logicians. Although Russell's strongest criticism concerns Meinong's superfluous ontology, it has also had an influence upon the understanding of his treatment of semantic, or logical, issues and Meinong's attitude towards the question of truth. For this reason, it seems appropriate to present an analysis of Russell's response to Meinong's ideas, before proceeding to discuss more recent comments on his conception of truth. Almost all the later critics refer to Russell's objections and often formulate their views in relation or in opposition to them. There has been, however, little interest in the problems and mutual misunderstandings concerning the notions of proposition and truth, involved in the discussion between these two philosophers, who represent different traditions of thinking, but share surprisingly many goals in their philosophy. Many contemporary analytic philosophers are still, "traditionally", opposed to Meinong. However, there must be a reason that more and more of them find certain ideas in Meinong's philosophy interesting. Naturally, in what follows, the focus will remain upon matters directly or indirectly relevant to the question of truth, while other issues and critics, making comments on other aspects of Meinong's philosophy, will be left aside. As a background for discussing the critical approaches to Meinong's philosophy, an interpretation of

128

ANNA SIERSZULSKA

Meinong's views will be used which is not "post-Russellian". It will be visible in the course of presenting the critical opinions that some of them actually come close to this interpretation at various points. Russell concerns himself with Meinong's conception from Über Annahmen and from Über die Erfahrungsgrundlagen unseres Wissens, in one quite long article "Meinong's Theory of Complexes and Assumptions", in a review of the collection of articles about the theory of objects "Untersuchungen zur Gegenstandstheorie und Psychologie", and in his famous "On Denoting", where he introduces the theory of descriptions as a remedy for Meinong's unacceptable ontology. Meinong tries to answer his objections, particularly in "Über die Stellung der Gegenstandstheorie im System der Wissenschaften" and in some notes to the second edition of Über Annahmen, but unfortunately he does it in a very unconvincing way. In consequence, his point of view is entirely disregarded, Russell being by far a clearer and more expressive writer. In "Meinong's Theory of Complexes and Assumptions" Russell still claims that his own philosophy is close to Meinong's in numerous respects. He attempts to explain Meinong in terms of his own views, which are not always in agreement with the original intentions. Two main objections contained in this paper are: (1) that propositions, whether true or false, must have subsistence – Russell understands false propositions as subsisting entities, which is not the case with Meinong; (2) that Meinong's idea of truth is erroneous because it consists in the correspondence of an immanent object with a transcendent one, which involves an unnecessary duplication of objects and confines truth to the level of judgments, as if it were a subjective psychic phenomenon – this, in turn, is a misunderstanding, Meinong is trying to keep as far as possible from psychologism. Russell acknowledges this later, in the review of the collection mentioned above.171

171

Bertrand Russell, "Review of Meinong and Others, Untersuchungen zur Gegenstandstheorie und Psychologie, " Mind 14 (Oct. 1905): 530-8. The Collected Papers of Bertrand Russell, vol. 4, p. 604.

MEINONG ON MEANING AND TRUTH

129

♦ A difference of opinion concerning the nature of propositions Russell identifies Meinong's objectives with what he calls propositions, but his notion of a proposition from that period differs from Meinong's in an important way. Russell thinks that propositions are complexes, while Meinong distinguishes complexes from objectives. What Russell also does not understand is why, according to Meinong, there cannot be any presentation of an objective as the object of a judgment. He maintains that such a presentation of a complex should be possible. He does not realize that an objective differs from a complex as regards its ontological character, and as regards the ontological status of the objects involved (objectives may concern non-subsisting objects as well as existing ones). Therefore, Meinong's factual objective differs from Russell's complex. The objects in reality, standing in certain relations are treated as constituents of complexes, parts of reality apprehended by means of objectives. Meinong's view on the connection between complexes and objectives is that complexes are only apprehended through objectives, but they are never identical to objectives. At the time of the dispute with Meinong, Russell's view is the opposite. Both judgments and assumptions have reference to what Meinong calls Objectives, which are the propositions concerned... Thus we have (1) simples, which can be presented; (2) complexes, which can be either assumed or judged, but not presented. The point of most importance for logic is, in my opinion, the connection of complexes and propositions.172

The connection between complexes and propositions is obvious on Russell's view, although at the same time it is puzzling. A proposition, the proper truth bearer, is a complex, but the constituents of a complex are neither true nor false, as ordinary objects never are in themselves. What is therefore so special about a complex that we face the question of truth when a complex appears as a proposition? It is this special and apparently indefinable kind of unity which I should propose to employ in characterizing the notion of a complex. The kind of unity in question belongs, as is evident, to all propositions; and the inadequacy of analysis appears, 172

B.Russell, "Meinong's Theory of Complexes and Assumptions", Mind 13, 1904, p. 206.

130

ANNA SIERSZULSKA

in this case, in the fact that propositions are true or false, while their constituents, in general, are neither.173

Complexes and propositions share a certain kind of indefinable unity of their constituents, which binds them into integral single entities and not just aggregates of objects and relations. We may judge truly about the way they are united, or not. This unity is the kind of connection which binds existing objects in reality, so that we can say that they are in a particular way related to each other, and then our proposition-complex is true. Russell's propositions are thus based upon the ontological structure of fragments of reality. It seems surprisingly easy to know this structure, so that it can appear as the unifying factor in propositions. The same objection could be directed at Meinong. The idea that the world and the thought can share their logical structure is somehow present in the writings of both. However, for none of them, it leads to denying the primary role of perception in empirical cognition. Yet, when we see Neptune and when we infer it, there is a wide difference of the two cases. This seems to be accounted for chiefly by two facts: (1) that a proposition perceived is not expressed in words, (2) that it always forms part of an infinitely complex spatiotemporal continuum, and is not, in perception, attended to in isolation.174

Russell claims that a proposition perceived is always a part of a spatiotemporal continuum, as if it were a fragment of the spatiotemporal reality taken out to become the object of our thought. But how can it be possible to apply a true-false distinction to a complex consisting of real objects? Unless we assume that a proposition is the complex as correctly or incorrectly represented by the act's content, which Russell might really mean to say. In the latter case, there would be no subsisting false propositions, because there are no negative or false complexes to be represented, exactly the same fault of which Russell accuses Meinong. Meinong's theory may be modified, (1) by denying his non-subsistent objects, (2) by denying that they do not subsist. I should propose to apply the former process 173

B.Russell, "Meinong's Theory of Complexes and Assumptions", Mind 13, 1904, p. 210 174 B.Russell, "Meinong's Theory of Complexes and Assumptions", p. 216

MEINONG ON MEANING AND TRUTH

131

to the round square, the latter to false propositions. There is, Meinong admits (p.12), one strong argument in favour of the subsistence of the objects which he regards as non-subsistent, and that is, that such objects can be subjects of true and therefore subsistent propositions. But this argument, he says, depends upon regarding a proposition as a complex, and its subject as a constituent of it; and such a view, he thinks, can only be taken figuratively.175

On Meinong's theory there are in fact false propositions – as non-subsisting pure objectives, prior to judgments. An accusation of psychologism concerning falsehood may be dropped, because they are not, after all, merely mental objects. Besides, since Meinong's objective is not a complex, there is no need for negative facts to be found in the world to fit the theory. True negative objectives subsist, but they do not have to be found in the world, it is enough that they are correct about the structure of reality. At this point of the development of Russell's philosophy, his own position in relation to these issues raises more doubts than Meinong's does. What happened later was that Russell abandoned the idea of a proposition as an entity in any ontological sense and treated it as related to the meaning of a sentence. Then, he does not postulate the subsistence of false propositions, for meaning entities need not subsist. It seems that Meinong's attitude to this issue is very similar, even though not articulated enough. At the time of his occupation with Meinong, Russell makes it plain that a complex simply is a proposition, and moreover, it is the product of a proposition. We may suppose that reality would remain a continuum if it were not for propositions, which pick out its fragments. Complexes, as soon as we examine them, are seen to be always products of propositions: one might be tempted to describe them, rather loosely, as propositions in which the truth or falsity has been left out. (...)The other way, which is pursued by Meinong, is to examine our apprehension of complexes, and to show that it contains or involves the apprehension of propositions. The supposed advantage of this method is, I imagine, that our apprehension exists: but it is impossible to apprehend an object unless the object itself is amenable to inspection.176 175

Bertrand Russell, "Review of Meinong and Others, Untersuchungen zur Gegenstandstheorie und Psychologie", Mind 14 (Oct. 1905): 530-8. The Collected Papers of Bertrand Russell, vol. 4, p. 598. 176 B.Russell, "Meinong's Theory of Complexes and Assumptions", p. 346

132

ANNA SIERSZULSKA

Russell does not see that the method Meinong is using, of examining the apprehension of complexes by means of objectives, is not merely a method. On Meinong's view, it is simply impossible to do it the other way round. Complexes can be apprehended only by means of objectives. Complexes themselves are not directly apprehended, because the connections between the objects they consist of have to be apprehended as objectives of judgments or assumptions, and thus complexes cannot be themselves the proper objects of judgments or assumptions. ♦ An objection to Meinong's 'psychologistic' treatment of false propositions Meinong's division of objectives into subsisting and non-subsisting ones results from his conviction that what is false and does not obtain should not possess any kind of ideal being. We should not have a whole sphere of ideal existence filled with false propositions; only the true ones belong to this sphere, together with ideal mathematical objects of different kinds. Russell is of the opposite opinion, because he feels that his propositionscomplexes would have no kind of being at all if deprived of subsistence. For it is plain that logic must concern itself as much with false propositions as with true ones; but false propositions, according to Meinong are the non-subsisting, merely pseudo-existing Objectives of erroneous judgments, and except through erroneous judgments they have no connexion either with existence or with subsistence. (...) But if we deny the distinction between the immanent object and the (so to speak) external object, the positions become reversed. We shall now hold that a presentation or judgment is only capable of being directed to an object because such an object subsists, and that thus the subsistence of the object is prior to the presentation or judgment.177

Since Russell rejects the idea of an immanent object of a judgment, it seems to him that false judgments would have no objects. Only if the subsistence of false propositions is assumed can such judgments have their objects. Apparently, Meinong's conception of the sphere of pure objects of intention does not appeal to Russell. He does not think that the being of such pure, but non-subsisting, objectives would be sufficient for providing 177

B.Russell, "Meinong's Theory of Complexes and Assumptions", p. 353.

MEINONG ON MEANING AND TRUTH

133

judgments with objective propositions. He does not understand what kind of being that would be which is beyond ontology, in a sphere of entities which are abstract but not necessarily ideal, which is precisely what we take linguistic meanings to be. According to Russell, what Meinong proposes amounts to a claim that the objectives of false judgments should be only mental. If a belief may be a content which has no object, then it may be true that, though we believe, there is nothing we believe in; and in this case correct beliefs would be distinguished from erroneous ones by the fact that they have an object, while the others have not. But this possibility seems too paradoxical to be maintained except in last resort (...) false propositions must have some kind of extra-mental subsistence.178

Trying to provide arguments against such a supposedly subjective conception of false propositions, Russell finds an example which makes it look paradoxical. If we take a complex proposition involving a false proposition, then even if it is true, it will have to remain only mental, because one of its constituents is supposed to be only mental. Meinong's theory turns out to be absurd from this point of view, except that Russell does not notice that objectives are never merely mental in the first place, and that the objects involved are not literally their constituents. Further, the proposition "p implies q" may be true though p be false: but in that case, since p is merely mental, the whole proposition will be merely mental, which we supposed true propositions not to be. And so throughout, the attempt to make a difference, as regards subsistence, between true and false propositions, leads to countless difficulties and to countless conflicts with what appear to be obvious facts.179

The final and decisive objection is that Meinong himself follows the principle that inferiora of a relation should be prior to their superius, and he does not recognize that in order for an objective to be apprehended in a judgment, the objective itself must be prior to such a situation taking place.

178 179

B.Russell, "Meinong's Theory of Complexes and Assumptions", p. 219. B.Russell, "Meinong's Theory of Complexes and Assumptions", p. 511

134

ANNA SIERSZULSKA

And Meinong's own principle, that the inferiora are prior to the superius, makes the Objective necessarily prior to the relation involved in judgment.180

This could be an entirely forgivable misunderstanding on Russell's part, but for the fact that the very idea behind Meinong's theory of objects is, above all, to provide a spectrum of pure objects to which the intentionality of mental acts could be directed. These objects are understood as in all circumstances prior to such acts, and capable of being grasped by many subjects at different times. At the same time they do not oblige anyone to any definite ontological commitment, for obviously many people can think about chimeras without believing in them. ♦ Russell's remarks about truth in Meinong's theory Russell's negative attitude to Meinong's views on truth is determined by the accusations of psychologism. When Meinong writes about the adequacy of a presentation, he explains that the content of a presentation of a square table need not be square, but the object presented by the content must be square, in order for the representation of the external object to be correct. Meinong does not say that the object presented by the content is only immanent, because that is not what he means at all. But this is the way Russell understands him, also in relation to objectives. If truth requires the correspondence of ideas with facts, it must be the correspondence of the immanent object of ideas with facts, not of the contents of ideas: for the content of a presentation of a square table is not square. When such correspondence subsists, there is a relation which we may call that of adequacy.181

On Russell's view, such an adequacy does not depend on a relation between any immanent objects of ideas and transcendent objects. A proposition understood as a subsisting complex is transcendent, it is beyond the subject. We may protest that Meinong's objective is basically beyond the subject as well, yet we will have to say that it is much closer to what we take abstract propositions to be than Russell's complex. And one may suppose, that if Russell could have seen this at that time, he would 180 181

B.Russell, "Meinong's Theory of Complexes and Assumptions", p. 354 B.Russell, "Meinong's Theory of Complexes and Assumptions", p. 347.

MEINONG ON MEANING AND TRUTH

135

have rejected Meinong's view all the same, for he would not have been able to accept abstract propositions. At the end of "Meinong's Theory...", Russell compares what can be called his identity theory of truth with Meinong's identity theory. According to Russell, a proposition is a complex of objects and relations, a true proposition is identical with a fact, and a fact is a complex as a part of reality: so we can see that a true proposition is a part of reality. This being a very unsatisfactory theory of truth, most philosophers will be likely to reject Russell's views on this question. The difference, as Russell sees it, between his theory and Meinong's , is supposed to be that for Meinong it is the immanent object of a judgment that is identical with a part of reality. Which would combine Russell's fault with a psychologistic approach to propositions, if it were correct. It may be said – and this is, I believe, the correct view – that there is no problem at all in truth and falsehood; that some propositions are true and some false, just as some roses are red and some white (...) The fundamental objection may be simply expressed by saying that true propositions express fact, while false ones do not. This at once raises the problem: What is a fact? And the difficulty of this problem lies in this, that a fact appears to be merely a true proposition, so that what seemed a significant assertion becomes a tautology. It is very difficult to avoid recurring to the notion that a proposition is a judgment, and it might be thought that this is why the statement that true propositions express facts seems significant. But even when this error has bee avoided, it seems to remain that, when a proposition is false, something does not subsist which would subsist if the proposition were true. (...) ...it is hard to regard A's non-existence, when true, as a fact in quite the same sense in which A's existence would be a fact if it were true. It may be suspected, however, that this apparent difference is not logical, but derived from the nature of perception: all the propositions we perceive are affirmative, and the word fact applies most naturally to propositions which are either perceived or analogous to such as perceived.182

Meinong would certainly object to false subsisting objectives, but that is because what Meinong understands as subsistence seems to differ from what Russell means by it. Russell means some kind of being which is not existence, and thus it may be suitable also for false propositions. Meinong 182

B.Russell, "Meinong's Theory of Complexes and Assumptions", p. 523.

136

ANNA SIERSZULSKA

means ideal being by subsistence. Only what is ideal may possess subsistence in his sense. Nevertheless, Russell's white and red roses have their counterparts in factual and non-factual objectives remaining in the sphere of Aussersein. Therefore, it is of course in total disagreement with Meinong's conception to say that 'a proposition is a judgment'. He would not have developed his theory of assumptions, and his theory of objects in general, if he had thought so. ♦ The ultimate rejection of Meinong's ontological ideas: 'On Denoting' What Russell mainly disapproves of in Meinong's philosophy, is related to logical and ontological questions. He cannot imagine a logic which would incorporate nonexistent objects, for this has to lead to problems and contradictions in a purely extensional theory. A hundred years later no kind of logic is able to surprise anyone, although the opinions on the subject of the 'proper' logic are still very much divided. Meinong would never have gained such a bad name among logicians if he lived today. However, his ideas are still sufficiently misunderstood to secure for him a bad name among ontologists. A reader of "On Denoting" will find out, in the first place, that Meinong invents denotations in such cases where there is no denotation, e.g. for phrases which concern something that does not exist. Moreover, these apparent denotations may consist of incomplete or contradictory objects – Meinong does not observe the law of contradiction in his ontology. Thus we must either provide a denotation in cases in which it is at first sight absent, or we must abandon the view that the denotation is what is concerned in propositions which contain denoting phrases. The latter is the course that I advocate. The former course may be taken, as by Meinong, by admitting objects which do not subsist, and denying that they obey the law of contradiction; this, however, is to be avoided if possible.183

And the reader of "On Denoting" must be very surprised that there might be someone who could develop such an unusual theory. The reason for this 183

Bertrand Russell, The Collected Papers of Bertrand Russell, vol. 4, p. 419-20.

MEINONG ON MEANING AND TRUTH

137

misunderstanding is that what Russell calls denotation would never even cross Meinong's mind in relation to non-subsisting entities. Such expressions as mentioned above can only have objects from the abstract sphere beyond being, which I propose to call 'merely semantic' objects. And to the same sphere belong all contradictory objects. Meinong would never allow them to have a place in any ontology. Because Aussersein is the realm of all intentional entities, without assuming this, it is impossible to accept Meinong's philosophy. Russell falls into his own trap of strict extensional semantics, because he is unable to deny the existence of something nonexistent, but only by applying his theory of descriptions, which is a very artificial device of assigning meaning to referring expressions. The need for such a device results from his semantic approach, on which assuming the subsistence of a nonexistent object leads to contradiction, while assuming its total nonexistence deprives the phrase in question of any meaning. But how can a non-entity be the subject of a proposition? "I think, therefore I am" is no more evident than "I am the subject of a proposition, therefore I am", provided that "I am" is taken to assert subsistence or being, not existence. Hence, it would appear, it must always be self-contradictory to deny the being of anything; but we have seen, in connection with Meinong, that to admit being also sometimes leads to contradiction. Thus if A and B do not differ, to suppose either that there is, or that there is not, such an object as "the difference between A and B" seems equally impossible.184

Thus, on Russell's view, "the difference between A and B" cannot have any meaning if A and B do not differ – the whole phrase is non-denoting and so deprived of meaning, unless his method of descriptions is used. But according to his identity theory of truth, the proposition that "A and B differ", which is false, can have subsistence, while "the lack of difference between A and B" is a fact in the world. At a first glance, some of Russell's ideas are at least as surprising as Meinong's , especially in such a combination, so that we could just as easily imagine Meinong criticising Russell, and not vice versa, even though Russell does not propose any round squares.

184

The Collected Papers of Bertrand Russell, vol 4, p. 420-1.

138

ANNA SIERSZULSKA

The meaning of 'round square' is provided by Russell according to the following pattern. A proposition about Apollo means what we get by substituting what the classical dictionary tells us is meant by Apollo, say "the sun-god". All propositions in which Apollo occurs are to be interpreted by the above rules for denoting phrases. If "Apollo" has a primary occurrence, the proposition containing the occurrence is false; if the occurrence is secondary, the proposition may be true. So again "the round square is round" means "there is one and only one entity x which is round and square, and that entity is round", which is a false proposition, not, as Meinong maintains, a true one.185

It is a different thing, on Meinong's view, to claim that some entity which we conceive of as being round and square at the same time, is round, according to which properties we ascribe to it, and it is still a different thing to claim that there is such an entity, in any other sense than as an abstract object of intention. Russell's analysis in this case is not an analysis of Meinong's idea, because Russell obviously does not mean abstract objects of intention when he writes 'there is', so on his interpretation it is false that a round square is round, since there is no such entity to begin with. It is not so on Meinong's conception. Meinong's principle of independence of so-being from being, which separates predication from the question of existence, has been affirmed by many logicians at present (while remaining unacceptable to many others). Mr. McColl (...) regards individuals as of two sorts, real and unreal; hence he defines the null-class as the class consisting of all unreal individuals. This assumes that such phrases as "the present King of France", which do not denote a real individual, do, nevertheless, denote an individual, but an unreal one. (reference is not denotation) This is essentially Meinong's theory, which we have seen reason to reject because it conflicts with the law of contradiction. With our theory of denoting, we are able to hold that there are no unreal individuals; so that the nullclass is the class containing no members, not the class containing as members all unreal individuals.186

It is difficult to tell what Meinong would say about a null-class containing unreal individuals. Probably he would not consider his unreal objects as 185 186

The Collected Papers of Bertrand Russell, vol 4, p. 425-6. The Collected Papers of Bertrand Russell, vol. 4, p. 426.

MEINONG ON MEANING AND TRUTH

139

members of a null-class, even though they possess no existence. His logic, with all its entities, might be more on the side of language than on the side of the world, contrary to what Russell attempted to do. In any case, he would certainly say that there is a non-subsisting object 'the present King of France', which is the object of intentional reference of everyone who thinks about the present King of France. And again, this has nothing to do with denotation. Intentional reference is a way to handle the fact that many people may talk and judge about such an entity as 'the present King of France'. The fact that nonexistent objects appear in our language and that they have meaning for us has always been of some interest for logicians. Therefore, at least in this respect, Meinong's cause defends itself simply enough. The situation is worse with his views upon the nature of a proposition and truth, which have become seriously distorted in their interpretation by Russell's critical reformulation of them.

Appendix to Chapter 5: MEINONG'S DEFENCE AGAINST RUSSELL'S OBJECTIONS ♦ Meinong on the famous paradox The attacks in "On Denoting" are mostly of an ontological nature, and some of Russell's remarks in the review of the collection of papers by Meinong and his pupils "Untersuchungen zur Gegenstandstheorie and Psychologie", also concern nonexistent objects. Since these issues are not primarily investigated here, I will only quote Meinong's most famous reply to this criticism. A reply which because of its ambiguity is generally understood in a completely opposite way to Meinong's intention. The discussion concerns Russell's questioning of Meinong's separation between the nature and the existence of objects, which is also called the principle of independence of so-being from being. Russell points out that we could, therefore, say that "the existing round square exists" and it would be true, which can never be accepted.187 Meinong answers:

187

"Review of Meinong and Others...", The Coll.P., vol. 4, p. 599.

140

ANNA SIERSZULSKA

It would be much worse if the new discipline, as B. Russell still seems to worry, were compelled to ascribe existence or subsistence to these impossible objects. The objection is based upon the validity of sentences such as that the existing round square 'exists', which seem to admit explicitly that among round squares there are also such that possess existence. (…) The difficulty lies much more in the predication of existence. By building the participle 'existing', one is actually in the position to ascribe existence to a formal object, in the same way that one can ascribe to it any property of the Sosein.188

Meinong is right that the problem appears because claiming existence differs from predicating other properties. However, instead of rejecting the idea that existence can be predicated as a property of the Sosein at all, i. e. that 'existence' is an ordinary property, he decides that there are two kinds of existence-claims, one which actually claims the existence of an object, by way of an objective of the type "x exists", and the other, like in the case of 'the existing golden mountain', which places the property of 'being existent' in the Sosein of the object. Meinong also distinguishes between constitutive and non-constitutive properties, while in the formulation of the paradox the property 'being existent' appears in both roles. From this indeed it follows that 'the existent golden mountain is existent' is a true proposition. It does not follow that the golden mountain exists for this reason, but it is rather understandable that nobody wants to consider such intricacies on hearing something that sounds so strange... This is one of the so called paradoxes of the Sosein. On a solution which is close to Meinong's two kinds of existence-claims, two kinds of predication are distinguished: exemplifying and mere encoding of properties (see Ch.9).

188

A.Meinong, Über die Stellung der Gegenstandstheorie im System der Wissenschaften, Gesamtausgabe, vol. V, pp. 222-3: "Schlimmer wäre es freilich, wenn die neue Betrachtungsweise, wie B. Russell weiter zu besorgen scheint, nun auch noch dazu zwänge, diesen unmöglichen Gegenständen Existenz resp. Bestand anzusprechen. Der Einwurf gründet sich auf die Gültigkeit von Sätzen wie etwa, dass das existierende runde Viereck "existiert", worin ja ausdrücklich anerkannt scheint, dass es unter den runden Vierecken auch solche gibt, denen Existenz zukommt. (...) Die Schwierigkeit liegt vielmehr in der Existentialpradikation. Indem man das Partizip "existierend" oder dergleichen bildet, gelang man wirklich in die Lage, einem Objekt formell ganz ebenso Existenz nachzusagen wie man ihm sonst ein Soseinspradikat nachsagt."

MEINONG ON MEANING AND TRUTH

141

♦ Meinong's notes in the 2nd edition of On Assumptions In the 2nd edition of Über Annahmen, Meinong tries to amend his work and defend his ideas against different critical remarks which are expressed by various philosophers of his time. These criticisms are historically interesting, but they cannot be all discussed here. Besides, none of the other critics of that time has had a comparable influence on the later interpretation of Meinong's works to that exerted by Bertrand Russell. Meinong refers to his criticisms directly several times, both in the footnotes and in the main text of the 2nd edition. We will have a closer look at these notes, since they include clues about how Meinong's conception differs from Russell's and how it should be understood in comparison to it. The issue of the proper object of an act of judgment emerges at the very beginning of the book, and because Russell criticises Meinong by claiming that his proper object of judgment is not a proposition, Meinong admits in a footnote that his previous rendering of this problem could indeed cause a confusion.189 This time he tries to explain more clearly that he speaks about the object of a judgment in two different senses. If I look at the snow-covered street and say, "There is snow outside", then 'snow' is the representational object, the objectum of this instance of knowing. But also "that there is snow" is its objective, even though the latter does not have that inherent opposition to the objectum that made the objective so easily noticeable in the instance of negative knowing. (...) The frequently employed designation 'intentional object', which is not entirely without its dangers, could not be applied in a judgmental context more effortlessly than when it is applied to a judgment's objectum. And yet this object has been offered to the judgment merely from the outside, and to that extent it does not belong to the judgment in the way that the objective does; the judgment is inherently turned to the objective, as the representation is to the objectum. Basically, the judgment apprehends the objectum through the objective; so that in the objective we have the true judgmental object, despite the objectum's obviously commanding position in our interests.190

We can see that, in one sense, the object of a judgment is the transcendent physical object the judgment is concerned with, in Meinong's terminology 189 190

A. Meinong, On Assumptions, Notes to Chapter III, note 8. A. Meinong, On Assumptions, pp.38-39.

142

ANNA SIERSZULSKA

it is called a judgment's objectum. Meinong says that this object is offered to the judgment from the outside, it does not really belong to it, because it is a part of the outside world which the judgment apprehends. In another, and more proper sense, an objective is the object of a judgment. The judgment apprehends the objectum through the objective, which means that the objective is the object closest to the content of the judgment. In other words, what the content of the judgment intends is a proposition, and the proposition involves a representation of the objectum. Russell does not understand this distinction of Meinong's , because for him the whole proposition, as a complex, could be represented, and the whole proposition, if true, is identical to the external object of a judgment: the fact. Meinong's proposition is of a different character, which makes Russell suspect that perhaps it is merely mental. Meinong strongly opposes such an interpretation of his views. If it is just a question of whether I did or did not ever regard the object as being a part of the representational experience, or even (qua object) as being merely mental, the belief that I did is a misconception (e.g., Bertrand Russell's , cf. 'Meinong's Theory..."); nevertheless, this misconception is one which may have been incurred by me. It was possibly occasioned by the fact that I did not discontinue the use of the old term "immanent object" [immanentes Objekt]. Still, I have never conceived of this object [Objekt] otherwise than as 'pseudoexistent'.191

He admits that he should not have used the old term 'immanent object', because it suggests that he is talking about something that is only immanent. He uses this term to describe the object as it is apprehended, and as it pseudo-exists in the mind, in opposition to the full richness of the ultimate object of intention. But on his view, the apprehended object is implected in the ultimate object as regards all its properties, so the ultimate object is a kind of limiting case for the apprehension. If it is an apprehension of an existent object, the ultimate object can never be grasped completely. The immanent object, as well as every object or objective that can be apprehended, belongs to the sphere beyond being of pure objects of intention. That is where it can be found, and can be intended by many subjects. Even if it does not possess either existence or 191

A. Meinong, On Assumptions, Notes to Chapter III, note 76.

MEINONG ON MEANING AND TRUTH

143

subsistence, it is never merely immanent. Therefore, Meinong 'agrees' with Russell that objectives concerning the non-existential sphere are equally legitimate. Which is not quite what Russell has in mind. Bertrand Russell (Meinong's Theory...) rightly mentions the possibility that more than the existential sphere might be taken in. Naturally, the domain of the objective is not abandoned in any of these extensions of the sphere of consideration.192

Russell does not mean that it is important to have subsistent true or false propositions about the sphere of nonexistent entities. He only means that it is important to have subsistent false propositions – in his understanding of subsistence (non-ideal) – but about the sphere of existent entities, for he naturally does not accept any nonexistent entities. According to Meinong, neither the first nor the second type of false propositions could have subsistence, i.e. the ideal mode of being. But both these types of propositions belong to Aussersein and they are not only immanent. The other major problem, to which Meinong is trying to react in the remarks that will be quoted below, is that Russell believes that objectives should be related to representations of them in the mind. In particular, he thinks that assumptions are just representations of propositions. This stands in opposition to Meinong's idea that an assumption is a different kind of act apprehending an objective, parallel to an act of judgment. It cannot be a representation of an objective, or proposition, because the nature of Meinong's proposition is such that it cannot be the object of a representation. This is why he has to introduce an assumption, which grasps the objective in a way similar to the way a judgment grasps it. Perhaps it will foster clarity of insight into the facts if I contrast the above delineation of them with Bertrand Russell's view. As far as words go, Russell indeed shares the position expounded here, that objectives can be apprehended by means of assumptions; but as far as the fact of the matter is concerned, he contests this position. For he also believes that assumptions themselves are nothing but representations and that their special character consists only in the fact that they are representations of objectives.193

192 193

A. Meinong, On Assumptions, Notes to Chapter V, note 60. A. Meinong, On Assumptions, p. 99.

144

ANNA SIERSZULSKA

Russell prefers to say 'complexes', he thinks that complexes coincide with objectives.194

Meinong's note makes it clear that there is a difference between complexes and objectives, contrary to what Russell maintains. But complexes could not be objects of representations either. According to Meinong, as they are only apprehended by means of objectives. Meinong states this very explicitly that an objective consists in the integrating structure of a complex. Due to a common, shared, structure of the objective and the complex, the objective can apprehend the complex. But if for the present we are still lacking in more precise knowledge of how assuming or judging begins to accomplish what representation alone is not able to accomplish, then the following can be taken as a very welcome confirmation of the view I have set forth. As Bertrand Russell has justly pointed out, one can arrive at the same result through a simple deliberation conducted exclusively from the standpoint of the theory of objects. (...)195

As a proposition, a true objective apprehends correctly the way objects are related. However, it is not possible to represent an objective, even though it is possible to grasp it. Meinong tries to explain this by showing what happens when we apprehend an objective of the existential type: "A is". Thus, even if I must without fail represent an A in order to apprehend an objective, that is certainly no reason why I should have to apprehend the object 'being' once again by means of representation (which would of course have to be another representing than that of A), and there is always the question as to whether I could apprehend it by means of representation at all.196

We may wonder why, if an objective cannot be represented in any way, it is all the same called an object, and even the object of a judgment. Russell's proposition, which is a complex, may be taken to be an object of some kind, but Meinong's 'integrating factor' to which the objective is identical, could hardly be understood as an object in the same way. It will probably become clearer when we consider a certain remark of Meinong about what can be called an object on his theory. 194

A. Meinong, On Assumptions, Notes to Ch.V, note 3. A. Meinong, On Assumptions, pp.201-2. 196 A. Meinong, On Assumptions, p. 101. 195

MEINONG ON MEANING AND TRUTH

145

The fact that contents appear here as objects could well puzzle someone who is unpracticed in such investigations. But this fact constitutes neither an error nor a difficulty. For anything apprehensible is an object, and contents in particular are therefore objects. This will be especially clear in the present connection, where what we have to do, of course, is to occupy ourselves with the object 'content' for a while and therefore apprehend it and make judgments about it.197

Thus, on Meinong's theory even contents may be considered as objects of some kind, because anything that can be judged about is an object. Objectives are not merely contents, for no contents as such would be able to possess subsistence, but they are objects, as one could say, of a 'semantic' character. A true objective is a sense-structure which obtains about the world, and a false objective is a sense-structure which does not obtain about the world. Both can be grasped either by judgments or by assumptions, but not by means of representations. Russell's account of assumptions is different, since he does not see Meinong's point here. As I see it, the whole course of these investigations suggests unequivocally that there is a far closer affinity between assumptions and judgments than there is between assumptions and representations. As paradoxical as it may sound, it really makes quite good sense to say this, that an assumption is a judgment without conviction; whereas there would be no intelligible sense at all in defining an assumption as, say, a representation that is determinate in respect of the antithesis of yes and no.198 The fact that Bertrand Russell judges to the contrary ("Meinong's Theory...") is traceable to a view that I have attempted to refute above...199

Nevertheless, Meinong does not succeed in refuting Russell's view, as he does not manage to make it sufficiently explicit and clear that he means by a proposition a completely different sort of entity, than Russell does. And the reason is, that when he writes about the identity of an apprehended objective with a factual one in a true judgment, the reader assumes that he is really speaking about a complex of the Russellian type. Another problem concerning Meinong's attitude to truth which Russell points out is related to assigning to truth a secondary role with respect to 197

A. Meinong, On Assumptions, Notes to Ch. VIII, note 20. A. Meinong, On Assumptions, pp.262-3. 199 A. Meinong, On Assumptions, Notes to Ch. X, note 35. 198

146

ANNA SIERSZULSKA

factuality of objectives. If an objective is true, it must be factual, and only those objectives which are facts can be true at all. Factuality does not depend on anyone's knowledge, it is independent and transcendent to the subject of cognition. Meinong does not think at all that the situation with truth is different, but the apprehension of truth, the process of cognition, is already subjective. It is exactly for this reason that he thinks about factuality as the primary property of true objectives. Strictly speaking, truth by no means constitutes a new and peculiar modal property of an objective – least of all a property having that fundamental epistemological significance that is usually attributed to the factor of truth.200 Also by Bertrand Russell, it would seem ("Meinong's Theory..."), though he might of course say in reply that it is easy to solve problems concerning truth when one has first heaped up all the difficulties into the concept of factuality. Perhaps, though, I can hope in that way to get a bit closer to the nature of these 'last things' of epistemology.201

Apparently, Meinong has doubts whether truth is really of fundamental epistemological importance. As he claims himself, he solved the problem of truth by describing it as factuality of apprehended objectives. He does not entirely identify truth with factuality only because he wants to bind truth exclusively to the acts of judgment, emphasising its close relationship with the process of cognition. One may assume quite legitimately, though, that truth simply coincides with factuality on his theory. It may seem surprising that this does not turn his conception into an ontological theory of truth. But while in an ontological theory, as of Thomas Aquinas, truth coincides with what exists in the world, Meinong's true objectives do not exist as fragments of reality. Their nature resembles a logical structure, and it seems to me that it would be proper to treat them ontologically in the same way as subsisting logical truths. If someone wants to uncover an empirical truth, he wants to find something similar to the answer whether a certain logical structure is identical to the structuring of a certain fragment of reality. Russell's reception of Meinong's philosophy precludes such an interpretation of his approach to truth. 200 201

A. Meinong, On Assumptions, p. 72. A. Meinong, On Assumptions, Notes to Ch. III, note 96.

MEINONG ON MEANING AND TRUTH

147

CHAPTER 6 AGAINST THE POST-RUSSELLIAN VIEW OF MEINONG'S PROPOSITIONS ♦ Objectives as objects – Findlay's exposition of Meinong's philosophy Findlay's book Meinong's Theory of Objects and Values is a classic source concerning Meinong's philosophy. Findlay's was the first entire book on Meinong's philosophy, and also the first source in English, after Russell, which made Meinong's thought available to the English speaking philosophical world. Findlay is very influenced by Russell's criticism of Meinong. When giving a short account of the Meinong – Russell debate, he agrees with Russell about the nature of Meinong's objectives. He is very reluctant to identify them with abstract propositions, both for reasons presented by Russell in 1904 and for reasons that might be given from the point of view of later Russell. He disregards completely the radical change in Russell's views on propositions and speaks about Russell's propositions as if they had always been meaning entities related to sentences. Since at the time of the dispute with Meinong they were complexes, it was natural for Russell to identify true propositions with facts and extremely strange that he ascribed being to false propositions. Later only facts remain complexes for Russell and such is Findlay's general understanding of a fact, quite in disagreement with Meinong's conception of factuality. Now Meinong's objectives and the Russell-Moore propositions of 1904 have one thing in common, they are the objects to which judgments are directed; in other respects they are totally different. In the first place we may note that the propositions of Russell all have being whether they are true or false. ... It is clear that Meinong's objectives differ completely from Russell's propositions, for with them it is not the case that their falsity implies their being. Objectives may be false, and have a great number of other essential and accidental properties, without either existing or subsisting.

148

ANNA SIERSZULSKA

There is a more serious objection to the identification of an objective with a proposition: it is extremely unnatural to say that any propositions are facts, and on Meinong's theory some objectives are facts. It is true that Mr. Russell does in one passage identify true propositions with facts.... It must be conceded, however, that hardly anyone would say that a fact simply was a true proposition; to those people who believe in propositions they are essentially Zwischendinge, entities which correspond to or describe facts when true, and which, when false, fail to do this in some way.202

Except for some followers of Russell's early thought and the advocates of a Fregean identity theory of truth, not many philosophers would say that a true proposition is a fact. Meinong belongs in this group as he indeed identifies a true proposition with a fact, but his proposition is really a meaning entity, not a complex, and his fact is what is correct, or a way things are, not a real complex either. An interpretation of the above statement depends on what we understand as a fact. For a Russellian, a fact is a fragment of reality, for a deflationistic identity theorist, a fact is what is asserted, for Meinong a fact is neither a fragment of reality nor merely what is asserted. Logical realism is also Meinong's doctrine in as much as he claims that thought and reality can share a common logical structure. Hence Meinong's theory of truth and falsehood is simply one of identity; the pseudo-existent objective of an experience is true if it is also factual, false if it is unfactual. Truth and falsehood, in spite of their august associations, are therefore properties of objectives which play a very unimportant and derivative part in the theory of objects; they are completely overshadowed in importance by the fundamental distinction between factuality and unfactuality. Since Meinong attributes to his objectives so many properties that no one would dream of attributing to propositions, and since the proposition-theory of truth is so totally different from Meinong's theory, it can only lead to confusion if we identify objectives with propositions.203

Findlay emphasises that Meinong's theory of truth is that of identity and the most important property of objectives is their being factual. This, together with the Russellian understanding of a fact, suggests to many readers that Meinong's notion of truth is ontological, that on his conception truth coincides with what exists. The main reason to object to such a claim 202 203

J.N.Findlay, Meinong's Theory of Objects and Values, p. 84. J.N.Findlay, Meinong's Theory of Objects and Values, p. 88.

MEINONG ON MEANING AND TRUTH

149

is derived from the character of Meinong's proposition, which is a kind of sense entity, an abstract structure that obtains in reality. We may say that truth coincides with the ideal subsistence of the obtaining objectives, but the subsistence of what is true does not make this conception of truth ontological. Findlay's interpretation of Meinong goes in the direction of a strongly ontological conception. He even suggests substituting the term 'state of affairs' (used to refer to objectives), for the word 'circumstance'. The merit would be, according to him, that it is also possible to talk about fictitious 'circumstances', while the term implies neither factuality nor anything merely mental. A circumstance is something that can be but need not be in the world. That the diagonals of a square are unequal or that whales are not mammals, would hardly be called Sachverhalte though they are objectives. ... On the other hand, the word 'circumstance' is suitable from most points of view; there are fictitious circumstances as well as actual (i.e. factual) circumstances. ... Of course one cannot say that one believes a circumstance or assumes it, or that a circumstance is true or false, but this is nothing but sheer gain, for it rids us of the temptation to psychologize objects.204

Objectives as circumstances would be simply something that either is or is not in the world. In particular it is obvious that they would have nothing to do with meanings, unless meaning is treated as denotation. And such is indeed Findlay's attitude, for he claims that Meinong's meaning is a twoelement word-object relation. Therefore, he also thinks that sentences denote objects in reality called objectives, or that maybe these objectives are created for sentences to refer to and they are not entities in their own right. It may of course be argued against Meinong that such entities as objectives are mere projections of language, and that their very projection arises out of a misinterpretation of language, the mistaken view that a sentence stands for something, that it is a name and not an assertion. (...) And while there are situations in which objectives are merely used to illuminate or describe first-order

204

J.N.Findlay, Meinong's Theory of Objects and Values, p. 90.

150

ANNA SIERSZULSKA

objects, there are as many situations in which objects are merely used to illustrate or to fulfil certain sorts of objectives.205

In one sense Findlay is close here to what in fact is the case with Meinong's theory of objectives, namely that it is based upon the logical structure of language. But he is wrong when he thinks that objectives are taken to be objects belonging to reality. They are more like logical structures involving real objects, which are found to obtain in the world. Such sense-related entities are hardly what any ontology would regard as genuine objects, yet Meinong's use of the notion 'object' does not primarily belong to ontology, it is also a psychological and semantic term. Findlay is well aware of the connection between language and consciousness, on one side, and the theory of objectives on the other. For him, this confirms the view that the theory of objects in general is a phenomenological study of objects of consciousness, which are indiscriminately and sometimes falsely given the status of genuine entities in Meinong's ontology. It may now be held, paradoxically enough, that what Meinong really brought out was not the ordered hierarchical structure of a realm of objects having a status aloof from our conscious references, but rather the structure of objects as they appear in and for our conception or our language, or, in the widest sense, our 'experience'. Meinong was in effect, we may suggest, a true empiricist and phenomenologist (...).206 Husserl, despite his constant stress on appearance, may be called a realistic subjectivist, even a crypto-solipsist, while Meinong, despite all his so called logical realism, is the true phenomenologist, the man concerned to describe what is before consciousness, rather than to construct or reconstruct it.207

Apart from being a 'phenomenologist', however, Meinong always remains a realist, which is not the case with some phenomenologists, including late Husserl, whom Findlay calls a crypto-solipsist. Meinong, to the contrary, is an empiricist for Findlay, mostly in the sense of investigating the phenomena of consciousness, but Findlay also admits that his philosophy is more on the side of the realistic approach to cognition. At the same time, 205

J.N.Findlay, Meinong's Theory of Objects and Values, p. 333-4. J.N.Findlay, Meinong's Theory of Objects and Values, p. 331. 207 J.N.Findlay, Meinong's Theory of Objects and Values, p. 332. 206

MEINONG ON MEANING AND TRUTH

151

Findlay considers his logical realism to be artificial, i.e. the idea that reality and thought can share a common structure. Yet, whether right or wrong, it is precisely this idea that lies at the bottom of Meinong's epistemological realism, his theory of truth and his theory of objectives. Therefore, the question arises what the difference is between Russell's logical realism and Meinong's ? The main difference lies in Meinong's conception of factuality, which does not postulate 'objectual' facts, while simultaneously Meinong is much less reductionistic in his attitude to abstract objects of higher order, such as objectives in the sphere of meaning. He does not think that they could be brought down each time to simple logical constructions and thus dispensed with. Findlay makes the following remark to praise Meinong with respect to this issue: Meinong may, however, be said to steer bravely clear of one of the pitfalls of the traditional empiricism: the doctrine that 'while there are in a sense all these higher order objects, all these entities of reason, yet ultimately they can be dispensed with. In describing the furniture of the lowest level, we could with some complication say all there is to say in terms of higher order entities.' This belief that everything is really to be found on the lowest rung of the ladder of logical types inspires Russell's doctrine of logical constructions, as it inspires nominalistic schemes like those of Quine and Goodman. In a remote way it also is the inspiration of Wittgenstein's language-games, where the suggestion is that by examining how people operate with words in quite simple situations, we shall be able to see around, and to remove puzzlement from, their more sophisticated and difficult utterances.208

Meinong's theory does not deprive the world of its intricacy and, simultaneously, it allows for unlimited complication in our expression and understanding of its structure. Every intelligible sense, concerning real, ideal or non-subsisting objects, that can be expressed by mental content, intends a legitimate objective. And there is certainly nothing wrong about having as many propositions as we can think of, as long as they are not supposed to be entities which all possess some kind of existence at the same time. Whoever would think that it is so, will certainly be likely to regard Meinong's theory of objectives as a naive projection of the intentionality relation. Findlay is a representative of this tendency in the passage below. 208

J.N.Findlay, Meinong's Theory of Objects and Values, p. 334.

152

ANNA SIERSZULSKA

He could only conceive intentional references in terms of objects logically prior to them, on which they necessarily depended: hence the many absurdities of his theory of objects.209

Findlay maintains that Meinong provides us with a surplus of objects and objectives of the most absurd sorts, as 'relata' to which our intentional states may be directed. Because he cannot imagine how intentionality would be possible without such artificial objects, ready and waiting for the mind to grasp them. This absurdity claim is a consequence of the way Findlay perceives Meinong's theory of meaning, as a post-Russellian wordobject pattern. Actually, the whole Meinongian Aussersein is a realm of all possible meaning-objects, including objectives, and there is nothing too absurd about meanings being contradictory or related to fictitious things, if such can be expressed in human languages. Another direction of criticism, also initiated by Russell, and concerning objectives in particular, is related to the idea that factuality prevails over truth in Meinong's theory. Since truth appears, according to Meinong, only in epistemic contexts – the question about truth is asked only in relation to cognition – he needs another property for objectives, which would belong to them timelessly and which would distinguish those that obtain about reality from those that ascribe to reality an incorrect structure. Like Russell, Findlay is very surprised by such a solution. Factuality occupies the central place in Meinong's list of modal properties, ousting the property generally called truth. (...) This is a property which an objective can possess in its own right at all times, whereas truth can only belong to it if it is apprehended by someone.210 Factuality is to objectives what subsistence and existence are to objecta; the being peculiar to objectives is being-the-case. (...) An unfactual objective is wholly without being of any sort, but it is still an objective which concerns certain objecta in certain ways, and which is indifferent to our apprehension.211

209

J.N.Findlay, Meinong's Theory of Objects and Values, p. 344. J.N.Findlay, Meinong's Theory of Objects and Values, p. 186. 211 J.N.Findlay, Meinong's Theory of Objects and Values, p. 187. 210

MEINONG ON MEANING AND TRUTH

153

Indeed, this is one of the strangest things in Meinong's theory, for it would have been much simpler to say that factual objectives are true timelessly. Yet Findlay gives justice to the fact that what is characteristic for objectives is not existence in reality, but being the case. Then, factuality is the same thing as being the case of a proposition, and so we can see that it is the same thing as being true. It requires some effort to observe, against presuppositions, that there is no room for Russell's complex here. However, it is not hard to notice that objectives are treated like propositions of a more abstract kind than complexes, since they can be premises and conclusions of reasonings. Such reasonings aim at confirmation of the truth of certain objectives by applying the criterion of coherence. Findlay mentions the application of such a criterion by Meinong, emphasising that, nevertheless, he does not propose a coherence theory of truth. The function of criticism is to examine the apparently evident and see whether it is really evident; it is in this connection that the criterion of coherence is useful, which some philosophers have identified with truth. (...) The elaboration of a symbolic system, the deduction of objectives from the minimum number of independent postulates, is immeasurably valuable precisely because it enables us to see which of our judgments have questionable evidence.212

Although Findlay is, on the whole, convinced about Meinong's sincerity as a realist in the matters of cognition and his attitude to Meinong's theory is favourable, he considers it, at the same time, to be extremely old-fashioned in many ways. One of the major problems is his complete misunderstanding of Meinong's conception of the sphere of pure objects. This misunderstanding leaves Findlay very puzzled, as well as several generations of Findlay's readers, about the most interesting idea in Meinong's conception. But to do what Meinong does, and merely to set the non-existent, false and absurd alongside the actual and the true, to treat them like the white and red roses of Russell's famous comparison, with nothing to explain the special prerogative of

212

J.N.Findlay, Meinong's Theory of Objects and Values, p. 255.

154

ANNA SIERSZULSKA

being real or being the case, is the very type of a dead, surd, unintelligible mode of conception.213

Russell has to withdraw from the theory ascribing being to false propositions that possess it ideally, because his theory of meaning is based upon extensional semantics. For the same reason he cannot tolerate nonexistent objects. A post-Russellian view upon these issues has to reject such ideas. But Meinong's conception does not really involve a whole range of ideally existing falsehoods. Meinong proposes a sphere of 'semantic' objects, considered independently of any existential assumptions. From the fact that they can be meant or intended, nothing about their ontological status follows. The post-Russellian view sees an advantage also in keeping a sharp distinction between matters of fact and the various ways to talk about them. Meinong's objectives as 'facts' are thought to be defective in this respect, since they display language and perspective sensitivity. The same fact in the world can be apprehended by means of different true and factual objectives. Thus, there is no uniform way to understand facts in reality, which raises objections as in this quotation from Findlay: Such quasi-empiricism has in recent times been greatly criticized: it has been held to have erred in failing to keep apart questions of language and conception, on the one hand, and questions of empirical fact, on the other, the ways in which we can variably talk about things and the way they are found to be.214

This is how Findlay sums up Meinong's attempts to remain a realist at the end of his exposition of Meinong's philosophy. Looking from the point of view of post-Russellian theories of truth, any language sensitivity of the structure of 'facts' implies an idealistic approach. But for Meinong, there are no facts as chunks of reality. So every objective, in every language or conception, is true if it is correct about reality as to the structure it expresses. Meinong already knows, in a way, that it is not possible to separate apprehension of reality from conceptions, and he undertakes the enterprise to preserve epistemological realism in spite of the conception dependence of our apprehension of the world. 213 214

J.N.Findlay, Meinong's Theory of Objects and Values, p. 342. J.N.Findlay, Meinong's Theory of Objects and Values, p. 347.

MEINONG ON MEANING AND TRUTH

155

♦ Objectives as states of affairs Due to the object-like status of objectives in Meinong's theory, it has always seemed to be a plausible translation of the term 'objective' to call it a 'state of affairs'. However, if Meinong had thought that it was the proper way to see objectives, he would have used the term 'Sachverhalt' himself. He abstains from doing so, because a state of affairs is something that is best perceived as positively present in reality. Objectives, on the other hand, are very often negative or false. This problem does not discourage thinkers who are more liberal in matters of ontology, and who are ready to admit an infinite number of non-factual states of affairs, like Roderick Chisholm. Chisholm would prefer to endow false states of affaires with ideal being, rather than say, after Meinong, that false objectives have no being whatsoever, but they just belong to the pure objects of Aussersein. Perhaps the most natural English term for Meinong's objectives is 'state of affairs' (...) I would say that there are states of affairs, some of which obtain and some of which do not obtain. But Meinong, with his doctrine of Aussersein, puts the matter differently. (...)And he uses 'is' where I have used 'obtains': objectives are of two sorts, those which, like the being of horses, are such that there are those objectives, and those which, like the being of unicorns, are such that there are not those objectives.215

Aussersein, the sphere of pure objects beyond ontology, gives false objectives a different status as opposed to the subsisting true ones. False objectives can be meant, intended, spoken about by many people, and yet they are not to be found among objects which possess being. This solution is certainly somehow inspired by the ontological doubts of Meinong's teacher Brentano, whose attitude Chisholm rightly compares to Russell's later negative view on whether there are any false propositions that possess being. Russell's attitude toward Meinong's theory – in 1918 – was very much like that of Brentano. "Time was", he wrote, "when I thought there were propositions, but it does not seem to me very plausible to say that in addition to facts there are also these curious shadowy things going about such as 'That today is Wednesday' when 215

Roderick M. Chisholm, "Homeless Objects", in: Brentano and Meinong Studies, p. 46-7.

156

ANNA SIERSZULSKA

in fact it is Tuesday...To suppose that in the actual world of nature there is a whole set of false propositions going about is to my mind monstrous."216 Russell's misgivings, at least in the passage quoted, are concerned only with those objectives which, according to Meinong, are such that there are no such objectives – those objectives that Russell called 'false propositions.'217

Russell's opinion quoted here is opposite to Chisholm's , which does not worry Chisholm in the least, since his understanding of states of affairs seems to be less concrete than what Russell thought that propositions would have to be, if there were such entities at all. Brentano's objections are much more serious for Chisholm, since they result in a situation when it is impossible to classify objectives as objects in any way that would agree with the traditional treatment of all objects in general. Brentano's objections suggest a dilemma: "If you say that objectives are abstracta, then you must have a view about the kinds of things that exemplify or instantiate them. How many things, if any, exemplify that objective which is there being no round squares? If you say that objectives are concreta, or individual things, then you must have a view about where they are. Is there being no round squares an objective which is to be found everywhere, or nowhere? If it's nowhere, how can you say it is a concretum? (...) And the incomplete objects, which we will discuss below, would also seem to be such that they are neither abstract nor concrete.218

This appears to be a real dilemma for Chisholm and he does not think that there is a good escape for objectives from such an intermediate status between abstract and particular objects. If objectives were particular objects, they would have to belong to the world in the way that factual states of affairs are supposed to belong to the world. But what about a negative factual objective, like "that trains do not fly"? Objectives can be conceived as objects of a different kind than states of affairs. Chisholm, however, presents another point of view. He claims that this is a general feature of Meinong's theory that objectives and also incomplete objects are such, that they are neither to be considered as particulars nor as abstracts.

216

Quote from: B.Russell, "The Philosophy of Logical Atomism" (1918) Roderick M. Chisholm, "Homeless Objects", p. 47-8. 218 Roderick M. Chisholm, "Homeless Objects", p. 48. 217

MEINONG ON MEANING AND TRUTH

157

Thus he says that the incomplete billiard ball, that object that is billiard ball as such, exists in every particular billiard ball. To be sure, he adds, no one of the actual complete billiard balls has the incomplete one as a part; for the incomplete billiard ball "would have to lose its incompleteness if it were to become a part of the complete billiard ball." (UM, p 211). But the incomplete one is involved or 'implicated' (implektiert) in each complete one.219

We know from Meinong that an incomplete object is not an existent particular object, for it would have to be complete in order to exist. And it is also true that it is not a universal, either, because Meinong simply does not believe in universals. Does it follow, therefore, that an incomplete object is not an abstract object of any sort? For one thing, it is not a proper abstract object, such that would possess ideal being, like mathematical objects. On Meinong's theory of objects, the incomplete object shares with the complete one in which it is implected, if there is such, some of its properties, but never its ontological status. In spite of that, it can be identified with the complete one and the existing object may be intended, or meant, through the incomplete object. Sometimes many existing objects may be intended by means of one incomplete object, e.g. 'train' may serve to refer to many trains. I suppose that it is possible to say that we have an abstract meaning entity here, related to the word 'train'. Although this is not an abstract object of the Platonic realm. The situation with objectives is perhaps similar, in the sense that one apprehended objective may be used to intend several factual ones, and many objectives to intend one maximally rich with respect to the others in its apprehension of a complex. But there is a difference, for there are no existents among objectives. The subsisting ones and the non-subsisting ones, they all are best seen as meaning entities. That an objective is something abstract and meaning-like is visible when we consider such objectives which are factual even if their constituents are false, as in Russell's example with the implication 'if p then q', which is true when both p and q are false. This does not suggest, contrary to what Russell thought, that the false objectives p and q must have subsistence, at least not according to Meinong, but this suggests that it may be an operation on linguistic meanings, not on objects of any more substantial kind. Chisholm 219

Roderick M. Chisholm, "Homeless Objects", p. 50.

158

ANNA SIERSZULSKA

notices, in relation to this example, that some true sentences may be regarded as truths about language. Yet, he still maintains that these are not meaning entities we are dealing with in such a case, but rather objectives as some 'imagined' states of affaires. How else are we to interpret such principles? (...) We might interpret them, of course, as truths about language. (...) But Meinong would say, and I think correctly, that we cannot adequately interpret what it is to say of a sentence that it has the property of necessity unless we presuppose his doctrine of objectives.220

It is not too controversial, one may suppose, that, especially in the case of truths of a non-referring sort, it is possible to get rid of the notion of proposition, because sentences can entirely assume the position of truth bearers in such purely logical contexts. But there are also many such contexts where this is not so easy. Chisholm, even with his interpretation of objectives as states of affaires, regards them as playing the role of propositions as well, and accepts that such a notion is needed to explain some linguistic meanings of a more complex kind. Interpretations of objectives as states of affaires, or even as complexes, are sometimes dictated by logical considerations, since such an interpretation is useful for the purposes of a certain kind of logic. To provide an example, Karel Lambert finds it convenient to treat Meinong's objectives as the propositions of early Russell, to use them as the background theory in his type of free logic. Second, another matter of considerable current interest on which Meinongian reflections might shed illumination, concerns what are called, in recent discussions in the foundations of modal logic, 'Russellian singular propositions.' A singular proposition of the present kind, say, the singular proposition that this is white, contains as constituent the very object, the word 'this' specifies, rather than, say, an individual concept, in the position occupied by that object.221

A Russellian proposition, such as Lambert finds convenient to use, literally consists of the objects which it concerns, and not of any concepts or meaning entities. It is, therefore, a complex, even though it may be just as

220 221

Roderick M. Chisholm, "Homeless Objects", p. 46. Karel Lambert, Meinong and the Principle of Independence, p. 6.

MEINONG ON MEANING AND TRUTH

159

well false. Lambert thinks that propositions understood in this way are in fact very close to Meinong's original idea of an objective. At any rate, the parallel between the singular proposition of the Moore-Russell kind and 'singular' objective is so close that the theoretical light to be shed on the account of truth in modal logic championed by Kaplan and others such as Kit Fine cannot but be substantial.222

This parallel Lambert speaks about might not be really so close, as has been shown above. Nevertheless, the important point for Lambert is how the theory of propositions combines with Meinong's principle of independence of the nature of an object from its existence. This principle is the most important contribution of Meinong from the point of view of this logician. It provides an all-inclusive domain of objects without a definite existential status, about which it is possible to make true predications even if no such objects are to be found anywhere. The usual treatment of predication in extensional semantics is connected with assuming the being of the object about which we predicate, in accordance with the following rule: Gs; So, s has being

For this reason, it is entirely impossible to have any true predications about nonexistent objects. A free logician perceives this as a certain deficiency, quite understandably, since nonexistent objects often happen to play a relevant role in expressing various ideas, they are not meaningless, and should not be dismissed so lightly from any investigations. So Lambert criticises this principle formulated above calling it the principle of interdependence of being-so and being. (...) Burge adduces reasons based on the theory of truth (in the Tarski style) in support of his acceptance of the validity of the inference pattern in question. In Meinongian language this might be called the principle of interdependence of being so and being (...)223

222 223

Karel Lambert, Meinong and the Principle of Independence, p. 8. Karel Lambert, Meinong and the Principle of Independence, p. 90.

160

ANNA SIERSZULSKA

However, Lambert's acceptance of Meinong's principle of independence does not result in admitting a world of nonexistent entities. He claims that accepting this principle, he need not share Meinong's world view. It is implied that Meinong's world view does embrace a realm of nonexistent objects. Reflection on the fact established in Chapter 3 that the principle of independence does not entail a world of non-beings, and thus does not encourage open-armed acceptance of the core of the traditional theory of predication, should explain why. It is possible, after all, to agree with leading Meinongian principles without sharing his world view. Such evidently is the case for those positive free logicians who share Russell's robust sense of reality, but who find his sense of truth and logical form far more anemic when compared with Meinong's.224

Thus, utilizing Meinong's principle which makes it possible to consider the properties of objects that have no being, Lambert at the same time does not notice, perhaps due to a well established presupposition about Meinong's ideas, that the nonexistent objects in Aussersein possess properties without possessing any sort of being. So it is obviously not true that Meinong postulates a realm of subsisting bizarre objects, though it is not excluded that many objectives of so-being about nonexistent objects may be true on his theory. This fits perfectly to what Lambert needs for his purposes, except for one thing: Meinong's objective is not a complex and it does not include literally the objects involved, as Lambert's proposition does. This is a strange thing to say on Lambert's part, because how can a proposition literally contain a nonexistent object as its constituent, when he genuinely ascribes to this object no sort of existence? It cannot... Therefore, Lambert's proposition must be just a certain logical structure involving perhaps some nonexistent objects, but not literally containing them. In this case, it would be the same as Meinong's objective, but not the same as Russell's proposition-like complex, and not the same as a state of affairs, either. Some commentators, like Peter Simons, think that objectives resemble states of affairs, but acknowledge simultaneously that they don't resemble Russell's complexes. This is because the relation between an objective and its inferiora is of a "more abstract" character than it is supposed to be in a 224

Karel Lambert, Meinong and the Principle of Independence, p. 121.

MEINONG ON MEANING AND TRUTH

161

complex. (It is discussed earlier that Meinong regards objectives as dependent objects only on the basis of logical priority of the objects they presuppose, not as founded upon these objects.) Yet the difference between a true and false objective is that between subsisting and non-subsisting entities, so it is an existential difference. On Russell's early view this difference is qualitative only, for all propositions subsist as ontological entities, unacceptable as it has turned out to be. But for Meinong only half of these objects of judgment – the factual objectives – exist, whereas for Russell all propositions exist, and the distinction between true and false propositions is qualitative rather than existential. Russell views the proposition as a complex whose parts are its subjects, whereas for Meinong the relation between an objective and its subjects is more abstract than part-whole.225

If objectives were to be propositions of the kind of meaning entities, the difference should be also qualitative rather than existential. The existential division between objectives seems to suggest that they are actually states of affaires. One might even say that it is an improvement over Russell's view that not all of these states of affaires have subsistence. Still, there are some problems with understanding objectives as states of affaires. There are objectives that involve nonexistent objects, and Simons remarks that Meinong is only able to accept 'nonentities' because he takes an objective to be an 'about'-relation, rather than a part-whole relation. Now, isn't an 'about'-relation typical for meaning entities specifically, in opposition to states of affaires? Many people would be inclined to think that it is. And the other problem is with subsisting objectives involving nonexistents. A subsisting state of affaires, no matter how it is conceived, is not supposed to involve nonexistents. When we say that there are chimeras, this is a non-subsisting objective, and we could perhaps call it a fictional state of affairs. However, when we say that there are no chimeras, which is a subsisting objective, what kind of a state of affairs is that? Simons criticises Russell's idea of treating propositions as actually consisting of objects, of which Meinong cannot be accused, and it is an important point to make that Meinong's objectives are not like Russell's complexes. 225

Peter Simons, "On What There Isn't: the Meinong – Russell Dispute", p. 171.

162

ANNA SIERSZULSKA

For Russell a proposition is a whole of which its subjects are parts, and like all complexes its parts must exist if it does. This conception of propositions forces him to reject nonentities. Meinong's subject / objective relation corresponds much more closely to Russell's 'about' relation. Russell gives no reasons for thinking that objectives cannot be conceived in Meinong's manner, and his own theory is beset with problems. He owes us an explanation as to how a concrete object existing in space and time like Mont Blanc can be literally part of a non-spatial, non-temporal proposition, or how Hans, Erna and love can get together into two propositions, one true, the other one false.226

On the other hand, he does not go as far as to conclude that an objective may be just a sense entity, merely a logical structure or coordination. It may be so even when it is factual and subsistent, because the division between subsisting and non-subsisting objectives corresponds to the division between logical structures that obtain in reality or not. Despite very obvious objections, like these above, Russell's early conception of propositions still finds followers. One might wonder why. The reason may be that a proposition turns out to be something quite concrete on this view, a complex of objects bound by some relation. Propositions which do not literally consist of objects, like Meinong's objectives, seem to amount to imprecise abstract senses related obscurely to linguistic utterances. Such propositions are considered to be nothing at all ontologically and therefore to be nothing a serious theory can deal with. The same is true in the case of objectives seen as 'about'-relations, potentially concerning nonexistent objects, such objectives cannot be states of affairs from an ontological perspective, they can be nothing but sense entities.

226

Peter Simons, "On What There Isn't: the Meinong – Russell Dispute", p. 184.

MEINONG ON MEANING AND TRUTH

163

CHAPTER 7 SOME EPISTEMOLOGICAL INVESTIGATIONS CONCERNING MEINONG'S PHILOSOPHY ♦ The notion of a true judgment A book by Reinhardt Grossmann, Meinong, is perhaps the most popular source of reference, written so far, on issues related to Meinong's epistemology. Grossman tries to present a general review of Meinong's attitude towards cognition, rather than concentrate upon the problems of truth itself and its bearers. The basic notion for his investigations is that of a judgment. The reason is that he takes objectives to be states of affaires and he claims that on Meinong's view we are not acquainted with them directly – i.e. with facts – but only through judgments. Factuality is hidden from us, according to this first reason, because we are not directly acquainted with objectives, but only with judgments. Hence, there must be something in or about certain judgments that allows us to infer that their objectives are factual and, therefore, that they are true. Meinong's defense of evidence, seen from this angle, is merely a version of the traditional representationalist move.227

The difficulty for Meinong is different than the one stated by Grossmann above. We are acquainted with objectives, for after all they play the role of the immediate objects of judgments, but not with their factuality. Because factuality is not a feature apprehended inseparably from a factual objective. Thus Meinong has to introduce the notion of self-evidence of a judging experience as a way to perceive factuality of objectives. On one side we have the judgment (with its objective) and on the other something in the world that it is supposed to agree with. So this seems to Grossmann to be a typical representationalist pattern, where we look for some kind of correspondence. And the conception is indeed representational, but not by 227

Reinhardt Grossmann, Meinong, p. 137.

164

ANNA SIERSZULSKA

way of correspondence. Besides, Meinong's objectives are not conceived of as simple representations. But how do we know that a judgment is correct? Grossmann underlines the problem concerning self-evidence as a merely psychological conviction, which cannot guarantee that the objective of the judgment is factual and possesses subsistence. What is truth? A judgment is true, Meinong answers, if and only if its objective subsists. But since mere psychological conviction does not guarantee that the objective subsists, a serious problem arises. How can we ever find out that a given objective, O, subsists?228

To answer this question, Grossmann refers to the way Meinong explains the process of perceiving reality in Über die Erfahrungsgrundlagen unseres Wissens. Meinong writes there that although our sensory perception is only phenomenal, there are noumenal properties of objects which we perceive indirectly through the phenomenal ones.229 The division into phenomena and noumena sounds quite Kantian. Yet Meinong's approach is radically different. He does not think at all that there is a serious discrepancy between what there is and what we can perceive, which is the starting point for Grossmann's considerations. Thus Meinong argues that there are noumenal properties which correspond to the phenomenal ones; and he asserts that it is evident that the same relations hold between these noumenal properties as between the phenomenal ones.230

Grossmann finds this answer difficult to accept and presents the following analysis. Since we cannot know that the noumenal properties stand in certain relations of comparison by means of a priori judgments, and since we cannot know that they stand in these relations by means of a causal inference, we do not seem to be able to know it at all. (...) Recall Meinong's quite unsatisfactory account of abstraction. Meinong was then faced with the problem of how abstraction was possible from such a simple entity 228

Reinhardt Grossmann, Meinong, p. 133. A.Meinong, Über die Erfahrungsgrundlagen unseres Wissens, Gesamtausgabe, vol.V, p. 94. 230 Reinhardt Grossmann, Meinong, p. 145. 229

MEINONG ON MEANING AND TRUTH

165

as a certain shade of a certain colour. He tried to solve the problem by assuming that one and the same content (idea) can intend different objects with more or less distinctness. He attempts to utilize this same assumption in order to explain how we are acquainted with noumenal objects. He claims that one and the same content can distinctly apprehend phenomenal properties and indistinctly apprehend the underlying noumenal properties. This answer makes little sense, as far as I can see. (...) For example, the difference which holds between two phenomenal properties a priori and with evidence for certainty also holds between the corresponding, indistinctly apprehended, noumenal properties; but it holds for them merely with evidence for surmise. Furthermore, the strength of this latter evidence will depend on the degree of distinctness with which the noumenal properties are presented. There is an obvious objection – one might say: Berkeley's objection – to Meinong's answer. If the arguments from science and the relativity of sensing really prove that sensory qualities are mere phenomena, do not these same arguments prove the relations of comparison are mere phenomena?231

The main objection here is that, if we can perceive the noumenal properties at all, we should also be able to perceive relations between them, but this seems implausible in the situation when the noumenal properties are given indistinctly and only with evidence for surmise. While in the case of phenomenal properties we can judge a priori and with evidence for certainty that, e.g. a red object differs from a yellow one, we are unable to make such a priori judgments about the noumenal properties. The strength of evidence for surmise accompanying a judgment of this kind depends on the degree of distinctness with which the noumenal properties are perceived by means of the phenomenal ones. But the degree of distinctness of perception is unstable. How, therefore, are we ever able to tell anything about the noumenal properties of objects? Grossmann rejects Meinong's explanation that the same content can distinctly apprehend phenomenal properties and indistinctly apprehend the underlying noumenal properties. Meinong simply expresses here his overall view that the object as it is perceived, is implected in the real object, but maximally rich, and thus a limiting case for what is there to be apprehended. Meinong can say this precisely because he does not think that there is a gap between phenomenal properties and their counterparts in reality. Grossmann, to the contrary, 231

Reinhardt Grossmann, Meinong, p. 146.

166

ANNA SIERSZULSKA

holds on to the classic sceptical view that we do not know whether what we perceive has any relevant connection with what there is. In consequence of such a point of view, the solution which serves for Meinong as an additional strengthening of his generally realistic approach to cognition, namely the coherence criterion, for Grossmann constitutes the only way of escape from the sceptical dilemma. What we actually do – and cannot help but do – is to compare objectives with each other. There is thus a criterion of truth after all. It consists in the 'coherence', logical and factual, of objectives. The so-called coherence theory of truth has a point, but it is an epistemological rather than an ontological point. Some states of affairs exist (Meinong would say: they subsist), others do not. This is the ontological part of the story of truth. But the ontological status of a state of affairs is not open to view. Factual and non-factual states of affairs, though they are inherently different, look alike to the mind. thus, The best we can do in this world is to compare states of affairs with each other, to trust some of them and to eliminate others on the basis of those we trust. If this account is true, then we are indeed faced with those consequences which, in Meinong's words, amount to our abandoning all confidence in our judgments and, hence, to our relinquishing all knowledge. There is then no solution to the problem of knowledge which would satisfy Meinong. There is then no criterion of knowledge other than one in contextual terms.232

Grossmann sees the coherence criterion as a rather desperate step, when every other solution has already failed. But Meinong never says that we should abandon all confidence in our judgments! He only says that such would be the outcome if, following Kantists, we assumed that there is no reason to suppose that our apprehension of reality is correct. Although Grossmann's presentation of Meinong's theory of cognition is very insightful, I cannot but disagree with him when he presents Meinong more as a sceptic than as a realist. I do not think this is right. The book by M.L. Schubert Kalsi entitled Meinong's Theory of Knowledge, contrary to what the title suggests, does not contain many comments concerning truth in Meinong's philosophy. The purpose of the author is different, much closer to the analysis of perception and memory, in relation to knowledge. When the question of truth appears in the course of discussion, the key notion is again that of a judgment. 232

Reinhardt Grossmann, Meinong, p. 138.

MEINONG ON MEANING AND TRUTH

167

I will not say much about truth in this book because it is not its proper subject matter. The concept of true judgments can be transferred to the corresponding objectives which in turn may be called true. In this sense, factual objectives are called true. Truth is founded on a relationship which is outside the judging person and which holds between a judgment and its objective, whereby of course the judgment is an experience of a person.233

Factual objectives are understood as complexes. Since they are called true, we get an ontological conception of truth here. The relevant relation holds between a judgment and its objective. A judgment in order to be true must apprehend a factual objective, which is out there in the world as a complex. So a judgment, as an experience of a person, is something mental, and an objective is a part of reality, a complex. There is not a trace in this account of any such role to be played by objectives as would be ascribed to propositions. We can see in the definition of knowledge presented in this book that knowledge also does not consist of objectives, but only of true judgments. Besides, these true judgments have to be, at the same time, self-evident, to count as belonging to what is called knowledge. This is a very epistemic approach to what knowledge consists in, seen from the point of view of the process of cognition exclusively. Knowledge or cognition finally consists of judgments which are true, i.e., penetrating and and evident. Not all true judgments are knowledge. But judgments which are true and also evident are knowledge. In other words: knowledge is penetrating apprehension with evidence. Evidence is an experience, it indicates the psychological situation. Truth is the objective factuality, independent of the subject. In a priori matters it can be said that the internal justification for truth is in such judgments which really are knowledge, i.e. which are true and evident.234

The objection may be raised that knowledge is usually considered to be something more permanent than the very moment of making a judgment, with its illumination of evidence. Therefore, knowledge cannot be regarded as merely consisting of true and evident judgments, rather than of objectives. In a priori knowledge, where we judge with evidence for certainty, this is especially transparent. When we use true objectives as 233 234

M.L Schubert Kalsi, Meinong's Theory of Knowledge, p. 103. M.L Schubert Kalsi, Meinong's Theory of Knowledge, p. 104.

168

ANNA SIERSZULSKA

premises of reasonings, we are not obliged each time to make judgments about their truth. We operate on objectives, not on judgments. This perspective, however, must be entirely alien to someone who treats objectives mainly as ontological entities. It is difficult to see them simultaneously as propositions which are supposed to constitute a coherent system of knowledge. Thus, M.L.Schubert Kalsi seems not to be able to find an explanation for the way Meinong uses the coherence criterion in his theory. If truth and knowledge are the same in respect to a system, then an objective, in order to be true, must be shown to belong to the system. That does not include such objectives which possibly belong to the system but are not yet shown to belong. (...) He could also have said that the showing and accepting of an objective's belonging to the system involves evidence in his sense and thus does not eliminate it. But he does not say it. (...) We must surmise: perhaps we could say that a system is a complex. i.e., an objectum whose members are objectives. A complex is formed by members and relates holding between them. But the members must be objecta, too. Thus, a system of objectives cannot be, according to Meinong's theory. Objectives can only be combined to form more complex objectives. They cannot be part of an objectum. Relates, as we know them, never hold between objectives. So the system cannot be.235

She considers it to be important that judgments about whether an objective belongs to the system must also be made with evidence, and so, the coherence criterion does not eliminate evidence. And so it does not, but I do not think that it is really Meinong's purpose to eliminate evidence. He rather tries to find a remedy for its insufficiency. Of course, introducing the coherence criterion, Meinong is mostly concerned with judgments possessing only the evidence for surmise, generally speaking – empirical judgments. This kind of evidence is insufficient to confirm the truth of the judgment's objective, and, due to this problem, he adds the coherence principle. Making inferences from objectives by applying the rules of logic, one could make judgments with evidence for certainty in the process, which could not be questioned as to its doubtful outcome. In this way, an empirical objective which is strongly supported by an evidence235

M.L Schubert Kalsi, Meinong's Theory of Knowledge, p. 106.

MEINONG ON MEANING AND TRUTH

169

experience and can also be demonstrated to be coherent with the system of objectives considered to be true, can be accepted as true. According to M.L Schubert Kalsi, Meinong's system of objectives is something completely unthinkable. To support her view, she presents a rather fantastic vision of a system of objectives as one huge complex consisting of smaller units, which are objectives. And obviously, Meinong could never have had anything of this sort in mind. For him objectives resemble logical structures, and in contexts where he talks about logical relations between them, they behave like propositions from which one can derive consequences or use them to prove certain conclusions. A very valuable point, on the other hand, is made in the introduction to her translation of Objects of Higher Order..., where M-L.Schubert Kalsi pays attention to an important difference between objectives and other objects of higher order. She observes that objectives are not considered by Meinong to be objects of higher order at first because they are not literally composed of objecta. When I was studying what we now call complexes, according to the later Meinong, I kept asking myself why objectives were not objects of higher order or, rather, why objectives were not complexes or not even members of complexes. The question can easily be answered, now. All objects of higher order were originally objecta. This excluded objectives automatically from being objects of higher order. Nevertheless, why not objectives? In order to qualify they would have to be either complex-like or 'hungry'. In the beginning of this introduction it was said that objectives are, figuratively speaking, composed of objecta (or other objectives). But, in fact, they are not. Objecta are their presuppositional objects, that is, the objects about which the objectives are. Later, of course, the need for presuppositional objects sufficed to qualify objectives as objects of higher order.236

She says that objectives presuppose other objects in the sense that they are 'about' these objects. And we may add that the relation of being about something is characteristic to semantic entities, like meanings. But she does not arrive at this conclusion, nor at the idea of treating objectives as propositions, which is why Meinong's concept of a coherent system of objectives remains incomprehensible on her account.

236

M-L.Schubert Kalsi, Introduction, in: A.Meinong, Objects of Higher Order, p. 21.

170

ANNA SIERSZULSKA

An account of Meinong's objectives as being closer to states of affairs or complexes, than to propositions, is also presented by Kevin Mulligan in his article "Judgings: Their Parts and Counterparts". Both early-Russell's and Meinong's propositions are understood as 'hybrid entities', something between abstract meanings and complexes. However, this is one of the interpretations of objectives which ascribe to them explicitly the role of propositions. Matters have been further complicated by the tendency of some philosophers to use 'proposition' to refer to states of affairs or to hybrid entities combining features of propositions (qua abstract entities) and states of affairs. (Moore and Russell started this tradition; to the same tradition belongs Meinong's use of Objective.)237

Assuming complex-like objectives, the notion of a judgment, or rather of a 'judging' as a kind of mental act on the part of the judging person each time he judges about a certain state of affairs, acquires the central place in any discussion related to truth. For objectives, like states of affairs, are regarded as actually consisting of spatiotemporal components. And even though there is a sense in which objectives play the role of meaning entities and propositions on this account, they differ from the latter substantially as to the way they are classified as entities (in agreement with the standard interpretation of Meinong's objectives as states of affairs). It is claimed to be especially the mark of the different status of objectives, compared to non-spatiotemporal ideal propositions, that they are only capable of multiple access, but not multiply exemplifiable. Now the distinction between the judging and the state of affairs is that the latter but not the former contains perceivable spatiotemporal components. (...) Husserl, Stumpf and Meinong regarded the most basic sort of positive states of affairs as being, like abstract meaning entities, ideal, but as differing from these in not being multiply exemplifiable. (Although they are of course capable of multiple access.)238

I suppose that what is meant by 'multiple access' is that many people can think about the same objective or state of affairs and in each case it becomes the object of their mental acts. It is a different situation than when 237 238

Kevin Mulligan, "Judgings: Their Parts and Counterparts", p. 119. Kevin Mulligan, "Judgings: Their Parts and Counterparts", p. 121.

MEINONG ON MEANING AND TRUTH

171

the same ideal content is 'exemplified' in the mind of each person making a judgment. A state of affairs is conceived of as an ideal entity, but it has spatiotemporal components, and so it cannot be exemplified in anyone's mind, although it can be intended in many acts. So it is within Husserl's theory at least, where states of affairs are the objects intended in judgments. Husserl's claim is best understood by comparing it with that of Frege, whose position is, with respect to the issue under discussion, intermediate between that of Marty and Husserl. A Fregean Gedanke can indeed be tied or linked to the real world of psychological episodes but it is not so tied by virtue of being instantiated. Rather, it is exemplified by a denizen of the real world, a judging.(...) On Husserl's view, both the table and its redness – what is often called a 'particularised property' or a 'moment' – are non-repeatable denizens of the spatiotemporal world. Each instantiates a repeatable denizen of the third world: Table and Redness.239

Husserl's meanings of sentences in Logical Investigations are primarily the ideal contents of judgments, which must be understood as not actually containing spatiotemporal constituents. The states of affairs intended in judgments, however, are taken to be identical with the ultimately intended states of affairs and these possess spatiotemporal constituents. But the conceptions of Meinong and Husserl differ in many respects, and I think that they differ with respect to this issue as well. On Meinong's theory, every particular objective of a mental act is a part of reality as well, for it exists in the mind. But not in the same sense that a state of affairs, or a complex, can be a part of reality. Meinong's objectives are genuine meaning entities and they neither contain any spatiotemporal constituents, nor do they exist only as intentional objects. If true, their structure is identical to that found to hold for certain fragments of spatiotemporal reality, but they share with reality only the logical structure, not its spatiotemporal character. On the other hand, objectives do not share the merely temporal character of intentional objects. This manifests itself in the fact that objectives in the spectrum of Aussersein are timeless, or beyond time. Yet, false objectives do not possess any ideal being, so we cannot say that they are 'multiply exemplifiable' in particular mental acts. Therefore, if being 'capable of multiple access' could also be applied to an 239

Kevin Mulligan, "Judgings: Their Parts and Counterparts", p. 123.

172

ANNA SIERSZULSKA

intentional relation towards an abstract meaning entity, it would be more proper to describe the relation between the mind-independent objectives and their realisations in mental acts in this way, than to use the metaphysically biased 'multiple exemplification'. Meinong's conception is also quite different from Brentano's , though it resembles Brentano's conception very closely as far as the theory of mental acts is concerned. But Brentano does not allow any meaning entities or propositions, to the extent that his definition of truth involves a requirement that an act of judgment should be identical to that of an ideal judging person who judges correctly, in order for it to be true, instead of saying that either the meaning or proposition expressed in the act of judgment should be correct. This is not true about Meinong and, therefore, I only partly agree with the following remark. But in fact Meinong accepts all the fundamental features of Brentano's account of judging. He, too, is committed to a non-propositional account of judging and so also of assuming. He, too, accepts that the quality of a judging comes in two opposed kinds.240

To be more exact, I do not agree that Meinong's account of judging is nonpropositional, although it shares the feature of 'coming in two opposed kinds' with the non-propositional account proposed by Brentano. The content of a judgment or assumption, and thus the objective, may be negative in itself, e.g. 'There is no snow in Cracow on the 1st of June'. But in addition to this inner yes-no polarity, the act of judging itself possesses the quality of affirmation or rejection. We judge that it is so, or not, that there is no snow in Cracow on the 1st of June. The problem is that the affirmation or rejection does not belong to the objective. This suggests that there is a certain drawback in Meinong's theory, as a remnant of Brentano's position which is the starting point for Meinong's investigations, but this does not mean that his account of judging is non-propositional. A related question is the subject matter of an article by Urszula egle , who analyses negative judgments in the conceptions of Meinong and Ingarden in order to prove that they do not require negative objects, such as negative states of affairs. Since she identifies objectives with Ingarden's 240

Kevin Mulligan, "Judgings: Their Parts and Counterparts", p. 127.

MEINONG ON MEANING AND TRUTH

173

states of affairs in her paper, she assumes that a negative objective, unlike a negative proposition, would need to possess an ontic foundation. The fundamental ontological question raised on the basis of the analysis of negative judgments concerns the existence of negative states of affaires. States of affaires, like objectives in Meinongian theory of objects, can be defined as objects of higher order, that is, objects whose occurrence is warranted by the basic objects which are their ontic (necessary) foundation.241

The analysis is, therefore, directed at demonstrating that there are no genuine negative objectives, only negations of positive ones. It would be equally strange to have negative objectives as to have negative states of affaires, in so far as this would imply the presence of some negative states of reality. The conclusion of this analysis is, for this reason, correct about an important point: that no negatives can be autonomous objects in reality. The analysis of negative judgments made by Meinong and Ingarden, shows that both philosophers assume a moderate standpoint which can be expressed in the following theses: 1. 'Negatives' do not belong to the realm of autonomous objects. 2. Ontologically 'negatives' have their foundation in positive states. 3. 'Negatives' are to be distinguished in the operation of our mind only subjectively.242

Points 2 and 3, as applied to Meinong's conception, have to be discussed further, because they are questionable in this formulation. Objectives are actually meaning-objects, which differ from all the other objects in many ways. As objects of higher order, they possess inferiora, but the objects presupposed in this sense, are only logically prior to them. They do not need any ontological foundations, just like propositions do not need any such foundations. In this much, point 2 of the above conclusion does not apply to Meinong's conception. However, it is certainly so that negative objectives are true not because of any negative states of reality, but because of positive ones, e.g. it is true that something is not oval, because it is in fact triangular, and these two properties exclude each other. Thus,

241 242

U. egle , 'Meinong and Ingarden on Negative Judgments', p. 268. U. egle , 'Meinong and Ingarden on Negative Judgments', p. 275.

174

ANNA SIERSZULSKA

negative objectives depend on positive states of reality for their truth, but not for their ontological foundations. As to point 3, it is not correct about Meinong's objectives, because negative objectives are not just intentional objects distinguished only subjectively. Their status is entirely intersubjective and mind-independent. On Meinong's conception we can say that there are true negative objectives, even though there are no negative complexes to be apprehended by them, or there are no negative states of affaires. And it is hardly an argument against negative objectives that they cannot be given in presentation, for Meinong puts a lot of emphasis on the fact that objectives, either positive or negative, cannot be represented. Objectives merely involve representations of their inferiora. Naturally, in the case of negative objectives there is no representation of a negative property, like nonredness. It is rather that the positive property of redness is presented and then negated. It is important to stress that the negation does not have to lie outside of the objective itself, merely as the quality of an act of judgment. We know further from the theory of judgment that the act of judging is founded upon the act of presentation. Thus the question arises whether negatives can still be given in presentation. The answer to this question has an important consequence for ontology because the ability of conceiving something in presentation determines its 'objecthood' in the sense of Gegenstandlichkeit.243

Although negation cannot be given in presentation, it can still belong to an objective, for in spite of being non-presentable, it is, nevertheless, conceivable. This agrees with Meinong's general idea of an object of intention, which is not really what can be presented, but what can be conceived in a very broad sense. A negative objective remains a semantic entity in its own right, even when it is interpreted as an objective of second order, as Urszula egle proposes. She claims that each negative objective consists of a positive objective of the first order, and an objective of the second order 'It is false that p'. Therefore, the objective 'Trento is not in Germany' is interpreted as "It is false 'that Trento is in Germany'". But the second order objective still includes an implicit negation (it is false), which is explicitly visible in the symbolic notation: '~p'. 'It is false that p' means 'It is not the case that p'. If true, this objective of the second order is a 243

U. egle , 'Meinong and Ingarden on Negative Judgments', p. 275.

MEINONG ON MEANING AND TRUTH

175

subsisting entity, and a genuine negative objective at the same time. We do not get rid of negative objectives in this way, even though we get rid of negative states of affairs. We would interpret the analysed cases of negation in its relation to falsehood semantically in the following way: '~p' means 'It is false that p'. For example for p: 'Trento is in Germany.' The sentence p is directed to an objective which according to the one interpretation is treated as falsehood, and according to another interpretation is treated as the objective which is nonconsistent with a factual objective given, for example in the form that Trento is in the autonomous region of South Tirol. In turn in the analyses of the item 'It is false that p' we have the objective that p which is an objective of the first order, i.e. Oi. The propositional attitude 'It is false' can be here interpreted as an attribute of the object Oi. The objective Oi with its attribute can be said to form an objective of the second order, i.e. Oii.244

The situation is very similar with negative properties, like 'being non-red'. The relevant negative objective is analysed as 'It is false about the object A that A is red'. In all these cases, Meinong's conception allows for transposition of negation in such a way that it precedes a sentence, which is not without significance for the general character of objectives. Objectives behave like propositions in this respect, not like states of affairs. From Meinong's illustration of 'non-red' we know that 'non-red' means 'being an object, such that it is false of it, that it is red'. Thus 'A is non-B' means 'it is false about A that A is B'. We can say this because the propositional attitude in (d) does not require a many-order interpretation which was needed for the previous negative cases. By using, however, the predicate 'is false' Meinong's interpretation seems again to presuppose a lower level objective which is positive (i.e. 'that A is red'). This analysis shows that on the syntactical level Meinong's logic (like traditional logic) allows for transposition of negation that deals with the namecategory into negation whose argument is a sentence.245

Altogether, we can say that even if a negative objective can be analysed as a negation of a positive objective, this gives a negative objective of the second order, but does not eliminate negative objectives as such. It also does not follow that negative objectives are distinguished only subjectively, as merely intentional objects. In fact, we have a situation 244 245

U. egle , 'Meinong and Ingarden on Negative Judgments', p. 271. U. egle , 'Meinong and Ingarden on Negative Judgments', p. 271-2.

176

ANNA SIERSZULSKA

where the negative objectives 'A is non-red' and 'It is not the case about A that A is red' subsist, while the positive objective, which is subordinate to the latter one, does not subsist. So it seems that there are negative objectives which, however, cannot be understood as states of affairs. If judgings were only to affirm or reject states of affaires, such an account of judging would be non-propositional. But on Meinong's theory, there are such objects as true negative objectives and they are not states of affaires. This gives us a right to assume that they can only be propositions, and the entire account of judging is propositional after all. ♦ Timeless truth Meinong believes that truth as such is timeless. Ideally subsisting objectives are potential objects of true judgments in an eternal sense. However, on an interpretation of objectives as states of affairs that contain spatiotemporal objects as their constituents, the eternal character of truth would be difficult to defend. Thus it seems to be a popular view that objectives have two functions which are played by the same entity. On one hand, they are real or unreal states of affairs, and on the other, they are propositional meanings of sentences. Such a view is typically represented by Peter Simons, whose papers on Meinong's semantics are discussed in Ch.3. Here I quote a passage on truth, written together with Wole ski. Meinong's theory of truth is not exactly a standard correspondence theory. The objects of judgment according to Meinong are mind- and language-independent Objectives (his name for what others called states of affairs), which also serve as the meanings of declarative sentences. Objectives simply subsist or fail to subsist. Subsisting objectives may be called 'factual' or just 'facts'. A truth for Meinong is however a particular kind of fact, namely one which is actually intended by some subject. So there is no correspondence between truths (the meanings of true sentences) and facts: they are identical. However a true individual judgment is one which intends, and in this sense corresponds to, a factual objective.246

246

J.Wole ski and p. Simons, "De Veritate: Austro-Polish contributions to the theory of truth from Brentano to Tarski", in: K.Szaniawski, The Vienna Circle and the LvovWarsaw School, p. 396.

MEINONG ON MEANING AND TRUTH

177

So objectives serve as meanings and they are mind and language independent, which is entirely suitable for bearers of timeless truth. But the meanings of true sentences are identical with facts, facts are states of affaires, and so, meanings are states of affaires. Of course, states of affaires, as facts, contain spatiotemporal objects. Is Meinong's semantics based on a simple reference relation, if meanings are states of affaires? Rather not, there is a visible difference between Meinong's and Russell's theory in this respect. Let us direct our attention once again to negative and fictional objectives, concerning nonexistent objects, or contradictory, or incomplete ones. These must be meaning entities, not simply objects of reference. In any case, the point of view of Simons is closer to accepting objectives as propositions than the way this matter is seen by Wole ski alone: Thus, on Meinong's view, truth has two dimensions: epistemological and ontological. Epistemologically speaking, truth is connected with acts of apprehending something. This means that there is no truth without subjects of knowledge. However, it does not entail that truth is predicated of 'epistemological' bearers, that is judgments or propositions considered as something belonging to a language. The epistemological dimension of truth is also connected with selfevidence as its criterion.247

Wole ski thinks that the bearers of truth in Meinong's conception are not epistemological, which means that they cannot be treated as propositions. On his interpretation, which is rather post-Russellian, objectives are ontological entities, such as complexes. So when he speaks about any absolute notion of truth in Meinong's philosophy, it is ontological truth. An ontological notion of truth consists in the claim that what there is, is true, independently of the subject of cognition. This seems to follow from the fact that truth pertains to objectives, not to judgments. But is ontological truth timeless? Is the same objective likely to be true after many years, when its objects have already disappeared? On an ontological conception that would not be possible, while this is what is required in Meinong's theory. Then something must be wrong with the ontological interpretation.

247

J.Wole ski, "Theories of Truth in Austrian Philosophy", Essays in the History of Logic and Logical Philosophy, p. 157.

178

ANNA SIERSZULSKA

It is interesting to ask whether Meinong's is an ontological or an epistemological theory. Although Meinong admits the Erlebnisbegriff of truth, he still maintains that truth in its proper sense is an attribute of objectives. This is an argument to the effect that he defended an ontological theory. A contrary interpretation may be based on Meinong's statement that truth consists in equality of intended and factual objective. In any case, even if Meinong's theory is interpreted ontologically, it is certainly different from the view of Thomas Aquinas that truth is coextensional with being qua being. Also it is not clear how far Meinong's theory of truth is semantic. The priority of the view of truth as a prima facie property of objectives pushes Meinong's theory in an asemantic direction. However, a similar qualification is here in order as in the case of its ontological character.248

Wole ski does notice that something must be wrong with the ontological interpretation, but he does not say exactly what. What is most certainly wrong is the ontological interpretation of an objective itself. An objective is introduced by Meinong as a counterpart of the content of acts of judgment and assumption. It is not originally conceived as belonging to the world. For this reason it can concern e.g. contradictory objects. As we know, some objectives concerning contradictory objects are true on Meinong's theory. This is definitely not characteristic of ontological conceptions of truth! Another problem, which is mentioned by Wole ski above, is whether Meinong's is an 'asemantic' conception, i.e. non-semantic. He is inclined to think that it is 'asemantic', but during the last years he has changed his position on this issue to vaguely undecided. He prefers to see sentences in the role of truth bearers, or, in the worst case, propositions as related to linguistic sentences. The difficulty with Meinong again lies in the character of objectives, which obviously serve as meanings of declarative sentences, but understood as complexes, they are not the abstract meaning entities one could accept as truth bearers in a semantic conception of truth. Again, we can argue, that since only abstract propositions can be timeless, and objectives, subsisting or not, are entities conceived of as remaining beyond time, they have the character of abstract propositions, and Meinong's conception of truth may not be as 'asemantic' as it seems to be at first sight. 248

J.Wole ski, "Theories of Truth in Austrian Philosophy", Essays in the History of Logic and Logical Philosophy, pp.156-9.

MEINONG ON MEANING AND TRUTH

179

The understanding of the notion of fact as a true proposition, rather than a factual state of affairs, is another element in Meinong's conception that leads to inadequate interpretations of his notion of truth. It is often assumed that Meinong's true proposition is simply the same thing as a factual state of affairs, in consequence of which the whole conception of truth turns out to appear as ontological and deprived of timeless truth bearers. In opposition to this tendency, it has been noticed by David Lindenfeld that Meinong's notion of a fact is different from what is usually understood under this notion, because factuality applies also to negative objectives, in the same way that modal properties apply to them. He defines truth in the following way: "A judgment is true whose objective is a fact". Lest this seem to be mere word-play, it should be noted that Meinong develops a new context in which the term 'fact' appears. A factual objective is not merely one of existence or so-being, but includes negative objectives; it is also a fact that centaurs don't exist. Factuality, then, is a distinct property of certain objectives, independent of their being or non-being and is related to such other properties as possibility, probability, and necessity – the so called modal properties. The statement, 'it is a fact that flying saucers don't exist' is along the same lines as 'it is possible that flying saucers don't exist.' It is to these objectives that the terms 'true' and 'false' should be applied, not the simple objectives of being or so-being. Meinong's definition, it should be noted, does not include a criterion for distinguishing a factual situation from a non-factual one; it is merely an attempt to describe what it is to say that something is true.249

As a result, there are negative facts on Meinong's theory. However, these negative facts do not resemble negative states of affairs or negative complexes, which would have to be difficult to accept. It is much easier to admit negative facts as negative true propositions. Naturally, it follows that a fact is not something in reality. Instead, it is the way something is, or is not, in reality. An agreement with facts so conceived constitutes the essence of Meinong's notion of truth. An objective as a way things may be, can be impossible, possible, or factual, probable, true or false, truth being one of its modal properties on this view. Like other modal properties, truth pertains to objectives beyond time. Lindenfeld is right that there is no criterion of truth given in Meinong's theory, just an explanation what it 249

David F. Lindenfeld, Transformation of Positivism, Ch. VI The Linguistic Analogy and the Theory of Objects, p. 158.

180

ANNA SIERSZULSKA

means to be true. Because self-evidence of judgments is only an epistemic substitute of a criterion, it is too subjective. Meinong's objective notion of truth is, for this reason, based on factuality, understood in a specific way. I suppose that we may accept that Meinong's factuality is by no means coextensional with any entities which are found in the world. And since factuality of objectives coincides with their truth, truth is not ontological on Meinong's conception. Truth pertains to objectives as propositions, which possess the property of factuality but are not to be found in the world. All objectives, factual and non-factual, belong to Meinong's Aussersein. This is why they can be bearers of timeless truth or falsehood eternally. ♦ Objectives in Aussersein That Aussersein is related to what can be meant and expressed by linguistic means, rather than to what can be considered to have being of some kind, is confirmed by the general attitude of the Graz school towards the study of intentional phenomena. One of the examples may be provided by the concept of producing, to which Liliana Albertazzi pays special attention in her contribution to the book The School of Alexius Meinong. The concept of producing encloses the element of rational structuring that is involved both in mental representation of objects and in the apprehension of objectives. In particular, the concept of producing is a distinctive feature of the Graz school's analysis, in that it coincides immediately neither with presenting (Vorstellen) nor with judging (Urteilen), nor even with assuming (Annahmen). Specifically, the theory of production deals with those aspects concerning the subjective mental activities of composing or dividing a whole of presentation, of introducing elements of scansion or stressing among the parts.250

This attitude towards mental acts implies that the objects or objectives given by their contents involve an element of rational structuring, which is not to say that some sort of idealism is at issue here. It is simply the case that human mind can focus on different aspects of reality, see it in different 250

L.Albertazzi, "Presentation and Production", The School of Alexius Meinong, p. 242.

MEINONG ON MEANING AND TRUTH

181

ways, and so express different meanings in relation to it. Thus, we have the perceptive elements and their mental organization intertwining in the apprehension of everything that is thought about. In fact, from the point of view of the act: 1.Internally to the current presentation, even in the brief duration of its extension, almost always acting in the shaping of the objects presented are, besides perceptive aspects, also mental aspects like assimilative or additive schemata or ones which compare among the contents conceived. Instead, from the point of view of the object: 2.Both the perceptive presentations and the mental presentations comprise elementary as well as founded objects. Perceptive presentations and mental presentations, therefore, may be equal with respect to objects but different with respect to the type of act, which is exactly what the theory of production maintained from the very beginning, as a specification of certain aspects of the theory of intentional reference.251

From this general doctrine it may be further assumed that what plays an important role in determining the internal structure of objectives, is the specific focus of consciousness on the part of the subject, whose attention may be directed to some relations perceived in reality while disregarding or remaining blind to others. For this reason, an objective should not be conceived as a fragment of reality, for even though its structure is supposed to be identical with that found in reality, it is not the world that determines which structures will be perceived by a subject of cognition. Due to the addition of rational organization to what is perceived, objectives belong very naturally to the sphere of meaning, even though they also objectively subsist as being correct about reality. There are still diverse misunderstandings in connection with the nonontological sphere of pure objects that Aussersein is designed to be. The most important one of these misunderstandings results from a popular belief that all objects in Aussersein must possess some sort of being after all, and therefore, all these round squares and golden mountains have to be considered as ontological entities according to Meinong's theory. It has been explained many times by Meinongian scholars, and in a most 251

L.Albertazzi, "Presentation and Production", The School of Alexius Meinong, pp. 256-7.

182

ANNA SIERSZULSKA

emphatic fashion by Dale Jacquette in The School of Alexius Meinong, that this is an entirely mistaken interpretation of Meinong's Aussersein. Meinong's doctrine of the Aussersein of the pure object posits a semantic domain of absolute ontic neutrality. Aussersein is literally beyond being, a realm of object theory objects that comprehends not an ontology, but an extraontology.252

By calling Aussersein a domain of extraontology, Jacquette means to say that all objects indiscriminately belong to this domain, ie. everything that can be thought about or meant or expressed in language is an object of Aussersein. Some of these extraontological objects can be identified with ontological entities, either existent or subsistent ones, but a great amount of them belong nowhere, apart from this abstract sphere. Thus, one can find all objectives in the sphere of Aussersein, and while some of them can be identified with subsisting objectives, which are ontological entities in the sense of possessing ideal being as truths, others can be found only and exclusively among the 'merely semantic' objects, as I call them. Objectives as 'semantic' objects can be bearers of truth and falsehood timelessly, and they can be intended by mental acts of many subjects. In Meinong's terminology we would have to say, to be exact, that objectives are timeless bearers of factuality or non-factuality. But for Meinong, to say about an objective that it is factual, is equivalent to saying that it is true in a non-subjective sense. Factuality is simply 'objective truth', and nonfactuality is 'objective falsehood'. So non-factual objectives are 'merely semantic' objects in Aussersein which are timeless bearers of objective falsehood. 'Tokens' of them are intended in particular mental acts of different people. Objectives are meaning entities, possibly correlated with certain complexes of objects in the world. In the case of negative objectives or those concerning nonexistent objects, of course there is nothing to correlate them with. Yet certainly, as Jacquette says, there is still much that can be said truly about purely extraontological entities. Meinong's mature object theory permits reference and true predication of properties to objects in both the ontology and extraontology, indifferently. It is enough to be an object, which is to say, something that can be thought about, 252

D.Jacquette, "Aussersein of the Pure Object", The School of Alexius Meinong, p. 373.

MEINONG ON MEANING AND TRUTH

183

regardless of whether the object belongs to the ontology as a spatiotemporal existent or abstract nonspatiotemporal subsistent entity, or to the extraontology as a beingless object or mere object of thought.253

What is specifically 'extrasemantic', I would say, about Meinong's Aussersein is that not only linguistic meanings belong to it, but absolutely everything that can occupy thought. For Meinong, this sphere is broader than that of linguistic meanings alone. There is naturally no necessity to agree with this conviction, for one may equally well maintain that what can be thought is very much determined by what our language allows us to think about. A compromising view might be that Meinong's Aussersein includes all the meanings possible to express in all possible human languages. The idea of a sphere of pure objects, considered regardless of their existential status, brings to mind Husserl's conception of epoche as a method applied in his phenomenological studies. But Husserl's objects of consciousness constitute his ideal transcendental ego, while Meinong's theory of pure objects simply deals with possible objects of intention that may occur in ordinary mental acts of different people. One has to agree with Jacquette's remark that Husserl's approach is on the whole more subjectivistic and idealistic. Husserl's methodology, despite his frequent assertions of phenomenology as scientific endeavour, is more subjectivistic then Meinong's object theory. Husserl indeed understands the phenomenological method of transcendental epoche as uncovering the transcendence of the pure ego, which Husserl in neo-Kantian idealist fashion, in his transcendental phase after the 1913 publication of the second edition of the Logische Untersuhungen, takes to be a precondition for the existence of the natural world. Meinong, by contrast, as we have seen, regards the Aussersein of the pure object more objectively as a domain of intended, or, better, potentially intendable, objects that are available to but semantically independent of thought.254

Meinong's domain of pure objects is not subjective and it just provides objects to be intended. In the case of such meaning entities, no ontological 253

D.Jacquette, "Aussersein of the Pure Object", The School of Alexius Meinong, p. 384. 254 D.Jacquette, "Aussersein of the Pure Object", The School of Alexius Meinong, p. 387.

184

ANNA SIERSZULSKA

specifications as to their status are needed, for the reason that they are all equal as potential objects of intention on Meinong's view. His theory of Aussersein accounts for the fact that people may mean and communicate about the same things, whether real or not, and they may also assert, or reject as false, the same objectives. Husserl's vision, that one must abandon one's existential presuppositions and search through one's consciousness to find reality, with a possibly subjectivistic and idealistic interpretation of this term, does not lie behind Meinong's idea of Aussersein. ♦ The Objection of Conception Dependence The objects in Aussersein are independent of thought in the sense that the subject does not create them in each of his acts, and many subjects may think about the same object or objective. But it may be argued that the 'choice' of objects or objectives that are actually intended (picked out from Aussersein) in the acts of subjects of cognition, is dependent on a number of factors. In everyday life, people may see things in a particular perspective due to their interests or goals. In scientific practice, the focus is provided by the point of view of the theory which is applied in a given case. Further, we may take into account the typical expression structures characteristic of a certain language. These more or less subjective factors decide, however, only about the 'choice' of auxiliary objects by means of which the ultimate object is intended. Several auxiliary objects may be implected in the ultimate one as non-overlapping subsets of its properties. So two people may really intend the same object, but from entirely different angles. This feature of Meinong's theory has been the source of the objection concerning the conception dependence of cognition. Such an objection is first mentioned by Findlay, but it is formulated fully and at length in the book by Smith and McIntyre, Husserl and Intentionality. The Meinongian argument for incomplete objects is the key to a Meinongian treatment of conception-dependence, though Meinong seems not to address that topic explicitly. Meinong holds that all objects that actually exist (or subsist) are complete objects but that, due to the finite capacities of the human mind, we can never conceive any such complete object.(…) According to Meinong's views, then, acts in which Smith conceives Napoleon as the vanquished at Waterloo and acts in which he conceives Napoleon as the victor

MEINONG ON MEANING AND TRUTH

185

at Jena – both of which can be said to be acts of intending Napoleon – are in fact directed to distinct (incomplete) objects and, for that reason, are distinct intentions. (…) Because intention is conception dependent, and because our conceptions are always limited in what they can include concerning ordinary objects, Meinong's object-approach to intentionality forces the conclusion that the objects (directly) intended in even our most commonplace acts are not ordinary, complete existents. The objects of all our intentions are "indifferent to being", they vary with different conceptions of ordinary objects, and they are 'incomplete': they are, in short, "intentional objects" par excellence. That Meinong's "objects" deserve to be afforded any legitimate ontological status is not obvious; and that they, and only they, are the objects of consciousness is hardly more plausible than the claim that all our acts are directed toward the creations of our own minds.255

That incomplete objects, which in general are meaning entities, do not require any special ontological status, is entirely clear on Meinong's theory. But incomplete objects may be implected in other objects, which do possess an ontological status. What happens then is that the existent object is intended through the incomplete one. So our acts do not have to be directed only at the creations of our minds, because when we think about something that exists, we ultimately mean the external object and not just the one which is presented in our mind. The objection follows that what is directly intended is just the auxiliary object and only this object can be accessible to the mind, while the ultimate object remains beyond our reach. …for Husserl the Sinn is in no way the object intended in the act. The act entertains the Sinn, which is thereby embodied in the structure of the act, but the act is not directed toward its Sinn. For Meinong, though, the incomplete object is intended in the act. Indeed it is the only object of which the subject is strictly aware. Finally, it follows that physical objects are not objects of intention at all, properly speaking, for Meinong. We have suggested that the Meinongian might say they are 'indirectly' intended. But 'indirect intention' is then not a species of intention proper; rather it consists in the complex relation composed of intention proper and the quite different relation of embedment. And so our intentions never reach physical objects, for Meinong, but reach other quite different objects that are at best related to physical objects in some further way.256

Husserl's Sinn is compared here to the incomplete object, for the latter also plays the role of a meaning element. But in Husserl's approach to the 255 256

D.W.Smith and McIntyre, Husserl and Intentionality, p. 56-7. D.W.Smith and McIntyre, Husserl and Intentionality, p. 167.

186

ANNA SIERSZULSKA

question of meaning and reference, senses belong to a different sphere of entities than objects of reference. There are two realms, that of ideal senses and that of objects possessing physical existence. One would be inclined to think that it is very strange that there is no such division in Meinong's theory, that both meaning entities and the objects of reference belong to the common sphere of Aussersein. But there is a reason for this in Meinong's approach to matters of cognition in general. Meinong maintains that it is the most natural thing that the objects of cognition really possess the properties we perceive. The meaning objects are just incomplete in their properties, in relation to the ultimate objects of reference. This is why these ultimate objects also belong to the sphere of objects of intention, for they appear as the limiting cases for all the properties that could be potentially intended. We can say that just as truth is the limit for probability of objectives, the complete object is the limit for the incomplete ones that are implected in it. So it is always an incomplete object that is intended, but it can be identified with a real existing object outside the abstract realm. Therefore, it is not the case that, for Meinong, the ultimate object of the act is just the projection of its content, as the passage below assumes. Husserl's account of the predicative content of a Sinn makes it clear that (except in very special cases of what he calls "adequate givenness": see Ideas §§138, 142) the object of an act is not presented in the act as a mere projection of the Sinn: if it were, the object itself would then typically be indeterminate or incomplete, in the manner that Meinong envisioned. Rather, the Sinn may be thought of as an indeterminate conceptual 'frame', into which the object is intended as fitting in completely determinate, but as yet undetermined, ways.257

The fact that Husserl's object of the act is intended as completely determinate, but yet undetermined, is contrasted with the incomplete object of intention in Meinong's theory. However, there is no convincing argument that there is a real difference between intending an object as determinate and yet undetermined, and intending an object as complete in principle, but incomplete in the intention. Meinong recognizes that we do not ordinarily take the objects of our intentions to be incomplete objects; rather, we intend them as being complete, even though we do not know what properties complete them. The objects of our ordinary intentions 257

D.W.Smith and McIntyre, Husserl and Intentionality, p. 199.

MEINONG ON MEANING AND TRUTH

187

ought, therefore, to have (in some sense) the property of being complete, even though they are (in the sense we have been discussing) incomplete objects. Meinong calls such objects 'completed' incomplete objects. The details of this view are rather complex and obscure and, as Findlay says, it faces 'formidable difficulties'. The crucial point, though, is that even with the admission of these 'completed' objects our intentions still fail to reach complete objects: Meinong holds that the very best we can do is to intend completed (but nonetheless incomplete) objects that 'do duty' for them.258

Husserl's Sinn, likewise, does not comprise all properties and characteristics of an object to be presented in a single intentional act. Besides, one may have doubts if his subject of cognition actually reaches the world of real transcendent objects in a more secure manner than Meinong's . One would rather think that it is the other way round. While Husserl's transcendental ego is occupied with the contemplation of its ideal content, Meinongian subjects perceive objects together with their properties on a much more direct basis. Also Lindenfeld, in his Transformation of Positivism, exposes the presence of some patterns of Kantian tradition in Meinong's thought. He especially underlines the fact that the auxiliary semantic objects resemble Kantian phenomena while his ultimate objects are the counterparts of noumena. In contrast to Frege, for whom senses, and thus, various ways to refer to an object, depended on the linguistic expressions which can be used, Meinong's auxiliary objects are related to concrete mental experiences as to the level of their completeness and as to the selected aspects of the object they can capture. He distinguishes between phenomena and noumena, or 'near' and 'remote' objects, in which the former are the objects of immediate, untutored experience, where the latter are the objects as they 'really are' in the physical world. (...) In this manner the different levels of objects emerge from the context in which the similarities and differences are defined, rather than being specified in advance, as in Frege.259

Meinong's ultimate objects, however, only apparently correspond to what Kant called noumena. Noumena are indeed supposed to be inaccessible to 258

D.W.Smith and McIntyre, Husserl and Intentionality, p. 83-4. David F. Lindenfeld, Transformation of Positivism, Ch. VI The Linguistic Analogy and the Theory of Objects, p. 161. 259

188

ANNA SIERSZULSKA

cognition, which is not true about external objects in Meinong's theory. Although they can never be grasped completely, they can be apprehended and their diverse properties can be represented quite correctly in the intentional acts of the cognizing subjects. It is not at all the case with Meinong that the veil of phenomenal auxiliary objects would cover the real transcendent world from our eyes. To say 'the meadow is green' is to apply an incomplete object 'green' in which the specific shade of green is unspecified, to a concrete object, meadow, which does have a specific shade. The meadow then is complete. But, as said before, we can never apprehend an object in its completeness, we can only approach this goal by bringing in more objectives of so-being. (...) unlike Kant, Meinong postulates differences of degree in completion. He speaks at times of the complete physical object as the 'ultimate object' (Zielgegenstand), and of the incomplete object that we can grasp as the 'auxiliary object' (Hilfsgegenstand).260

Lindenfeld notices this difference from Kant in Meinong's conception, that there is a gradation of the degrees of completion. What he perhaps does not notice, is that introducing such a gradation implies a point of view entirely opposite to Kant's , because a gradation of completeness suggests that our grasp of objects may approach completeness to a greater or smaller degree. In Kant's theory the emphasis is upon the rationally preconditioned distortion of perception and impossibility of cognition. In Meinong's , the emphasis falls upon the success of cognition, in spite of difficulties. Meinong does not over-simplify the cognitive situation by denying that any problems actually arise. He is aware of the inaccuracy, aware of the sceptical arguments, and of the subjective and rational elements taking part in the process. But he remains a realist, always and in every case. When he hesitates to speak about truth, he speaks about probability, supported by evidence for surmise, as well as by the coherence of the system of objectives. As Lindenfeld remarks, and Chisholm before him, the theory of probabilistic evidence of objectives is Meinong's most original way to escape the sceptical consequences of the subjectivity and uncertainty of judgments. 260

David F. Lindenfeld, Transformation of Positivism, Ch. VI The Linguistic Analogy and the Theory of Objects, p. 162.

MEINONG ON MEANING AND TRUTH

189

(...) the legitimacy of such probabilistic evidence does not simply depend on intuition, but on the circumstances: one can arrive at the judgment that something is probable only on the basis of a number of objectives – not on a single one. One of Meinong's commentators, Roderick Chisholm, has called this doctrine of evidence for presumption 'one of the most important contributions of the philosophy of the twentieth century." Its innovation is in its denial that the only alternative to certainty is doubt and scepticism. Rather, we may legitimately be said to know something of which we are less than certain. This probabilistic evidence also applies to inductive knowledge of the type based on repeated experience – that is, a posteriori knowledge.261

The theory of probability degrees assigned to objectives in empirical contexts, provides a ready counter-argument to the objection of conception dependence. Because whichever conceptual scheme or theoretical approach is applied in the task of providing a description of reality, the resulting description is correct only if it gives an objectively true account of reality. This approach finds its expression in the degree of probability assigned to these objectives which are accepted as approximating truth most closely. Thus, according to Meinong's objective formulation of the notion of truth, regardless of the language and perspective, no matter which auxiliary objects are grasped in the intentional acts: a presenting act is only adequate, if its auxiliary object can be identified either with an existing or with a subsisting entity, and a judgment is only true if it apprehends a subsisting objective. In other words, it is not important which aspects of reality are apprehended and which are omitted, as long as those that are apprehended are correct. In the case of empirical judgments, there is no certainty whether they are true in this objective sense, but such uncertainty in the process of cognition does not imply that we have no possibility to acquire objective knowledge. That is because we know that many of our judgments are highly probable. Meinong's theory takes this fact into account and, simultaneously, it denies that there is any serious discrepancy between the ways we see the world and the things in themselves. On this view, conception dependence of the intentional grasp of reality does not result in idealistic consequences. No matter from which angle, we still apprehend something real. 261

David F. Lindenfeld, Transformation of Positivism, Ch. VI The Linguistic Analogy and the Theory of Objects, p. 168.

MEINONG ON MEANING AND TRUTH

191

PART IV MEINONG'S THEORY IN THE PERSPECTIVE OF PHILOSOPHICAL SEMANTICS CHAPTER 8 REFERENCE IN A MEINONGIAN SEMANTICS There are two dominant traditions in contemporary philosophical semantics. One of them has its roots in the semantic theory of Bertrand Russell and it is based upon the principle of extensionality. There is the traditional version of extensional semantics, where only existent objects and classes of them are taken into account, and its extended version, allowing also nonexistent objects and classes of them to play the role of the extensions of linguistic expressions. According to the extensionalist view, linguistic expressions have meanings as long as they have extensions. A semantic theory of this kind relates words to objects and sets of objects, without introducing any meaning entities that would mediate between the two elements of the semantic relation. The other tradition in semantics started with Gottlob Frege, who tried to account for the situations when we speak about the same object in many different ways, and also for the problems arising in connection with propositional attitudes. Although Russell's theory of descriptions made it possible to speak meaningfully about non-existent objects even within the traditional extensional semantics, satisfactory solutions to the other issues were not provided, and therefore, Frege's views turned out to be more appealing for many 20th century philosophers. Eventually, combined with Husserl's theory of intentionality, Frege's ideas have led to the development of the intensional line in semantics, the essential characteristic of which is that a distinct meaning element is taken into account. This meaning element, called sense

192

ANNA SIERSZULSKA

or noematic sense, is an abstract content related to each particular linguistic expression. From the point of view of extensional semantics, abstract contents or senses could never be treated as legitimate entities. However, some interesting indications how to treat sense entities in a more objective way when they are referents of non-denoting terms or in intensional contexts, and preserve extensionality in ordinary contexts, can be found in Meinong's writings, whose theory of meaning remains somewhere in between the two major semantic traditions. What is particularly worth preserving in Meinong's theory of meaning, is the introduction of abstract meaning entities as referents in the cases when meaning objects have to suffice as the proper objects of reference, i.e. when there is no other denotation available for a given linguistic expression. In all the other situations, existent objects are ultimately referred to, by means of incomplete meaning objects which are identified with the referents. Therefore, all linguistic expressions that refer to ordinary existent objects, have these ordinary objects as both their ultimate referents and denotations. There have been attempts to classify Meinong as belonging either to the intensional or to the extensional semantic tradition. In fact he belongs to both, because his theory is in perfect agreement with the traditional extensional semantics in the case of extensional contexts yet shares the advantages of intensional semantics for the other kinds of objects and for intensional contexts. We may observe, on the basis of Meinong's investigations, that the semantic relation between a word and its referent does not have to be simplified within the extensional approach to avoid speaking about meaning elements at any cost. On the other hand, there is no need to introduce any distinct meaning entities in the case of reference to ordinary objects in well determined extensional contexts, where Meinong identifies meanings with the ultimate referents of expressions. Accordingly, inspired by Meinong's work, both extensional and intensional logical systems have been developed, all of which have the special feature of admitting nonexistent objects into their semantic domains (in a more or less reserved fashion).

MEINONG ON MEANING AND TRUTH

193

♦ Intensional reference and denotational reference Some questions concerning Meinong's theory of intentionality and reference have already been discussed in Ch. 1 and Ch. 3. Here, I will repeat my previous definitions of different kinds of objects, which will be needed for further discussion. The content of a mental act (Inhalt) – a subjective sense entity, related to a particular act and non-repeatable. The immediate object of a mental act (unmittelbares Objekt) – an objective sense entity, repeatable (in an intentional relation) as the object of many different acts of different subjects. The ultimate object of a mental act (Zielgegenstand) – the object which is the ultimate target of intention: it may be, but need not be, merely a sense entity. It is the object of mental and linguistic reference. If external and existent, it may be the denotation of a word or expression. The auxiliary object of a mental act (Hilfsgegenstand) – the immediate object in its relation to the ultimate object, i.e. playing the role of a meaning entity. It is the meaning of a word or expression. The nearer object (nächster Gegenstand) and the remote object (entfernter Gegenstand) of a mental act – another way to refer to the auxiliary and ultimate objects. The incomplete object (unvollständig) – an expression used to describe the immediate object, in contrast to an existing ultimate object, which is complete in all its properties. The structure of an act of presentation according to Meinong: the immediate (nearest, auxiliary) object (the content is psychic)

Act

(content)

[the ultimate object] [there may be no ultimate object]

A very conservative 'Russellian' notion of denotation will be adopted, such that all objects denoted have to be either existent objects or proper abstract objects (ideal objects). In relation to all the other objects we can speak about reference, but not about denotation. Non-subsisting objects will

194

ANNA SIERSZULSKA

obviously belong to the 'broad' extensions of predicates, but not to their denotations. This is not in agreement with the current logical practice of using the term denotation in relation to all entities that are members of the domain, regardless of their character. But it will help to maintain the distinction between traditionally legitimate objects and 'semantic' entities of a Meinongian conception. Thus, if a linguistic expression has a denotation, what is presented by the act is a meaning entity, described in the theory as 'an immediate, nearest, auxiliary object', but we ultimately refer to the object denoted, described as 'the remote, ultimate object'. For example, we can ultimately refer to Napoleon by means of an auxiliary object such that its properties would be expressed by the description 'the victor of Jena'. In fact, we can have a whole range of abstract objects of the latter kind, which may be used to refer intentionally to the ultimate object. They are also characterized by various degrees of incompleteness. Naturally, a Meinongian semantics must admit nonexistent objects as well. They also belong to the meaning entities, but they cannot be identified with any existent ultimate objects. They have no being, except that they are considered as objects of Aussersein. I interpret Aussersein as the spectrum of all possible intentional objects beyond ontological concerns that contains meaning entities on an equal basis with real objects. Within this spectrum, I classify 'nonexistent' objects as objects of merely semantic character. The following table presents my interpretation of the relations between meaning, reference and denotation of expressions in a Meinongian semantic theory. Expression - denoting

- non-denoting

Meaning

Reference Denotation ultimate object - ultimate object auxiliary existing existing semantic object or subsisting or subsisting [Hilfsgegenstand] [Zielgegenstand] [Zielgegenstand] merely semantic merely semantic objectobjectnon-subsisting None non-subsisting [ausserseiender [ausserseiender Gegenstand] Gegenstand]

MEINONG ON MEANING AND TRUTH

195

According to this table, non-denoting expressions have meanings, which are abstract 'merely semantic' objects. Their meaning objects are the objects of intensional reference at the same time. But the objects of ultimate reference for denoting expressions are not merely semantic objects. They are ordinary existent objects with which the auxiliary objects can be identified on the basis of being implected in them, as subsets of their properties. A descriptive definition of pure objects in a Meinongian semantic theory, and a specification of different kinds of these objects, would look like this: All pure objects of intentional reference 'O' – objects in Meinong's broad understanding of this notion, all objective sense entities belonging to Aussersein, including objecta and objectives ('ProP' – propositions). All pure objects can serve as objects of reference, but not all of them can be regarded as denotations (only those that can be identified with existing individuals 'RO' or proper abstract objects 'IO'). Real objects 'RO' also belong to the spectrum of objects of intention as the limiting cases of intentional presentation. They are complete and existent. They can serve as denotations of expressions. Ideal objects 'IO' – the proper abstract objects in Meinong's theory, objects of mathematics and figures of geometry. These are subsisting entities. They do not exist but they are traditionally accepted as denotations. Auxiliary objects 'AxO' – incomplete semantic objects which are implected in real objects by possessing some but never all of their properties. (To be 'implected' is Meinong's term: a meaning-object is 'implected' in another object, either real or not, if all its nuclear properties are also the properties of the object in which it is implected, richer in the properties attributed to it. A meaning-object implected in an object that has being is taken to have 'implexive being'.) Objectives 'ProP' – Meinong's propositions, the meanings of sentences, they are divided into subsisting factual objectives (s-ProP) and non-factual objectives (n-ProP). Factual objectives apprehend relations and complexes. Semantic objects 'SO' – all objects which can play the role of meaning entities, including ideal (proper abstract) objects, auxiliary objects, factual objectives and merely semantic objects. Only real objects do not belong

196

ANNA SIERSZULSKA

here, since they are never the meanings themselves but rather what is ultimately intended and denoted by linguistic expressions. We can say that all semantic objects have the character of abstract objects as mindindependent entities that can be the objects of intention of many different subjects. ('Semantic objects' is not Meinong's term. The term is introduced to group under a general category all 'meaning-objects' in Meinong's theory that are not real existing objects but they still possess a non-subjective character.) Merely semantic objects 'MS' – semantic objects that do not possess any being themselves and they are not implected in any other objects which do possess being. They are what is commonly understood by the term 'nonexistent objects', which is wrong on Meinong's theory. (The term 'merely semantic objects' is introduced to replace Meinong's 'nonsubsisting' entities.) Merely semantic objects consist of non-subsisting objecta (MSO) and non-factual objectives (n-ProP). Diagram 4 on the next page is a graphic representation showing the interdependencies and the semantic roles played by different kinds of pure objects. Meinong's wide realm of objects includes both senses and referents of all linguistic expressions. This can be seen as a special theoretical advantage of Meinong's conception, and not a fault in the least. Because on the basis of this feature of his conception, a two-step reference theory can be formulated to provide a uniform theoretical treatment for all the traditionally non-extensional contexts. Apparently, sense-objects instead of contents work better in formal accounts of linguistic phenomena in such cases. Some of Meinongian-style semantics develop a two-step reference theory of this kind, while others split the pattern of reference with respect to each of the categories of referents, real and semantic objects. In agreement with the simple and clear Fregean distinction, with the addition of a specifically Meinongian modification, I will propose to analyse reference of all kinds of expressions in current Meinongian semantics as a two-step reference pattern. An expression primarily refers to an auxiliary abstract object (Fregean sense), which mediates its ultimate reference to the real object, if there is one (Fregean reference).

MEINONG ON MEANING AND TRUTH Diagram 4

197

198

ANNA SIERSZULSKA

Let's call these two steps of reference: intensional reference and denotational reference. The distinction is Meinongian in spirit, but it concerns current Meinongian-style semantics rather than Meinong's theory as such. The immediate intensional reference leaves us at the level of abstract semantic objects, where all expressions are referential. The ultimate denotational reference is identical with denotation and may provide some expressions with empty denotations. Thus, for non-subsisting objects this procedure will lack its second part – there is no denotational reference – and we will say that the expression only refers to an abstract semantic object (what I call 'semantic' objects are all possible objects of intensional reference). For intensional contexts, the ultimate reference of the expressions under examination will be their denotational reference, but the intensional reference will allow us to account for the difference in meaning and the resulting non-substitutivity of co-extensional expressions in such contexts. When we look at the comparison between Meinong's and Frege's conceptions of meaning in Ch. 3, we can see that the expression 'the Morning Star' refers to an object, which is identical with the object referred to by the name 'Venus'. But the reference is primarily to the object associated with the expression 'the Morning Star', with all the properties that are specifically associated with this expression. It is only in the second step of the procedure that we realise that this object is identical with the object of reference of the name 'Venus'. These two steps are in fact always present, because I am inclined to think that even proper names are associated with senses. Yet, for ordinary denoting expressions in extensional contexts, the mediation of intensional reference can be disregarded completely and we will say that they simply refer to ordinary objects, which is the usual strategy of Meinongian logicians. To sum up: A. For all kinds of terms in a Meinongian semantic theory we can distinguish between: - intensional reference 'the Morning Star' 'Chimera'

– to sense-objects, abstract referents – Morning Star – Chimera

MEINONG ON MEANING AND TRUTH

199

- denotational reference – to real objects, denotations 'the Morning Star' – Venus 'Chimera' – none The final two-step pattern involves a characteristic gradation of reference: Expression – Intensional Referent – Denotational Referent 'the Morning Star' – Morning Star – Venus 'Chimera' – Chimera – none For denoting expressions intensional reference is disregarded except in intensional contexts. B. Likewise, in the case of sentences we can distinguish between: - intensional reference – to objectives (propositions), abstract sense entities - denotational reference – to truth values (truth values are introduced by analogy with Frege, Meinong himself would rather have 'factuality and 'non-factuality') In intensional contexts, the intensional reference of sentences to propositions is more relevant. In ordinary extensional contexts, the denotational reference to truth values is the primary one. The main difference from other conceptions of reference is that, instead of senses and referents, we explicitly introduce objects of reference of two kinds: objects of intensional reference and objects of denotational reference. This is a device to provide a uniform theoretical treatment of all traditionally nonextensional linguistic contexts, related both to the occurrence of non-denoting expressions and propositional attitude operators. The conception of Meinongian reference presented above agrees with two kinds of interpretation of expressions introduced by Pa niczek (see Ch.9), and with my general view of a Meinongian-style semantics as proposing a double theoretical approach to objects. Admitting the intensional reference as a legitimate referential option for some nondenoting expressions, serves to mark a peculiarly Meinongian idea that all linguistic expressions which are intelligible refer to something, if only in the abstract sphere of semantic objects.

200

ANNA SIERSZULSKA

♦ The meanings of sentences and states of affairs Before proceeding to discuss different semantic characteristics of objectives considered primarily as the meanings of sentences, it will be useful to return one more time to the issue of their nature as theoretical entities. An objective may be defined as follows: an objective – an abstract objective sense of a mental act of judgment or assumption and an abstract sense of a declarative sentence. It is repeatable (in an intentional relation) in many acts of judgment. It does not contain any objects as its constituents. It is not existentially dependent, nor founded upon other objects, but it presupposes other objects in the sense of logical priority. An objective is the means to apprehend a complex and, at the same time, it is the integrating structure of the complex, which can obtain in reality and can be grasped in a judgment. The meaning / intending relation for Meinong's act of judgment: a sentence a mental act

the intended objective

=

the factual objective

≠

a complex (state of affairs)

In agreement with Meinong's version of logical realism, truth is the identity of logical structure between a fragment of reality and an objective intended. The objective obtaining in reality is a factual objective, but a factual objective is not an actual state of affairs. The closest entities to states of affairs in the sense of possessing real constituents are Meinong's complexes, but they cannot be identified with states of affairs or at least not on all conceptions of states of affairs. Complexes differ from objectives, because they are not abstract entities – they are fragments of reality apprehended by means of objectives. Meinong's 'factual objective' is not defined as an actual state of affairs or a complex, and therefore, the lower level objective which is the meaning of the subordinate clause in sentences like "It is a fact that snow is white", is not considered as a state of affairs, even though it is considered as a kind of 'object', if only of an abstract character. For this reason, the subordinate

MEINONG ON MEANING AND TRUTH

201

clause starting with 'that...' is not to be understood as a name of a state of affairs, as it could be e.g. in Husserl's conception. I said above in passing that propositions functioning as subjects were not presentations of judgements, but of the corresponding states of affairs. This point must be noted. Judgements as concrete experiences can, like things, be objects of possible perception, imagination and perhaps of some non-physical representation. They can then function as subject-objects in judgements, as happens when we judge about judgements. When such subordinate judgements are expressed, and not merely indirectly referred to (e.g. as 'this judgement', 'your judgement' etc.) a sentence will occupy the subject-position. But where a sentence occupies this position, it will not always serve to name a judgement. Judging about judgements differs from judging about states of affairs: having a presentation of or naming a judgement is likewise different from having a presentation of, or naming some state of affairs as a logical subject. If I say, e.g., 'That S is P is delightful' I do not think that my judgement is delightful.262

Two kinds of contexts need to be distinguished here. When we say “It is a fact that ….." and also when we say “It is true that…..", these are objectives of higher order on Meinong's conception. So the clause “that…" expresses the objective of lower order, and the higher order objective states about the lower order objective that it is true or that it is a fact. Therefore, the judgment is about the lower objective, but not about another judgment (in agreement with Husserl), and not about a state of affairs (in opposition to Husserl). On the other hand, when we take a sentence like “It is delightful that snow is white", we could probably say that it is delightful that things are as stated, but naturally not that the objective is delightful, and also not necessarily that the state of affairs is delightful if we do not believe that there are states of affairs. Apart from that, we may have doubts whether the latter sentence is a genuine expression of a higher order objective which really takes another objective as its subordinate object, or just a complex objective in which the that-clause does not form a name of anything. Inspired by conceptions close to Husserl's, J-L. Gardies explains that when we nominalize the content of a proposition, we obtain an object which may be well described as a state of affairs, while the subordinate 262

Edmund Husserl, Logical Investigations, p. 623.

202

ANNA SIERSZULSKA

clause following 'that' in a complex sentence, is actually a name of the state of affairs. That-clause stands for a state of affairs in this case. It is possible to nominalise not only a functor or the set corresponding to its extension, but also the content of an entire proposition. (...) And who would deny that the name that Peter is superior to Jack is a name very different from superiority, that it designates a fact which is quite distinct from the relation designated by the latter name? One may say that the term that or the fact that is a functor that forms the name of a state of affairs.263

However, there is nothing to prevent us from saying that what comes after the functor 'the fact that' is a name of a proposition, under the condition that we treat propositions as mind-independent objects, and that a fact consists in a proposition being true. In this way, it is possible to avoid postulating states of affairs, for these are objects which do not seem to be found anywhere in reality. We are left with abstract propositions as the meanings of sentences and, likewise, of the subordinate clauses of sentences. These are also not to be found in reality, but no claims of this kind should be made in relation to propositions. The main advantage of abstract propositions as meanings of sentences is that they can be correct about reality without being literaly pieces of it. Nevertheless, these two notions, of a proposition and of a state of affairs, are defined in various ways, making them sometimes almost identical and sometimes very different. In general, one must agree with Gardies that there is a lot of confusion in this area. The confusion between a proposition, its name, the state of affairs it asserts, and the name of this state of affairs is one of the major inadequacies of contemporary logic.264

At least one distinction of the above is easy enough to make, the one between the names of either states of affairs or propositions and the significations of these names. That is because names are linguistic expressions and their significations are not. Names 'stand for' something, but what they indeed stand for is another matter. Let's have a look at what

263 264

Jean-Louis Gardies, Rational Grammar, transl. by Kevin Mulligan, p. 193. Jean-Louis Gardies, Rational Grammar, p. 205.

MEINONG ON MEANING AND TRUTH

203

Gardies himself assumes that linguistic phrases stand for in the case of the following instance of Tarski's T-equivalence: What Tarski expresses by saying: the sentence 'snow is white' is true if and only if snow is white is that one can always establish an equivalence between the reality of a state of affairs and the sentence stating this state of affairs. (...) In the case of the four modifying predicates, however, it is not enough to say that the first two take as a subject a name designating a state of affairs, the last two a name designating a proposition, sentence or statement. Rather, these predicates stand in a double relation to one another, such that an equivalence can be formed between the proposition formed by is (or by is not) and the name of the state of affairs on the one hand, and, on the other hand, between the proposition formed by is true (or by is false) and the name of the proposition stating this same state of affairs. This equivalence is indeed the source of all confusion in this area.265

There is in fact some confusion also in relation to the above fragment, but for this reason it will be helpful in showing what the problem is with propositions, states of affairs and Meinong's objectives. Let us have a look at the above explanation of Tarski's T-equivalence. 'Snow is white' in the first clause is a name of the sentence in the object language, and snow is white in the second clause may be said to be a name of a proposition or even of a statement, but rather not a name of a state of affairs. So to explain in a more orthodox way what Tarski expressed, we would say that the T-equivalence formulated in metalanguage states that the sentence in the object language is true if and only if the proposition expressed by its translation into the metalanguage obtains. Therefore, we can say further that in the first clause appears the name of a sentence in the object language, while the second clause contains, as Gardies might say, the 'name' of a proposition in the metalanguage. We know that the second clause contains a translation of the sentence the name of which appears in the first clause. And so, the sentence in the object language must name the same proposition that is also designated by snow is white in the second clause. Yet, until now, we have no name of a state of affairs involved in 265

Jean-Louis Gardies, Rational Grammar, p. 208.

204

ANNA SIERSZULSKA

this explanation. Why not? In the first place, because the intuition of Gardies that truth should involve an agreement between a state of affairs in reality and a proposition does not find a straightforward expression in Tarski's T-equivalence, but merely in the requirement that in order to be true, the sentence in the object language must fulfil the satisfaction conditions specified by the theory. Secondly, because it must be equally possible to express truth conditions for false sentences and because we do not normally say that sentences should be translated from one language to another so as to preserve their designation of the same state of affairs. The point which I want to make here is that we could say so exclusively on the condition that we assume a sufficiently flexible conception of a state of affairs, which would include negative and unfactual states of affairs. Thus, the only difference between a proposition and a state of affairs would be that the former includes its subordinate objects, while the latter does not, but both would be 'reified' senses of sentences. This is a different notion of a state of affairs than the one I take to be basic. Let us make a distinction between a few related notions: a) an abstract state of affairs: a 'reified' sense of a sentence, does not have to consist of real objects, may be unfactual or impossible, absolutely every declarative sentence is a name of an abstract state of affairs. This notion is not typically meant by the term 'state of affairs'. b) a concrete state of affairs: a fragment of reality corresponding to a sentence or to a proposition, it consists of real objects, universal or particular properties and relevant relations, may be only positive, but never negative, sometimes merely possible, but never impossible, some sentences and propositions do not correspond to any concrete state of affairs. c) an abstract proposition: a 'reified' sense of a sentence, understood as a logical structure or a function, does not consist of any objects, may be negative, unfactual or impossible, absolutely every declarative sentence expresses an abstract proposition. d) a Russellian proposition: consists literally of real objects and a certain logical structure binding them, may be positive or negative, true or false, but not impossible, usually cannot contain nonexistent

MEINONG ON MEANING AND TRUTH

205

objects (except in the conceptions of Lambert and Zalta). This notion is not typically meant by the term 'proposition'. Now it is possible to say more precisely which of these notions are accepted here as close to Meinong's objective, and which of them are rejected. It belongs to the 'received' interpretation of Meinong's philosophy that objectives are states of affairs. I object to this interpretation only mildly in sense a), because an abstract state of affairs is almost indistinguishable from an abstract proposition as defined in c). However, I prefer to use the notion of an abstract proposition, since states of affairs are more often understood in sense b), and objectives are definitely not states of affairs in this sense, the reasons for which are explained in the previous chapters. I also do not think that Meinong's objectives should be interpreted as propositions in sense d), which shows some similarities with b). Sentences do not have entities of type d) or b) as their meanings, for the reason that, as it seems, one should either abandon the Russellian notion of a structured proposition, or drop the issue of 'merely semantic' objects. ♦ The reference of sentences in a Meinongian semantics It will be assumed in what follows that sentences have objectives both as their meanings and as their intensional referents, while the extensions of sentences are truth values. This is not to imply that truth values are objects of some kind on Meinong's theory, as they are for Frege, even though they are treated as 'objects' of the Meinongian domain by some Meinongian logicians. Before passing to the role of objectives as sentence referents, let us recall what has been said in Ch.3 in favour of treating them as functionlike entities, in a very broad non-technical sense: 1. An objective is described as the integrating factor in a complex.266 This means that an objective is not the complex itself, and the components of 266

A.Meinong, On Assumptions, p. 202: "Generally speaking, then, where there is a complex, there is also an objective as integrating factor in it, and one who wants to apprehend the complex cannot do it otherwise than by apprehending the objective, too. From that, it now becomes directly understandable, in the first place, why representing fails in the presence of complexes in the broader sense just employed – objects of higher order (...)"

206

ANNA SIERSZULSKA

the complex are not components of the objective. The objective is the integrating logical structure of the complex. In other words, the objective is the structure which obtains in reality without being a part of reality, in which it resembles a function. 2. The inferiora of an objective, which is an object of higher order, are external in relation to the objective. In the same way, the arguments are external in relation to a function: they do not belong to it. And therefore we can have true (obtaining) objectives concerning non-existent entities. 3. It is generally accepted that there are negative propositions, but no negative states of affairs. On Meinong's conception, a negation may be attached to an objective either externally or internally, and transposition of the negation is allowed as well. This is a feature typical for abstract (not complex-like) propositions that can be ascribed the status of a sentential function. 4. Objectives are treated as premises in formal inferences. Meinong is entirely in favour of using a symbolic language and other formal devices in such contexts. He even maintains that the body of knowledge should be formed by a logically consistent system of objectives. 5. Occasionally, Meinong treats objectives as incomplete structures, which can take different arguments. This is very well visible in situations where he substitutes one of an objective's 'arguments' by another. In such cases, the unchanged part of the objective plays the role of a propositional function with a slot to be filled by a variable argument. 6. Just as there are first-level and second-level functions in Frege's conception, there are also first-order and second-order objectives on Meinong's theory. Second-order objectives have other objectives, instead of objecta, as their 'arguments.' The above arguments show that Meinong's objectives are indeed very similar to Fregean functions in various ways, even though Meinong himself never used the term 'function'. The point when they become not so similar any longer is when Frege leaves his broad concept of a function in order to dwell upon the more precise mathematical one. Meinong's objectives cannot be compared to functions in this technical sense. In particular, the more technical sense of the notion requires that the exact

MEINONG ON MEANING AND TRUTH

207

law of correlation between the potential arguments of a function should be explicitly specified. A certain correlation between the objects of an objective can be observed as well, but the rules and the nature of this correlation are rather vague in comparison, as they are determined by means and within the context of natural language. For this reason, the similarity between Meinong's notion of an objective and Frege's notion of a function can only be considered to be of a very general character. In any case, let us investigate the proposal that objectives are functions which take arguments from the all-inclusive Meinongian domain of objects. The essential question that must be asked at this point, when it has been argued that objectives possess the character of functions, is "What is the value of the sentential function of an objective?" We will speak about 'sentential functions', for naturally an objective is taken to be a propositional function already saturated with arguments to form a proposition which can be evaluated as to its truth. The term 'sentential function' has obviously nothing to do with Meinong's terminology, it is Tarski's term which is standardly used to describe propositional functions after the substitution of all free variables with constants. We assume that the objective is the function itself, and therefore it cannot play the role of the value of the sentential function. We also have to rule out facts, as not suitable candidates to the position of function values. The reason is that Meinong's 'fact' reduces to factuality, i.e. to the 'obtaining' of an objective. So it is not an object of any kind, and could not become the value of a function. Metaphorically speaking, Meinong calls factual objectives 'facts', but such 'facts' are merely the functions that yield the value true, and not the values of these functions. In the discussion of Bergmann's views on this issue, in Ch.3, where facts are understood by Bergmann in the complexlike way, the problem arises how objectives, which Bergmann definitely classifies as simple entities (not consisting of other objects), can play the role of facts. This is nothing to blame Meinong about, since objectives are not facts in this substantial sense. What is important, and what Bergmann is right about, is that an objective is indeed a simple semantic entity. Being a simple entity, however, does not have to imply, contrary to what Bergmann thinks, that it cannot carry enough complexity to reflect the structure of a complex. This is how an objective can be factual, or true, by which it is meant that it obtains in a function-like manner.

208

ANNA SIERSZULSKA

Still no function can be its own value. We are left with a choice between either truth values or complexes as the values of sentential functions. This is practically the choice between rejecting or accepting objectual facts as constituents of reality. If we agree that sentences have truth values as their extensions, we simultaneously subscribe to the view that there are no facts to be found in the world. Let's quote an example of why this may be taken to be incorrect. A basic point is obscured by thinking of truth values as values. If one thinks in terms of facts, one easily sees that the function cannot be a constituent of the value, the fact, and also be what maps an argument onto such a value. The relevant constituent of a fact is a property or relation: what maps a term onto a fact is a function.267

The discussion in the paper quoted above mainly concerns complex relational properties, frequently found in Meinongian style logics. A property of this kind is a function. It is argued that such a relational property cannot both be a constituent of a fact, in the literal sense, and a function that maps an argument (the object possessing the property) onto this fact. The objection is directed against complex relational properties. On the view according to which the values of sentential functions are facts, such properties cannot be introduced without causing an inconsistency. But with respect to a Meinongian theory, this argument does not work. Complex relational properties that have the character of functions can well be maintained within a Meinongian logic precisely because complex-like facts are rejected! No property is ever a constituent of a fact, it is merely a fact that an object has the property. The proper values ascribed to the sentential functions are the truth values, not actual states of affairs. Which makes it also easier to ascribe some value to such sentential functions that do not obtain. According to the two-step reference pattern for Meinongian semantics, sentences refer primarily to objectives as their intensional referents, and by virtue of their objectives, or rather by virtue of the obtaining of their objectives, sentences refer to truth values as their denotational referents. The notion of 'denotation' as it is used here is very restricted and it applies 267

H. Hochberg, "Abstracts, Functions, Existence and Relations in the RussellMeinong Dispute, the Bradley Paradox and the Realism-Nominalism Controversy".

MEINONG ON MEANING AND TRUTH

209

only to real or ideal objects as denotations of terms and to truth values as the extensions of sentences. If there is a reference to other entities, these are called intensional referents. Terms refer to abstract objects as their intensional referents when these objects are merely nonexistent objects of discourse, or when differences in sense occur in propositional attitude contexts. Such 'sense-objects' as the Morning Star or the Victor of Jena are not called 'senses' on a Meinongian theory. The main theoretical difference is that they are considered to be legitimate semantic referents and to possess the properties ascribed to them. Therefore, they are objects rather than senses, which is why I propose to call them intensional referents. In the case of sentences, the intensional referents are always Meinongian propositions, i. e. objectives. They are function-like entities and they are also regarded as abstract semantic entities. Reference of sentences to objectives determines the ultimate denotational reference of sentences to truth values. In general, it is impossible to establish the values of sentences without a recourse to their intensional referents. The two-step reference pattern for sentences: Linguistic Expression – Intensional Referent – Denotational Referent SENTENCE

–

OBJECTIVE

–

TRUTH VALUE

Where in the case of sentences, the intensional reference is to objectives (propositions, abstract sense entities), and the denotational reference is to truth values. As it has been claimed already: in intensional contexts, the intensional reference of sentences to propositions is more relevant. In ordinary extensional contexts, the denotational reference to truth values is the primary one. With the help of the two-step reference pattern for sentence reference introduced here, it is possible to explain, against the old objections of Quine, why reference of sentences to objectives does not resemble the reference of singular terms to their objects. If we try to make a graphic comparison between singular term reference and sentence reference, it will look as follows:

210

ANNA SIERSZULSKA

Singular term

(intensional reference)

AxO

(denotational reference)

O [object]

Sentence

(intensional reference)

ProP

(denotational reference)

{1, 0}

The reference of singular terms to their objects that Quine and others speak about in this context would be classified on our analysis as 'denotational reference' to a real object of denotation. On the other hand, it is plain that the reference of a sentence to an objective is an instance of 'intensional reference' to a meaning-entity. The counterpart of this meaning-entity on the singular reference pattern is the auxiliary object – the intensional referent of a singular term – but not its denotational referent. Besides, an objective is a simple entity as a sense-object, while simultaneously it always carries complexity in itself, a sentence could never refer to an ordinary object of denotation. So there is a difference between the level of reference involved in this discussion. Sentences do not refer to ordinary objects of denotation, they refer to 'objects' from the realm of semantic entities. On the basis of this analysis it may be concluded that it is not required to avoid commitment to objectives by applying devices such as substitutional quantification, since they are reified complex senses of sentences rather than objects of any kind. Generally, it is assumed in Meinongian-style semantics that sentences 'refer to' or 'stand for' propositions, which are their intensional referents, as I propose to say, in contrast to their denotational referents or extensions which are truth values. For instance, Terence Parsons in Nonexistent Objects calls the truth value of a proposition its extension, while the sentences of his formal language 'stand for' propositions, which is roughly, but not exactly, what is meant by 'expressing' propositions, as he says. All this supposes that we have something in the language to stand for propositions I'm going to suppose that the sentences of O do this. This sounds slightly peculiar, since traditionally sentences either stand for nothing or else stand for truth values. There are two ways to dispel the peculiarity. First one could regard it as a mere matter of terminology; perhaps my 'stands for' is just an idiosyncratic misspelling of 'expresses'. Second, one could suppose that it is really that-clauses that stand for

MEINONG ON MEANING AND TRUTH

211

propositions, and within O the word 'that' is always written in invisible ink.(...) Let me call the truth value of a proposition its extension (...)268

This relation of standing for propositions can be regarded as a kind of reference to propositions. Sentences do not simply denote truth values, they stand for propositions in the first place, and the propositions, in their turn, possess truth values. But if propositions have truth values, they must be abstract semantic entities, just as Meinong's objectives are. They cannot be singular, or structured propositions, in the Russellian sense of complexlike entities. What can we understand, therefore, by Parsons' words when he speaks about Russellian meanings? …Russellian 'meanings', where the meanings are propositions, propositional functions, objects, sets, etc. (I call these 'Russellian meanings' because it was Russell who committed the sin of calling concrete things 'meanings'.)269

He says that on this view propositions are also objects of some kind, and object-like meanings are regarded as 'Russellian', but propositions are mentioned here as meaning entities together with propositional functions and sets, which are abstract. So if propositions are objects on Parsons' theory from Nonexistent Objects, they are probably abstract objects, especially that the Meinongian objects they consist of are constructed as sets of properties and need not be existent. Another example will be Dale Jacquette's Meinongian logic. His propositions can possess three truth values, in a semantics modeled upon a trivalent system proposed by Jan Łukasiewicz. This semantics is designed to "accommodate in the most straightforward way the non-standard truth values of propositions attributing nuclear properties to incomplete nonexistent objects."270 Jacquette's propositions involving nonexistent objects can be undecided in their truth value if they ascribe a property which is not determined for a given object, but they can also be true, if they ascribe truly a property which a given nonexistent object is said to possess. Thus propositions concerning nonexistent objects are undecided in 268

Terence Parsons, Nonexistent Objects, pp.108-9. Terence Parsons, Nonexistent Objects, p. 128. 270 Dale Jacquette, Meinongian Logic: The Semantics of Existence and Nonexistence, p. 95. 269

212

ANNA SIERSZULSKA

some cases, but they can also be true in other cases. It is possible to have true propositions about nonexistent objects. On this basis we can suppose that propositions of Jacquette's theory must be abstract entities, and that a true proposition neither needs to be a fragment of reality, nor needs to denote such a fact in the world. All that can be said about Zalta's proposal is that it departs very far from Meinong's original conception, which is not necessarily a fault in itself, but one may wonder if he still presents a Meinongian-style solution. The doubt appears as we read that the notion of a proposition he assumes is straightforwardly Russellian: Our conception of propositions derives from Russell. Intuitively, propositions are structured entities that have constituents and a logical structure based on exemplification. The constituents of a proposition are the relations and objects that it has as parts. These parts can be organized into all sorts of complex logical structures. The metaphysical truth or falsity of these logical complexes is basic.271

The metaphysical truth of a Russellian structured proposition seems to amount to the literal existence of such a fragment of the world. There is nothing else that makes a proposition true because this kind of proposition is the fact in the world itself. The metaphysical truth of a proposition is therefore basic for Zalta. note: On our conception, the truth of a proposition is basic – it is not evaluated with respect to anything else. Facts are just true propositions.272

On the other hand, the truth of sentences is supposed to be derivative, and not metaphysical but merely semantic, because it depends on the truth of the propositions which are denoted by sentences. But in propositional attitude contexts and in nonexistence contexts the propositions denoted by sentences are of a completely different character. These are abstract propositions consisting of abstract objects that Zalta calls 'cognitive contents' of the terms. The embedded sentences denote propositions, but whether they denote a proposition constructed out of the denotations of its terms or one constructed out 271 272

Edward Zalta, Intensional Logic and the Metaphysics of Intentionality, p. 57. Edward Zalta, Intensional Logic and the Metaphysics of Intentionality, p. 182.

MEINONG ON MEANING AND TRUTH

213

of the cognitive contents of its terms depends on whether the report in question is de re or de dicto.273

Depending on the context, Zalta claims that sentences sometimes denote fragments of reality and sometimes abstract propositions. We can observe that there is a duality in the reference of sentences which reflects the division of propositions into complex-like Russellian ones and abstract ones. Such a solution may perhaps be welcome if someone wants to preserve a substantial notion of a fact as part of the world and at the same time offer an account of truth in non-extensional contexts. The split-reference pattern for Zalta's sentences looks as follows: Sentence –

Complex-like Proposition (fact)

Sentence –

Abstract Proposition

(in extensional and modal contexts) (in attitude and nonexistence contexts)

The objection in this case is just that the solution is not essentially Meinongian. On Meinong's theory, there is a uniform treatment of objectives. Regardless of the existence or nonexistence of the objects that the objectives are about, the objects are never actually contained by the objectives, but they remain external to them. It is clear that all Meinongian semantics accept some propositions about nonexistent objects as true, and so they have to accept that there are abstract propositions and that truth values are the extensions of sentences, but there are no facts that these sentences would refer to. Thus sentences refer to propositions which are more naturally accepted as abstract entities – intensional referents of sentences – while their extensions depend upon whether the propositions referred to actually obtain.

273

Edward Zalta, Intensional Logic and the Metaphysics of Intentionality, p. 169.

MEINONG ON MEANING AND TRUTH

215

CHAPTER 9 THE SPECIFIC FEATURES OF A MEINONGIAN SEMANTICS If the domain of pure objects is understood in the characteristic Meinongian way, it contains both real and meaning-objects of different kinds. This feature of a Meinongian-style semantics is responsible for what may be called a double theoretical approach to objects. Two senses of being, of quantification, of predication, of extension and of linguistic reference are required. These typical characteristics of a Meinongian semantics will be discussed in this chapter, while the next chapter will deal briefly with some issues related specifically to propositions, and some consequences of a one world or possible worlds approach in a Meinongian semantic theory. The present work concerns Meinong's ideas about truth, and so the semantic or logical issues are only investigated as related to this subject. Only four representative Meinongian logics will be considered here, all of which are not especially strange from the perspective of classical logic. These are the theories of Parsons, Zalta, Jacquette and Pa niczek. All of them share the model-theoretic paradigm in different extensional or intensional, modal or non-modal versions. ♦ Two notions of reference and two notions of extension It is the feature of a Meinongian semantics that, regardless whether it assumes a one world or a possible worlds approach, there is usually a single domain of objects. The objects are either all in one set, or the domain is subdivided, but not in such a way as to separate the abstract and ordinary objects, which appear side by side with each other. We can imagine someone thinking that this is wrong, or at least very strange. But this all inclusive domain of objects is not a reflection of reality, it is actually a reflection of the natural epistemic situation, when we can consider diverse epistemic options all at the same time.

216

ANNA SIERSZULSKA

No wonder that reference to abstract objects of different kinds (auxiliary objects, non-subsisting objects or propositions) is allowed and legitimate in Meinongian semantics. The notion of denotation traditionally applies only to ordinary existent objects, and maybe it would be better to leave it this way, so as to avoid misunderstandings. The distinction between intensional reference and denotational reference is introduced as useful for the explanation of intensional contexts, and also to differentiate between the ordinary denotations of expressions and the reference of nondenoting terms. Two notions of reference: reference to abstract objects and reference to real objects, are present, although not explicitly distinguished, in all Meinongian semantics. Reference in Meinongian logics The semantics of Zalta's intensional logic takes relations, including properties, to be basic and fundamental, which means that they are not defined by means of their extensions in the actual and each possible world. Relations are considered to be indefinable and known to us by acquaintance, so their intensional differences are always preserved. Relations possess two kinds of extension, exemplification extensions consisting of ordinary individuals and encoding extensions consisting of abstract individuals. Both 'abstract' and ordinary objects are members of Zalta's domain, and abstract objects are also called 'individuals', even though they are represented just as sets of properties or complex relational properties which they encode and exemplify. Zalta's 'abstract' objects include all such entities that do not possess existence – they are the counterpart of what I call 'semantic' objects so as not to confuse them with proper abstract object of mathematics. 'Abstract' individuals are 'denoted' by non-denoting expressions and by denoting expressions in epistemic contexts. We can notice that something like a split-reference pattern is assumed in the way Zalta analyses propositional attitude contexts on the ground of his intensional logic. (...) 'the Morning Star' has a reading on which it denotes Venus, and a reading on which it denotes the A-object that encodes all of the relevant information implied by being a Morning Star. In ordinary contexts, the properties relevantly implied are the same: being a heavenly body, disappearing in the morning, being the last

MEINONG ON MEANING AND TRUTH

217

heavenly body to disappear in the morning, etc. At least, these are the properties typically connected with 'the Morning Star.' The property of being an Evening Star is not typically connected, however (at least not until we discover that the two are the same).274

Zalta's denotation shows a certain duality. Sometimes 'the Morning Star' denotes Venus and sometimes it denotes an abstract object – the Morning Star. He observes that the property of being an Evening Star is not typically implied by the expression 'the Morning Star', and it is not encoded by the abstract object denoted by this expression. Clearly, Zalta uses the term 'denotation' with respect to his A-objects in the same sense that I use the term 'intensional reference', as distinct from denotational reference. Nevertheless, Zalta denies that he utilizes two semantic relations. Moreover, the formal language we shall use to preserve at least some of Frege's insights concerning senses does not utilize two semantic relations. We will get by with just the one semantic relation of denotation. The senses of the terms of English will be denoted by special terms of our language.275

Implicitly, he has two kinds of 'denotation' in his theory, because denoting senses must certainly be distinguished from denoting real objects. And perhaps the term 'denotation' is not very fortunate every case. But due to this duality, his theory captures the typically Meinongian idea of senses as abstract objects associated with subsets of the ultimate object's properties. Nevertheless, the theory does not reflect the Meinongian gradation of reference, first to sense-objects, next to things. What is entirely untypical for Meinongian semantics is that at the bottom of this theory of abstract objects as sense entities, Zalta is an fact an advocate of the theory of direct reference, whenever possible. He is reluctant to introduce senses into the relation of reference, even in the form of his property-encoding A-objects. Convincing arguments by Kripke, Donnellan, Putnam, and Kaplan suggest that terms directly denote, that is, no intermediate entity plays a role in securing or determining the denotation of a term. Since many philosophers think that Fregean

274 275

E.Zalta, Intensional Logic and the Metaphysics of Intentionality, p. 229. E.Zalta, Intensional Logic and the Metaphysics of Intentionality, p. 154.

218

ANNA SIERSZULSKA

senses are the entities that determine or secure the denotation of the term with which they are associated, Fregean senses are rejected altogether.276

In the end, he assumes that although there are no entities that secure or mediate the reference of expressions, terms can be regarded as possessing senses, which may have explanatory value in some contexts.277 Still, he claims that names and also definite descriptions are rigid designators in all possible worlds, they always denote the same objects as in the actual world, even though there are sense-objects ascribed to them. These senseobjects are signified only in de dicto attitude contexts, and otherwise they play no role at all in determining the reference of an expression. In general, our method of analyzing the attitudes is consistent with the results of the theory of direct reference. Every name or description of type t has a sense that is an A-object of type t. In attitude contexts, such terms signify their senses when they are in de dicto position. No problems are encountered when building up the intermediate proposition signified by embedded sentences containing such terms, since the sense and the denotation of the term are of same type. Given our work in Chapters 9-11, the Twain/Clemens case and the woodchuck/groundhog case do not present special puzzles. Their representations are consistent with the view that the names in question are rigid.278

It is perhaps worth observing that Zalta finds sense-objects, in opposition to senses as abstract mental contents, unquestionably more useful in the theoretical representation of non-extensional contexts, exactly because they are perceived as objects of some sort ('intensional referents' on my distinction). Zalta's split-reference pattern with two notions of 'denotation' could be represented as follows: 'the Morning Star' –

Venus (denotation1): in extensional and modal contexts

'the Morning Star' –

A-object the Morning Star (denotation2): in attitude contexts

'Chimera'

A-object Chimera (denotation2): in all contexts

276

–

E.Zalta, Intensional Logic and the Metaphysics of Intentionality, p. 153. E.Zalta, Intensional Logic and the Metaphysics of Intentionality, p. 154. 278 E.Zalta, Intensional Logic and the Metaphysics of Intentionality, p. 223. 277

MEINONG ON MEANING AND TRUTH

219

What Zalta calls denotation, Jacquette calls designation. There is also one semantic relation in his object-theory logic. Naturally, any Meinongian object, existent or not, can be designated by the terms of his language. In contrast to 'denotation', the notion of 'designation' does not suggest that the object which is designated should be a real object, so Jacquette's terminology seems to be more suitable for the logic of nonexistents than Zalta's. The most important primitive semantic relation is designation, not of existents only, but of any object or set of objects in the semantic domain of existent and nonexistent Meinongian objects.279

The basic system he presents is not modal and does not assume that there are any possible worlds. To make his logic intensionally sensitive, he introduces three different definitions for the identity of objects. The highest level of identity in all contexts is expressed by Leibniz's Law of the identity of indiscernibles. Jacquette's definition is based upon the distinction between nuclear (ordinary) properties and extranuclear ('external') properties of objects (about this distinction later, in the section concerning predication). For any objects x and y, x is intentionally identical to y if and only if, for all nuclear properties P, x has P if and only if y has P.280

This is called the intentional identity of objects, to be distinguished from their referential and extensional identity. To be on the safe side, since the objects which are referentially identical are not identical in all their properties, Jacquette proposes to use the notion of 'referential codesignation' instead of 'identity'. This definition employs the notion of converse-intentional properties, which include being believed, feared, etc. referential codesignation: the objects are referentially identical if and only if they share all non-converse-intentional nuclear properties.281

279

Dale Jacquette, Meinongian Logic: The Semantics of Existence and Nonexistence, p. 101. 280 D. Jacquette, Meinongian Logic: The Semantics of Existence and Nonexistence, p. 118.

220

ANNA SIERSZULSKA

We can understand that objects are referentially identical if they share all their nuclear properties except for those which are called 'converseintentional'. In this sense, 'the winged horse' and 'Pegasus' are referentially identical, even though someone may believe that he saw the picture of the winged horse without believing that he saw the picture of Pegasus. Finally, there is the extensional identity, or extensional codesignation: (...) objects are extensionally identical if and only if either the objects exist and are referentially identical, or are nonexistent. This entails that Cicero and Markus Tully are extensionally identical, even as the objects of thoughts in which incompatible properties are attributed to them, since Cicero is an existent object and is referentially identical to Tully. Any arbitrary nonexistent impossible or incomplete object is extensionally identical with any other.282

All nonexistent objects are extensionally identical, or perhaps it would be better to say that they are extensionally indistinguishable. To be more precise, their denotation is the empty set. The term 'extensional' is used by Jacquette in the traditional, existence-implying sense. But referentiality does not imply existence, and so nonexistent objects can be referentially distinct. However, if they are referentially identical they must be also intentionally identical. Existent objects, to the contrary, can be referentially identical but intentionally distinct. The split-reference pattern for Jacquette's theory would look like this: 'the Morning Star' –

Morning Star (represented designation in attitude contexts)

'the Morning Star' –

Venus (designation in extensional contexts = denotation)

'Chimera'

–

Chimera (designation in all contexts)

'Chimera'

–

the null set (denotation in all contexts)

281

as

description,

D. Jacquette, Meinongian Logic: The Semantics of Existence and Nonexistence, p. 118. 282 D. Jacquette, Meinongian Logic: The Semantics of Existence and Nonexistence, p. 120.

MEINONG ON MEANING AND TRUTH

221

(The objects Morning Star and Venus are extensionally and referentially identical, but they may be intentionally distinct.) According to Parsons, not all grammatically correct expressions are referential, but there are some non-denoting names and descriptions which refer to nonexistent objects of the literary tradition and mythology. For one thing, it is intended that some names refer to objects that don't exist. In particular, it is intended that 'Pegasus' refers to a certain object, to the winged horse of Greek mythology.283

Objects such as Pegasus are genuine objects of reference for Parsons, on equal rights with existing real objects. He does not make any difference or distinction between the reference to abstract objects and to ordinary objects, which does not have to mean that he believes that non-denoting terms have a denotation. To counter this impression, he formulates an 'orthodox theory' by means of which he presents a model of his objecttheory logic. The orthodox theory assumes that there is a nonempty class of individuals (existing objects), there are properties of individuals, classes of properties of individuals and relations among classes of properties of individuals. These are used to picture the objects of the Meinongian domain in the following way: Next, we will represent objects by classes of properties of individuals, that is, by classes of nuclear property representatives. In particular, we will now call any such class an object representative. We can picture an object's having a nuclear property as follows: we say that an object representative has* a nuclear property representative if and only if the latter is a member of the former. (...) We will say that an object representative exists* if and only if there is some individual i such that the object representative is the class of (orthodox) properties (of individuals) that i has. Finally, we will designate relations among classes of nuclear property representatives (i.e. relations among object representatives) extranuclear relation representatives. And we will say that an object representative has* an extranuclear property representative if and only if it has it (in the orthodox sense of 'has').284

On the orthodox picture, all objects are constructed as sets of properties, but while existent objects can be identified with individuals, nonexistent 283 284

Terence Parsons, Nonexistent Objects, p. 120. T. Parsons, Nonexistent Objects, p. 89.

222

ANNA SIERSZULSKA

objects cannot. Therefore, we can observe that the reference of linguistic expressions is gradual. Primarily, they refer to their representations as sets of properties, and so to the Meinongian objects, but whether they have a denotation depends on the presence in the basic domain of an existent individual that can be identified with a given Meinongian object. Furthermore, nonexistent objects disappear on this view as reduced to sets of properties only. Their possession of nuclear (constitutive) properties amounts to set membership, while other properties, like being incomplete or impossible (extranuclear), are properties of sets of properties, not of objects. The interesting issue is that Parsons himself is opposed to his orthodox theory, in the sense that, even though the nonexistent objects do not exist, he actually claims that there are particular nonexistent objects, like Sherlock Holmes or Zeus, playing the role of the referents of linguistic expressions, because when we think about such objects, we do not think merely about sets of properties.285 Parsons presents a modal intensional system which is sensitive to the differences of the sense of expressions specifically in propositional attitude contexts. It is worth noting that for Parsons also proper names are associated with senses and they are not regarded as rigid designators in attitude contexts. The senses of names are not reducible to definite descriptions in any straightforward way. I have heard it suggested that proper names, unlike definite descriptions, do not manifest de re – de dicto scope ambiguities. This is sometimes thought to follow from the view that they are rigid designators, where a rigid designator is defined to be a name that names the same object with respect to every possible world. But all that follows from this is that proper names do not manifest de re – de dicto ambiguities with respect to modal notions. Propositional attitude words are another question entirely. Note: There is another use of the term 'rigid designator' according to which a term is a rigid designator just in case it contributes nothing to the proposition except its referent (Kaplan); in this sense, names are clearly not rigid designators, since coreferential names are not interchangeable in epistemic contexts salva veritate.286

Parsons' note about Kaplan's notion of a rigid designator concerns what I call denotational reference, since indeed, names which are coreferential in 285 286

T. Parsons, Nonexistent Objects, pp. 93-7. T. Parsons, Nonexistent Objects, pp. 121-2.

MEINONG ON MEANING AND TRUTH

223

this sense are not interchangeable in epistemic contexts. So to disambiguate the situation, we could say that Kaplan's rigid designators would contribute to the proposition only their denotational reference, while in fact names contribute both their intensional and denotational reference to a proposition, which is why they are not rigid designators that could be exchangeable in all contexts. In order to distinguish the contexts of nonrigidity in the formal representation, Parsons introduces propositional attitude constants to represent belief and other attitude verbs. Within attitude constructions, coreferential terms are not assumed to be exchangeable. The representation of Parsons' two-step reference pattern might look as follows: 'the Morning Star'

–

Morning Star (a Meinongian object, in attitude contexts)

–

Venus

'Chimera'

–

Chimera (reference in all contexts)

–

empty set (denotation)

(Venus is an individual of the basic domain; ultimate reference in attitude contexts and basic reference in extensional and modal contexts = denotation) We can see the division between denotational reference in the case of existent objects and intensional reference for the expressions referring to nonexistent objects. There is also the Meinongian gradation of the reference of an expression, first to the intensional referent and then to the denotational referent. Pa niczek is perhaps closest to making an explicit distinction between two kinds of reference in his Meinongian semantics. In fact, he does not use two notions of reference, but he introduces the notions of primary and secondary interpretation.287 The primary interpretation of a term is what he calls M-referent, and it is an individual from the basic domain of interpretation. The secondary interpretation of a term is M-meaning which is a M-object constructed as a set of sets of individuals which are the

287

J. Pa niczek, The Logic of Intentional Objects, p. 108.

224

ANNA SIERSZULSKA

extensions of the properties ascribed to this M-object, e.g. golden mountain {set of golden things, set of mountains}. Every name expression is correlated with a meaning represented by M-meaning. With some of those name expressions – denoting names- an M-referent is correlated, although not necessarily an individual one as in Fregean semantics. (...) M-meanings and M-referents are of the same ontological category- they are just M-objects. (...) in case of empty names: these M-meanings do not represent or mediate the reference- they are the referents per se.288

When there is an existent referent, the mediation of M-meaning is disregarded and the reference is said to be directly to the M-referent. Sometimes such M-referents may be not individuals, but 'general objects' (sets of individuals), when the expression is a general one. For some name-expressions there are no M-referents in the domain of interpretation (i.e. no denotations, or one could say that there is no denotational reference). In such cases, the expressions refer to M-objects which are their M-meanings. According to my distinction, non-denoting expressions in Pa niczek's theory possess only intensional referents. It is an interesting issue how the two objects of primary and of secondary interpretation, namely the M-referent and the M-meaning can be identified with each other. The M-meaning is an incomplete meaning object, possessing only the properties specified, the M-referent is an ordinary individual of the domain, which is complete with an infinite number of properties. They are not identical in any strict sense. To explain this specifically Meinongian relation of identity by virtue of being implected in each other, Pasniczek uses the notion of ontic identity. One can object that the referential identity [between M-meaning and M-referent] is not the identity of M-objects at all. However, we call this relation an ontic (or metaphysical) identity (..)289

One object may be identified with the other if its properties form a subset of the properties possessed by the other object. In this way, a general object, such as most sense-objects related to linguistic expressions, can be identified with a particular individual, even though the relation of identity 288 289

J. Pa niczek, The Logic of Intentional Objects, p. 111. J. Pa niczek, The Logic of Intentional Objects, p. 120.

MEINONG ON MEANING AND TRUTH

225

does not obtain according to Leibniz's Law. Pa niczek calls this relation an ontic or metaphysical identity between the object as it is intentionally presented and the same object in reality. On this ground, it is possible for Pa niczek's theory to preserve the Meinongian gradation of reference. The primary interpretation of an expression is its ultimate referent, but the reference is mediated by the object of secondary interpretation – M-object (my intensional referent). The M-object, constructed as a set of sets of individuals, is certainly quite remote from an ordinary referent, yet, depending on the properties ascribed, it may be identified with an individual in the domain. The pattern of reference for Pa niczek's M-logic: 'the Morning Star' –

'Chimera'

–

Morning Star (M-object: meaning)

–

Venus (M-referent: an existing individual)

Chimera (M-object: meaning and referent)

–

∅

Generalizing the patterns of reference for different Meinongian semantics which have been presented above, we can say that linguistic expressions refer to their denotational referents in extensional and modal contexts only. In propositional attitude contexts and in the case of nondenoting expressions the reference is to semantic objects which can be called their intensional referents. In the theories of Parsons and Pa niczek, there is also a gradation of reference typical for Meinong's conception, because a Meinongian object constituted by a set of properties is ultimately identified with an individual in the basic domain. Expression

–

Intensional Referent (set of properties)

–

Denotational Referent (individual)

The theories of Zalta and Jacquette do not follow this two-step pattern of reference. In their theories, the reference is split depending on the context of interpretation.

226

ANNA SIERSZULSKA

Expression

–

Denotational Referent (in extensional and modal contexts)

Expression

–

Intensional Referent (in attitude contexts and in the case of non-denoting expressions)

The feature which is shared by both the two-step gradation pattern and the split-reference pattern is that the entity which plays the role of the 'sense' of expressions in both cases, is understood as a kind of object. This object is the intensional referent of an expression whenever the expression either does not possess a denotational referent, or the context of interpretation involves a propositional attitude. Intensional referents are genuine and legitimate objects of reference in a Meinongian semantics. They allow for a simpler and clearer treatment of a variety of semantic issues. A 'broad' notion of extension Apart from the two-step reference pattern and two kinds of referents, a Meinongian semantics involves either two notions of extension, or just one very broad notion. The broad notion of extension corresponds to Meinong's uniform treatment of all objects as possessing properties, regardless of their being or non-being. The Meinongian domain contains not only abstract and merely nonexistent, but also impossible objects, and thus the extensions of predicates become quite something else than ordinary extensions. It should be observed that in modal Meinongian logics the possible worlds are not constituted solely of would-be existent objects. The objects belonging to the extensions of predicates in each world show some diversity as to their existential status. An extension is, therefore, no longer a set of ordinary existent individuals, because nonexistent objects may be its members as well. Lambert explains that this is characteristic of the contemporary notion of extension, which differs from the traditional notion. Contemporary talk about extensionality has somewhat distorted the traditional notion. For example, the extension of a general term, in the contemporary parlance originating in Carnap's classic Meaning and Necessity, say, 'winged horse', is customarily taken to be simply the set of objects it is true of, as opposed to the

MEINONG ON MEANING AND TRUTH

227

intension of that term (a property, a function, etc.) So construed it would comprise, for instance, merely possible objects (such as Pegasus) as well as actual objects (such as Meinong). It is not surprising therefore, that traditional and current talk of extensions can yield different verdicts about whether a given theory of predication violates the principle of extensionality, a principle which demands that coextensional general terms be everywhere substitutable salva veritate.(...)in the contemporary mode, Meinong's logic may be every bit as extensional as 'extensional logic' ; in the traditional mode, it certainly is not...290

Therefore, if the round square is an object in our Meinongian domain and we accept that the sentence 'The round square is round' is true, then the round square belongs to the extension of the predicate in question. We have an extensional semantics always when properties are defined extensionally as sets of objects, no matter if these objects are existent or possible or impossible. The distinctive feature of an extensional semantics is that coextensional properties cannot be distinguished, as opposed to intensional semantics, making a difference between the senses of expressions. It is a specific characteristic of a Meinongian semantics that it can be intensional even though it does not introduce possible worlds. We will see on the example of Jacquette's semantics that if the broad extensions of predicates are taken into account so that they include semantic objects referred to in attitude and nonexistence contexts, this results in an intensionally sensitive semantics. Only Pa niczek's one-world semantics is entirely extensional, for only the existent individuals from the basic domain are included in the extensions of predicates in all cases. Zalta's semantics, on the other hand, is essentially intensional, since all properties are intensionally distinct by acquaintance, prior to defining their extensions. A Meinongian semantics may have to introduce semantic objects into the extensions of predicates, or, like Zalta's, it may introduce double extensions. Zalta's A-objects do not exemplify properties, they only encode them and for this reason, while existing objects belong to ordinary extensions of predicates, the 'abstract' objects belong to their encoding extensions.

290

Karel Lambert, Meinong and the Principle of Independence, pp. 64-5.

228

ANNA SIERSZULSKA

Since encoding is a kind of predication, we may regard properties as having a second kind of extension. In addition to having an exemplification extension, which consists of just those individuals that exemplify the property, they also have an encoding extension, which consists of just those individuals that encode the property.291

We can see the parallel within Zalta's theory between two kinds of extension and two kinds of denotation of expressions. This is related to the fact that the domain contains individuals of two kinds: existent and 'abstract' individuals. Zalta makes it clear that his denotations do not correspond to ordinary exemplification extensions, because in that case his system would be purely extensional, its intensionality would be lost. In what follows, we shall not use the method of intension and extension – the terms of our language receive only denotations, relative to an interpretation of the language and an assignment to the variables. Moreover, Substitutivity is preserved intact in our logic – :there are no denoting terms, or contexts, for which it fails. Truth is always preserved when terms having the same denotation are substituted for one another. This last fact could be a source of confusion, if one were inclined to think of the denotation of a term as corresponding in some way with the extension of a term. For then one could describe our system as purely extensional (...)292

The denoting expressions of the language denote either existent objects or 'abstract' sense-objects, depending on the context. The former belong to the exemplification extensions of predicates and the latter to the encoding extensions of predicates within each 'possible world'. Zalta's denotations of both kinds are not relative to world-time pairs, because only the extensions of predicates change from one world to another. As we have already seen above, names and definite descriptions are rigid designators. Their denotations are always A-objects in the case of nonexistents. Otherwise, Aobjects are used for explanatory purposes in propositional attitude contexts, on which occasions terms denote such sense-objects instead of ordinary individuals. The immediate consequence of this situation for the truth value of propositions is that the membership of existent individuals in the exemplification extensions of predicates need not coincide with the 291 292

E. Zalta, Intensional Logic and the Metaphysics of Intentionality, p. 29. E. Zalta, Intensional Logic and the Metaphysics of Intentionality, p. 9.

MEINONG ON MEANING AND TRUTH

229

membership of the A-objects – denoted instead of these individuals in attitude contexts – in the encoding extensions of the same predicates. An existing individual will typically exemplify more properties than the properties encoded by a corresponding A-object in a certain attitude context. Thus 'abstract' objects are not admitted into the ordinary extensions of predicates together with existent objects, but there is a second notion of extension concerning 'abstract' objects specifically, the encoding extension, which is important for the functioning of Zalta's semantic system. Parsons has only one broad notion of extension which includes nonexistent objects, such as he considers to be legitimate ones, i.e. present in the linguistic practice. He admits that the primary extension of a predicate consists of real objects, but these are not all the objects belonging to the full extension of the property for nuclear properties. We can suppose that the extension of an extranuclear property may sometimes consist only of real objects (e.g. the property of being existent), or entirely of nonexistent objects (the property of being incomplete). (...) whereas the primary extension of a nuclear property is a subclass of the real objects, such a property will also be possessed by certain unreal objects. The class of all objects which possess a property will be called the full extension of that property.293

The identity conditions for nuclear properties on Parsons' intensional logic are defined according to their full extensions and not just their ordinary extensions. The extensions of identical properties must be indentical even with respect to the membership of impossible objects. The notion of nuclear property being used here is something like C.I.Lewis's notion of the comprehension of a predicate. Lewis identified this with the set of all possible objects to which the predicate is applicable. Our identity conditions for nuclear properties are actually even tighter than this, for nuclear properties are identical only if they coincide for impossible objects as well. (This may make them closer to Lewis's notion of the intension of a term (...)).294

293 294

T.Parsons, Nonexistent Objects, p. 79. T. Parsons, Nonexistent Objects, p. 75.

230

ANNA SIERSZULSKA

Naturally, the same property may have different extensions relative to world-time pairs, but since Parsons identifies world-time pairs with different interpretations of the language that agree with the basic interpretation "about which predicates stand for which properties"295, we can speak about the identity of nuclear properties as the full extensions of these properties on a single interpretation. In opposition to those Meinongian logicians who admit nonexistent objects into their basic domains of interpretation, Pa niczek has just the ordinary notion of extension, because his basic domain contains only ordinary existing individuals and properties are defined on this domain exclusively. The nonexistent M-objects are constructed out of sets of individuals which constitute the ordinary extensions of properties. There are, strictly speaking, no nonexistent objects on full rights in Pa niczek's semantics. The M-objects that can be identified with existents turn out to be ordinary individuals, but there are no individuals in the domain to be identified with nonexistent M-objects. So Pa niczek's notion of extension is untypically traditional. In contrast to Pa niczek's austere proposal, where the basic domain includes only existent individuals, Jacquette's domain contains every conceivable entity. This is a true Meinongian domain, one might say. The model consists of three ordered components: a domain D, an interpretation I on the domain, and a truth valuation of propositions under the interpretation V, . (...) Domain D thus contains an object corresponding to every grammatically welldefined term, constant, predicate, functor, definite descriptor, and lambda abstract. Some of the objects are sets of objects, where every condition on any objects determines a set. (...) A nonstandard version of Zermelo- Fraenkel set theory drives the semantics.296

The membership in the extension sets of properties is non-standard in Jacquette's semantics, because there are many nonexistent objects which he considers to be undetermined as to their property possession. Therefore, properties are identical on his view only if the same objects belong to their 295

T. Parsons, Nonexistent Objects, p. 99. D. Jacquette, Meinongian Logic: The Semantics of Existence and Nonexistence, Walter de Gruyter & Co, Berlin-New York, 1996, p. 103. 296

MEINONG ON MEANING AND TRUTH

231

extensions and the same objects are also undetermined with respect to these properties. Since the property sets contain both existent and nonexistent entities, he calls them intensions instead of extensions of properties. This use of the notion of property intension does not coincide with the standard use of this notion, which is reserved for possible worlds options of the extensions of properties, where the property intensions can actually differ in the membership of objects from one possible world to another. No such thing happens when we have a one-world semantics. We rather have a broad extension of a predicate in this case, i.e. an extension of a predicate that contains both ordinary and nonexistent objects. But it is true that Jacquette's semantics is intensional, i.e. sense preserving, because the strict identity conditions for properties, concerning also the set membership of intentional and nonexistent objects, prevent properties from becoming identical just on the basis of the extensions that consist of existent objects. The predicate semantics for the object language of the logic can now be settheoretically defined.(...) This provides an intensional counterpart of the identity of a property with the existent and nonexistent objects in the intensions of corresponding predicates, parallel to the identity of a property with the existent objects in the extensions of corresponding predicates in standard extensional semantics. The intension of a predicate is non-Fregean, though indirectly related to Frege's property-'senses' by virtue of the identity conditions for objects, determined by their Soseine, or associated unordered sets of nuclear constitutive properties.297

Jacquette's intension of a predicate, or what I call its broad extension, captures the 'sense' of a property by means of including intentional objects, which play the role of linguistic meanings in attitude contexts. Thus a property is defined by all the meaning contexts in which it can be truly predicated. We may assume that if two properties are impossible to distinguish in all meaning contexts, then they are intensionally identical. The diagrams for property membership presented by Jacquette (p.109) show that for contradictory properties P and – P their broad extensions will have an overlapping common part that corresponds to the joint 297

D. Jacquette, Meinongian Logic: The Semantics of Existence and Nonexistence, p. 108.

232

ANNA SIERSZULSKA

membership of impossible objects in their extensions, and there are also some objects that do not belong to either of the two sets, because they are undetermined as to the possession of this property or its converse. ♦ Two kinds of properties and two kinds of predication Meinong's early subscription to the 'bundle' theory of objects, which holds that objects are bundles of properties, is reminiscent of the empiricist tradition, but Meinong's theory is not quite on the same lines. In the first place, Meinong's existent objects are bundles of particularized properties, which are moments of a concrete substance and not just clusters of abstract properties. Apart from this, the theory in its most interesting part concerns intentional objects, and its innovative character lies in Meinong's combinatorial approach, as it has been pointed out by Pa niczek: (...) whereas the classical bundle theories of properties basically pertained to existent (individual) objects, Meinongian theory has a combinatorial character.298

At this level, the theory does not deal solely with what can be perceived to exist, but with whatever is an intelligible object that can be apprehended on the basis of human intentional ability. Meinongian semantics reflect this feature of the theory of objects in their definitions of how objects are conceived. Parsons' axiom of objects is a good example: 'the axiom of objects' guarantees that, for any set of nuclear properties that is expressible in the language, there is an object that has exactly those nuclear properties.299

All possible sets of properties, compatible or contradictory, generate a wide spectrum of objects, some of which would never be capable of existence. But, as Meinongian logicians repeat, nonexistent objects are not entirely useless from the theoretical perspective. One might ask why it is important to make any distinctions among the properties of existent and nonexistent objects. The reason for this is that Meinong's principle of free assumption of an object's Sosein (the set of 298 299

J. Pa niczek, The Logic of Intentional Objects, p. 61. T. Parsons, Nonexistent Objects, p. 159.

MEINONG ON MEANING AND TRUTH

233

properties that constitute an object) results in some paradoxes without such distinctions. Some examples are: 'the existent golden mountain', 'the thought which is a thought about itself', 'the object which is a Meinongian object'. There are two distinctions designed to prevent the paradoxes: of two kinds of properties and of two kinds of predication, both originally introduced by Meinong's student, Ernst Mally. The distinction between two kinds of properties, described by Mally as 'formale' and 'ausserformale', was in fact accepted by Meinong himself as he defended his assumptions principle against Russell's attacks. Meinong calls them 'konstitutorische' and 'ausserkonstitutorische' – to give the distinction more explanatory value, and the distinction is translated by Findlay as a distinction between 'nuclear properties' – those belonging to the Sosein of an object, and 'extranuclear properties' – which are external to the Sosein of an object, such as being existent, impossible, incomplete. This distinction has been incorporated into the semantic theories of Terence Parsons, Richard Routley and Dale Jacquette. The second distinction, between predications of two kinds, originates from Mally's terms 'determinieren' and 'erfüllen', which mean, according to Findlay, being determined by a set of properties and satisfying the properties. The distinction between two kinds of predication has been utilized in a few Meinongian semantics under different names. There is the 'internal' and 'external' predication of Hector-Neri Castaneda and Jacek Pa niczek, the 'constituting' and 'exemplifying' of William Rapaport, and the 'encoding' and 'exemplifying' of Edward Zalta. The aim of such a distinction is to mark the difference between the ordinary exemplification of properties by real objects, and the possession of properties by semantic objects, since it is assumed that the latter cannot really exemplify any properties that are ascribed to them. There has been some discussion around these two distinctions concerning such questions as whether one of them is more appropriate than the other, and whether they are intertranslatable so that one can be explained in terms of the other. The distinctions seem to reflect two different approaches to predication, which brought to their extreme could be expressed by (a) the strangeness claim about objects, and (b) the strangeness claim about properties.

234

ANNA SIERSZULSKA

(a) the strangeness claim about objects: no properties can be properly predicated of some objects (which are semantic objects) => some objects are not ordinary objects of predication (b) the strangeness claim about properties: some properties (extranuclear) cannot be properly predicated of any objects => some properties are not ordinary properties According to (a), some objects can only encode or include properties, but they cannot exemplify properties, so they are not objects on equal rights with ordinary objects. According to (b), some properties do not constitute the nature of objects, so they are not ordinary (nuclear) properties. The situation is further complicated by the fact that the strangeness claims, both about objects and about properties, are not applied consistently. It is said that semantic objects do exemplify some properties, like being abstract objects, or being incomplete objects. Such 'extranuclear' properties, in turn, although normally non-constitutive, can have their nuclear 'watered-down' versions. So most Meinongian semantics make use more or less explicitly of both distinctions and they try to describe the problem in terms of both, if only by polemics. In Parsons' logic of nonexistent objects, apart from the nuclearextranuclear distinction, which plays an important role in the theory, the distinction between 'having' and 'including' properties is also mentioned. The obvious thing to do here is to make a distinction between a property's being in the set correlated with an object, and the object's having the property. In the former case, let us say that the object includes the property. Presumably, real objects have exactly the properties they include, but for others there will be a divergence. (...) When applied to language, this theory has the consequence that predication is ambiguous: Fa can mean either that a has Fness or that a includes Fness.300

While real objects have all the properties they include, fictional objects need not have any of the properties they include, or maybe they can have some and only include others. Parsons is quite precise about his preferences, but his 'orthodox' theory says that fictional objects have 300

T. Parsons, Nonexistent Objects, p. 171.

MEINONG ON MEANING AND TRUTH

235

extranuclear properties and only include nuclear ones. In the end, he decides that at least the treatment of all nuclear properties should be uniform for all objects. But for nuclear properties no such divergence seems appropriate. Perhaps fictional objects have all the nuclear properties that they include, as on the present theory, or perhaps they have none (...)301

Thus, both real and fictional objects of Parsons 'have' all their nuclear properties, and the distinction between two kinds of predication is dropped on this level. But is it dropped on all levels? It may be quite revealing to quote a passage from Cocchiarella. The possession of an extranuclear property is in that case correlated with the orthodox Russellian predication of a higher order property or relation. Such a difference in logical type and level in the representation of nuclear and extranuclear properties and relations, needless to say, might well be thought to be indicative of an underlying structural difference in Parsons's theory between two types, or forms of predication, viz., a Meinongian predication of nuclear properties, and a Russellian predication of extranuclear properties and relations.302

Parsons' is a second order logic, and, at least according to his 'orthodox' theory, his extranuclear predicates signify second order properties, e.g. 'being existent' is a property predicated of a set of properties that represents an individual. Extranuclear predication is higher order and always ordinary – the extranuclear properties are genuinely possessed by the set of properties in question. The nuclear predication, in agreement with what Parsons says, is sort of indifferent to the issue of the object's actually having the properties or not. The primary relation, also for existent objects, is that of the properties being included in the set representing an individual. Whether the object really has them, or does not have them, seems to be another matter. So we can see that, in a way, two kinds of predication are involved in Parsons' theory. Zalta is in favour of two kinds of predication, and against the nuclearextranuclear distinction. He emphasises very much the fact that 'abstract' objects do not exemplify properties in the same way real objects do. 301 302

T. Parsons, Nonexistent Objects, p. 172. Nino Cochiarella, "Meinong Reconstructed vs. Early Russell Reconstructed".

236

ANNA SIERSZULSKA

Things that exemplify the property of being a detective exist, have a location in space and time, are made of flesh and bones, think, solve crimes, and so on, whereas things that just encode the property of being a detective are abstract and do not exemplify any of these characteristics. They might exemplify these properties according to their respective stories, but this is not the same as exemplifying them simpliciter.303

Zalta's abstract A-objects encode most of their properties. However, Aobjects can also exemplify some properties, and incidentally, these are exactly the properties which are called extranuclear according to the distinction he rejects. For this reason, his theory can be supposed to contain the nuclear-extranuclear distinction implicitly. He could avoid this outcome, if he tried to be more consistent in applying the nonexemplification principle to 'abstract' objects, so that they would not be able to exemplify any properties at all. As it is, it seems (which is pointed out by Jacquette) that there is no clear principle when and which properties can be exemplified by 'abstract' objects. The limitation of Zalta's informal principle is that while it rules out exemplification by an abstract object when no existent object exemplifies the properties an abstract object encodes, it does not work in the opposite direction to determine when an abstract object exemplifies the properties it encodes.304

There is only the negative principle when the properties encoded cannot be exemplified, namely, when the A-object is not an auxiliary meaning-object that can be identified with an existent individual. The properties encoded by a nonexistent object cannot be exemplified. But which properties are merely encoded by a nonexistent object and not exemplified by this object? It seems that these are the extranuclear properties. Jacquette also makes the objection, that strictly speaking, if Zalta wanted to be consistent in his theory as independent from the other distinction, 'abstract' objects should not exemplify any properties, even the property of being abstract. It is not enough for Zalta to reform his use of what might now be called extranuclear predicates 'E!' and 'A!'. Even if he were to agree that abstract objects can only encode properties, and no abstract object exemplifies or really has a 303 304

E. Zalta, Intensional Logic and the Metaphysics of Intentionality, p. 17. D.Jacquette, Meinongian Logic, p. 37.

MEINONG ON MEANING AND TRUTH

237

property, and write 'xA!' or '~xE!' instead of 'A!x' or '~E!x', the philosophical problem remains. (...) To make a special case for properties like 'abstract' and 'nonexistent' is tacitly to rely on a distinction between constitutive and nonconstitutive or nuclear and extranuclear properties, which categories of properties Zalta indeed refers to respectively as 'nontheoretical' and 'theoretical'.305

Zalta could change his notation and represent 'being abstract' as 'xA!' – x encodes A, instead of 'A!x' – x exemplifies A, but we have to agree with Jacquette that the philosophical problem remains. Does an abstract object exemplify the property of being abstract or does it encode this property? I would be inclined to think that abstract objects exemplify no properties at all. Yet they really are abstract objects, aren't they? Making use of Zalta's expression, 'being abstract' is a 'theoretical' property and so it may be the most proper to understand it as a higher order property of sets of properties representing objects. Perhaps we could simply say that it is not predicated at the level of ordinary properties on which the 'encoding – exemplifying' distinction operates. Some logicians, like Kit Fine306, Jaquette, or Pa niczek, think that the two distinctions are either intertranslatable, or that it is possible to get rid of one of them in favour of the other. Jacquette thinks that the nuclearextranuclear distinction is more basic and the dual predication can be reduced to this distinction, although his attempt to do so is not quite successful due to some misconceptions concerning Zalta's theory, which he is trying to reduce. The reduction of the dual copula or dual modes of predication to to the nuclearextranuclear property distinction is easy to accomplish, since the two predication modes arise entirely in connection with whether or not an object has the extranuclear property of existence.307

There are also voices in defence of the distinction between two kinds of predication, which appears to be more clear and more elegant. Pa niczek, for example, proposes internal and external predication, with constitutive properties predicated internally.308 The innovation is that it differs from 305

D.Jacquette, Meinongian Logic, p. 26-27. Kit Fine, "Critical Review of Parsons' Nonexistent Objects". 307 D. Jacquette, Meinongian Logic, p. 17-18. 308 J. Pasniczek, The Logic of Intentional Objects, p. 122. 306

238

ANNA SIERSZULSKA

object to object which properties are those predicated internally or externally, so there is no preset list of properties of each kind, as in the case of nuclear and extranuclear properties. This solution combines the two distinctions, but it differs from Zalta's encoding-exemplifying division. There is no encoding for existent objects, and no exemplification for ordinary properties of abstract objects on Zalta's theory, while Pa niczek's internal and external predications apply to all objects indiscriminately. In opposition to the proponents of intertranslatability, Zalta thinks that the translation between the two distinctions is impossible, at least on the ground of his theory, because the same property can be both exemplified and encoded depending on whether the object is ordinary or 'abstract'.309 There is indeed a certain lack of correspondence involved here, as the two distinctions originate from the two different approaches to predication, which I describe above as the strangeness claim about objects and the strangeness claim about properties. The mutual interrelations between 'nuclear-extranuclear' and 'encodedexemplified' distinctions can be understood as follows: If P is an extranuclear property then P is exemplified; extranuclear => exemplified, but not conversely; If P is exemplified then P is an extranuclear property, or a nuclear property of a real object. If Q is encoded then Q is a nuclear property; encoded => nuclear , but not conversely; If Q is a nuclear property then Q is encoded by an abstract object or exemplified by a real object. Judging from the above one may have some doubts whether these two distinctions are really counterparts of each other, or even if they are indeed intertranslatable in any straightforward way. Perhaps it is useful to make both distinctions, for they concern slightly different issues. Another option might be to assume that all properties possessed by objects, both by way of encoding and exemplification, are ordinary properties, coinciding with what is called 'nuclear' properties. The 309

E. Zalta, "On Mally's Alledged Heresy: A Reply", online.

MEINONG ON MEANING AND TRUTH

239

'extranuclear' properties would not have to be considered as properties of objects at all. They could be substituted by logical operators of some kind whenever there is a need to express the idea that the objects we speak about appear in a certain context as existent, fictional, incomplete, etc. Meinongian logics already use a special existence symbol 'E!' , which is also called the 'existence predicate', but its semantic role is somewhere between a predicate and a strong existential quantifier. Perhaps it would be possible to introduce instead some special operators to express what is conveyed by means of 'extranuclear' predicates. Alternatively, one could assume that such properties are higher order properties, or that they are predicable only in a metalanguage, since they are usually the properties not of objects but of sets of properties that constitute objects. Thus, being contradictory, nonexistent and incomplete are not properties of the object the round square, but of the set of properties {being round, being a square}. Someone might object that being a fictional, abstract, or mythological object, on the other hand, does not apply to sets of properties, but to the objects, like Pegasus. Yet we can still argue that such properties belong to the meta-discourse. In consequence, it might be possible not to employ the slightly awkward distinction between nuclear and extranuclear properties. As to the other distinction, between two kinds of predication, there is no absolute need to employ it either, because the paradoxes may be taken care of by means of introducing logical operators for special contexts or higher order predicates for 'external' properties. All the same, I think that distinguishing between exemplification of properties by real objects and mere encoding, or including, of properties by abstract objects, has some explanatory value and may contribute to avoiding many misunderstandings on the part of more Russellian-minded philosophers. ♦ Two senses of 'there is' The principle of independence of so-being from being Admitting 'merely semantic' objects into a Meinongian domain and claiming that predications about them may be true, requires a formal justification, which is provided by the principle of independence. Like

240

ANNA SIERSZULSKA

most of the logic related to Meinong's philosophy, the principle has its origin in the work of Ernst Mally. In a book devoted to this principle, Karel Lambert presents its two formulations, a strict and a looser version. The strict version makes use of the notions of Sosein and of nuclear (constitutive) properties, in order to express the idea that an object may have a Sosein consisting of a certain set of nuclear properties, even though the object does not possess being. The strict version says that the principle: "There are nuclear properties P1, P2... such that the set of P1,P2... attaches to s ; So, s has being" - is invalid.310

The loose version, in turn, says that the principle: "There is a property P such that P is possessed by s ; So, s has being" - is invalid.

Lambert explains the difference between these two versions as follows: This looser version of the principle of independence differs from the strict version in two key respects. First, no mention of a set of properties (a Sosein) attaching to an object occurs in instances of the looser version. Second, the properties satisfying particular instances of the looser version can be extranuclear in contrast to the strict version. (...)311 Since the looser version is simpler and also affords continuity between the authorities and the current work, it will henceforth be adopted. This means leaving behind the nuclear-extranuclear property distinction vital to Meinong's theory.312

It seems that the advantage of the looser version is that predication may be true about an object even when the only property ascribed to it is a single, extranuclear property. So the object need not be constituted properly to be regarded as an object by Lambert, because his main concern is preserving the truth of predications for otherwise irreferential expressions, which is characteristic for his variety of free logic. 310

Karel Lambert, Meinong and the Principle of Independence, p. 28. Karel Lambert, Meinong and the Principle of Independence, p. 29. 312 Karel Lambert, Meinong and the Principle of Independence, p. 30. 311

MEINONG ON MEANING AND TRUTH

241

Lambert classifies free logics as positive, which are those according to which some predications concerning nonexistent objects are true, negative free logics – with only false predications of nonexistents, and neuter, on which such predications are truthvalueless.313 His own view is that some predications about nonexitent objects can be true, although such terms are irreferential. In the Russellian tradition, the 'is' of predication implies the 'is' of being, but the 'is' of predication and the 'is' of identity are treated differently. Identity statements about nonexistents may be true, even though predications must be false. The Meinongian tradition, represented also by Lambert, treats the 'is' of identity and of predication on equal rights, since both "The round square is the round square" and "The round square is round" are true sentences. None of these true sentences implies the 'is' of being, though. However, terms such as 'the round square' do not stand for beingless objects, according to Lambert, since in free logic there is no need for beingless objects, it is sufficient to admit irreferential terms.314 He remarks that, even though there are some free logicians who hold that such terms refer to nonexistent objects, none of them quantifies over such objects.315 Quantifying over nonexistent objects The most typical feature of Meinongian logics is supposed to be quantifying over nonexistent objects ('merely semantic' objects). And, on the standard view, this marks their ontological commitment to such entities. But those of the Meinongian-style logicians who actually quantify over nonexistent objects, hardly ever agree that, by this fact, they are committed to the existence of what in principle does not exist. They say that 'there are' nonexistent objects, although none of them have either existence or any kind of being. The mystery of this paradoxical position lies in the distinction between two senses of 'there is'. One of these senses is expressed by the ordinary existential quantifier ranging over all objects in the Meinongian domain. The typical reading that it receives has nothing to do with an object's existential status: 313

Karel Lambert, Meinong and the Principle of Independence, p. 116. Karel Lambert, Meinong and the Principle of Independence, p. 76. 315 Karel Lambert, Meinong and the Principle of Independence, p. 97. 314

242

ANNA SIERSZULSKA

∃ x – x belongs to the domain, or there is an object x in the domain The other sense of 'there is', the one which is standardly supposed to be expressed by existential quantifier, and which affirms an object's actual existence or, more generally, its being, is conveyed in Meinongian logics by a special existence predicate: E!x – x has being, or an object x in the domain has being The question appears what is 'E !'. It is not a second existential quantifier, though it seems to play a similar role. It is called a predicate, which seems to be fine when it is a second order predicate. But it is also employed in first order logics, where it must certainly be regarded as a very special predicate, because existence is definitely not an ordinary property of objects. In any case, it is by means of 'E !' that the ontological commitment usually associated with the existential quantifier is expressed. For instance, Jacquette writes: The 'existential' quantifier with bound variables ranging over existent and nonexistent objects in the Meinongian domain cannot serve as a criterion of ontological commitment. Instead the test must be to examine canonical formulations of a theory for occurences of extranuclear existence predications to determine which objects the theory claims to exist.316

Thus, Quine's criterion as to which objects a theory is committed, is abandoned. No singular terms can be eliminated by means of Russellian descriptions and so it is not enough to be the value of a bound variable to have being. Being must be explicitly stated by means of the existential predicate in order to determine to which objects it pertains. This is a requirement not only with respect to real existent objects, but with respect to any object that is claimed to have being of any kind. (...) the extranuclear existence property 'E !' is intended to subsume both of Meinong's existence and subsistence categories.317 316 317

D. Jacquette, Meinongian Logic, p. 95. D. Jacquette, Meinongian Logic, p. 116.

MEINONG ON MEANING AND TRUTH

243

Once such a proclamation of being is made, Jacquette is glad to admit that the ontological commitment of a theory has been stated explicitly and beyond any doubts. Which is good, because it enables us to see clearly which theories are committed to which entities. A theory about unicorns on this analysis turns out to be ontologically committed to the existence of unicorns, even though no unicorns exist.318

It does not matter entirely whether entities of a certain kind really exist in the world or not. If the theory claims their existence, it is committed to them, as the theory of 'ether' was committed to ether, for example, even though it was false. Jacquette thinks that this approach is useful in application to certain scientific theories that are committed to nonexistent theoretical entities like the ideal gas or the frictionless surface or infinitesimals. Also in Zalta's theory, apart from the usual existential quantifier, there is the existence predicate. Clearly, there are two notions of existence that may be distinguished in our logic. One is expressed by the predicate 'E!.' The other is expressed by the quantifier '∃', even though we have used the words 'some' and 'there is' to read this symbol.319

Consequently, because the quantifiers do not carry any commitment obligation, Zalta may quantify over the whole domain including 'abstract' objects. The nonexistence of an object doesn't diminish its status as an object over which we may quantify.320

A generous policy concerning quantification is assumed by Parsons as well, whose existence predicate is used by him to formulate explicitly an axiom according to which there are such objects that no such objects exist, paraphrasing Meinong's well known saying "there are objects that there are no such objects".

318

D. Jacquette, Meinongian Logic, p. 69. E. Zalta, Intensional Logic..., p. 102. 320 E. Zalta, Intensional Logic..., p. 68. 319

244

ANNA SIERSZULSKA

(...) it is explicitly assumed that variables may range over all objects, not just those that exist. There is an extranuclear predicate of existence, written 'E !', and there are axioms which entail that there are objects that don't exist: (∃x) ~E !x321

This is in agreement with what we are told at the beginning of his book, that the theory presented will claim that there are, in fact, nonexistent objects: The theory given below will say that there are unicorns, there is such a thing as Pegasus, etc., but that none of these exist.322

The only problem is with the meaning of the phrase 'there are' in the existence-neutral sense, which I characterize above as 'belonging to the domain'. Doesn't it really imply any kind of being at all? Parsons says that he does not know and has no intention of taking sides on this issue. Russell objected that if there are objects that do not exist, they have to have some other kind of being. I have never been able to find more than a terminological issue here. If there is an issue about whether nonexistent objects have some kind of being, I intend to remain neutral on the issue.323

Some people think that there is an issue, though, and that either there are no nonexistent objects or else they must have some other kind of being. For this reason, they find it safer not to quantify over such objects, or to quantify over them very carefully, so as not to suggest more than they in fact want to assert. The substitutional reading of quantification is an example of such a strategy. (...) some proponents of substitutional quantification assume that there are names that fail to refer but which nonetheless appear in true sentences of the form 'N is an A', and then they hold that some sentences of the form 'There is an A' are true even though no object is an A.324

In other words, if one accepts certain instances of predication as true, one also has to agree that certain instances of quantification are acceptable, without accepting that there are, in any sense, the objects quantified over. 321

T. Parsons, Nonexistent Objects, p. 155. T. Parsons, Nonexistent Objects, p. 11. 323 T. Parsons, Nonexistent Objects, p. 10. 324 T. Parsons, Nonexistent Objects, p. 12. 322

MEINONG ON MEANING AND TRUTH

245

On a substitutional reading of the existential quantifier, we accept that at least one instance of the substitution of a name from a certain inventory of names must be true. But we do not assume anything about the existence of the objects signified by the names we substitute. In relation to substitutional quantification two issues are important – as pointed out by Susan Haack325. First, if it is assumed, and it is assumed on a Meinongian theory, that singular terms are not eliminable by means of Russell's theory of descriptions, then quantification over variables need not be the only way to express ontological commitment to objects, because singular terms can perform this task by either denoting existent entities or not. It follows from this observation that we can agree to have non-committal quantification over all entities in a Meinongian domain and there is no real need for substitutional quantification. Second, substitutional quantification only postpones ontological questions, shifting them from the quantifiers to the names. Now we can make a choice which names we admit to be substituted for the variables. A Meinongian semantic theory will obviously admit names of 'merely semantic' entities. The truth of the instances of the subsitution of such names for the variables quantified over depends in the first place upon whether we accept that predications concerning the objects signified by these names can be true. From this second issue raised by Haack it seems to follow that the objectual treatment of the instances of quantificational substitution will eventually enter through the back door in any case... Fortunatelly, this is not a problem for a Meinongian semantics due to the consequences of its rejection of singular terms elimination. Parsons' quantifiers are objectual, but Pa niczek claims, for example, that his quantification over M-objects is substitutional. There is just a spectrum of constants which stand for the constructed Meinongian objects of the secondary domain. The constants are substituted for the variables under the range of the quantifiers without assuming any commitment to the objects signified by these constants. One can easily notice that the 'quantification' over M-objects in M-logic (...) may be considered as a substitutional one.326

325 326

Susan Haack, Philosophy of Logics, pp. 46-50. J. Pa niczek, The Logic of Intentionality, p. 80.

246

ANNA SIERSZULSKA

We must remember that Pa niczek avoids any commitment to M-objects by constructing them out of sets of ordinary individuals in his domain. Therefore, the constants representing M-objects merely stand for sets of properties, and not for objects in any ordinary sense. M-objects are represented by constants and since these are of a different category than individual variables, according to Quine's criterion they do not involve any ontological commitments. The role of constants in M-language is similar to that of symbols representing sets in the simple theory of sets which merely 'go proxy' for sets.327

Such an approach is not too non-Meinongian, for Meinong's 'merely semantic' objects do not seem to amount to much more than sets of properties. Still, there is certainly a difference between a mere collection of properties and an object possessing these properties. In Pa niczek's semantics, existent M-objects are identified with ordinary individuals, so they are not mere collections of properties in the end. Categorematic reading of terms which follows their semantic interpretation, does not mean that we equate these interpretations, i.e. sets of sets of individuals, with objects themselves. Neither do we equate properties with extensions of predicates, i.e. sets of individuals. We only represent objects and properties by set-theoretic constructs.328

Existent objects are only represented by means of set-theoretic constructs, but they are clearly not intended as such constructs by the theory. Nonexistent objects, however, remain set-theoretic constructs on Pa niczek's account, which stands in opposition to the way other Meinongian-style philosophers have usually understood these objects. A passage by Parsons is quite representative in this respect: I don't think that concrete objects are sets of properties. For example, I don't think that you are a set of properties. And I think the same about Sherlock Holmes, the gold mountain and the round square. Occasionally, people have said to me (in an

327 328

J. Pa niczek, The Logic of Intentionality, p. 79. J. Pa niczek, The Logic of Intentionality, p. 59.

MEINONG ON MEANING AND TRUTH

247

effort at friendly compromise), "When you say 'object', you really mean 'set of properties', don't you?" No, I don't.329

So are the Meinongian logics which quantify over nonexistent objects committed to the being of these objects? Not necessarily, but they are usually committed to the idea that these are really objects of some kind that deserve a serious theoretical treatment.

329

T. Parsons, Nonexistent Objects, p. 93.

MEINONG ON MEANING AND TRUTH

249

CONCLUSION: THE COSTS OF PRESERVING EPISTEMOLOGICAL REALISM A philosophical theory of truth that aspires to provide a useful theoretical insight into the nature of cognition from the perspective of scientific purposes is primarily expected to reassert rather than undermine the objective status of scientific investigations. Otherwise all scientific enterprises would appear to be futile. This goal can be best accomplished from the position of epistemological realism. The question that arises is whether and under which conditions it is possible to formulate a theory of truth that preserves the intuitions behind epistemological realism without oversimplifying the complexity of the epistemic situation and without disregarding the sceptical doubts evoked by the subjective aspects of cognition. If we try to specify the conditions that should be fulfilled in order to call a theory of truth realistic, we will be likely to arrive at the following six points: 1. A non-subjective treatment of the object of cognition 2. A mind-independent status of the truth bearer 3. A representational character of the relation between the truth bearer and reality 4. Avoiding the problem of correspondence as a relation between truth bearers and truth makers 5. Maintaining the substantial character of truth as opposed to deflationary conceptions 6. Maintaining the absolute character of truth as opposed to relativistic positions Meinong's conception of truth as it is reconstructed in this study seems to satisfy all the above conditions and therefore it may be regarded as realistic. Nevertheless, Meinong's attempts to preserve epistemological realism are not without minor and major drawbacks, which is the price to

250

ANNA SIERSZULSKA

be paid for defending a consistent and cognitively positive conception. The means and the costs of preserving epistemological realism within Meinong's theory of truth will be reassessed briefly with respect to the above six conditions. 1. A non-subjective treatment of the object of cognition Meinong's distinction between the subjective content of a mental act and its non-subjective immediate object presented by the content is a device to provide a uniform non-subjective treatment of all objects that cognition may deal with. The resulting notion of an object is very broad, because it includes all 'objective sense-entities'. This approach does not discriminate theoretically any possible area of discourse. The typical objection is that the domain of all objects of discourse together with all objects of intersubjective intentional reference is by far too broad, beyond the bounds of even a very generous ontology. The reply is that a semantic domain may be allowed to cross the limits of ontology so as to contain a sufficiently rich inventory of possible objects of intentional and linguistic reference, which may be useful for theoretical purposes. The category of 'semantic objects' is introduced to group under a general category all 'objects' in Meinong's theory that are not real existing objects but they still possess a non-subjective character. 2. A mind-independent status of the truth bearer As entities belonging to Aussersein, Meinong's objectives are mindindependent truth bearers that can be intended by different subjects. They also play the role of the meanings of declarative sentences and as such they are considered in this study to be 'objective sense entities' that possess a function-like character. This interpretation of objectives is prompted by a specific feature that distinguishes them from other objects of higher order: the subordinate objects of an objective are always external in relation to the objective itself. The notion of being saturated or satisfied by external arguments helps to provide an account of Meinong's factuality. 3. A representational character of the relation between the truth bearer and reality Meinong's version of logical realism assumes that the structure of an objective as intended reflects the structure that obtains between objects in

MEINONG ON MEANING AND TRUTH

251

reality (a factual objective), with the reservation that reality is not divided into such entities as factual objectives. The objective as intended and the factual objective are not exactly identical in the sense that the objective intended involves representations of its subordinate objects while the factual one obtains between real objects. They can be identified on the strength of the ontic identity of the objects that are represented with the ultimate objects, and because they consist in the same coordination between these objects (identity of logical structure). It may be objected that according to logical realism of this kind, the structure of human thought and language is projected upon reality, as pretending to the status of its inherent structure. On the reverse view, such as can probably be attributed to Meinong, it may be possible to perceive directly which coordinations between objects obtain or do not obtain in the world, and so our language can reflect correctly the ways things are in reality. 4. Avoiding the problem of correspondence as a relation between truth bearers and truth makers The criticisms of the idea of correspondence are related to the impossibility to represent perfectly in the mind any existing thing. If truth were to be understood as correspondence, it would admit of degrees and this must lead to relativism. The only theoretical option suitable for accomodating the absolute notion of truth is the identity theory. Since an identity theory on which facts understood as states of affairs are both truth makers and truth bearers is impossible, the remaining alternative is a theory on which truth bearers are 'facts'. Meinong's identity theory of truth seems to be of the latter kind. His notion of a fact is abstract or 'non-objectual', it is the obtaining of a true objective. True objectives are factual due to reflecting the structure of reality. Meinong's identity theory of truth is representational, even though it is not a relational theory with respect to objectual facts. A relational theory may appear to be more natural at first, because a representational account of the meaning of linguistic expressions is always relational – senses represent objects – and, by analogy, the senses of sentences should also represent objects: states of affairs. However, while there are objects in the world to be represented, there are no states of affairs to be found in the world. Meinong proposes a representational theory of the meanings of expressions combined with a semi-

252

ANNA SIERSZULSKA

representational account of the meanings of sentences: except for the representations of objects, objectives intended include a nonrepresentational part that expresses the logical structure or coordination of objects which is supposed to obtain in reality. If the objects are not coordinated in reality in the way specified, the objective is false. Truth without facts consists in correctness without correspondence. 5. Maintaining the substantial character of truth as opposed to deflationary conceptions Meinong presents a realistic identity theory of truth which preserves the substantial character of this notion. This is largely the result of proposing a representational theory of meaning, which links Meinong's truth bearers to reality. A conception of truth is taken to be realistic when a truth bearer is true only if it obtains objectively, or if its realistic satisfaction conditions are fulfilled. In contrast, on an antirealistic conception of truth to be true means to be assertible according to certain assertibility conditions (truth is relative to these conditions). On a deflationary minimalistic conception, truth is related merely to the usage conditions of the predicate 'true' (truth is relative to language use). The minimalistic approach is a consequence of skepticism concerning the possibility to define the nature of truth. But apart from deflationary minimalistic conceptions, there are also realistic minimal theories of truth, including some Fregean-type identity theories and Meinong's theory among them. Such conceptions agree with the deflationary approach that truth is indefinable, although they do not agree that there is nothing substantial involved in this notion. Truth is an indefinable but substantial property of truth bearers. 6. Maintaining the absolute character of truth as opposed to relativistic positions The notion of truth in Meinong's conception is absolute: the truth or falsity of all objectives is definite, non-relative and eternal. As a side effect of this strong notion of truth, however, we encounter a conception that possesses a certain deterministic flavour. On Meinong's absolute view, there is a preset inventory of existing objects and of factual objectives which did, do and will obtain in the future. A possible line of defence might be that future existents and factuals do not obtain also now and throughout the whole

MEINONG ON MEANING AND TRUTH

253

eternity. It is simply the case that the world will necessarily have only one course of events in the future, so some claims about this course of events will turn out to be true and others will not. This need not imply, as we can hope, that there is just one fatalistically determined way for things to be. The perspective changes once we pass to the questions of cognition. There is a back door in Meinong's absolute conception of truth, which can be called 'Meinong's epistemic view'. The epistemic view is related to empirical cognition and the epistemic undecidability of empirical judgments. On his solution, the uncertain objectives are regarded as possessing an unknown truth value, and they are ascribed degrees of probability to be true. The different options of interpretation on the epistemic view can be represented semantically as possible worlds. Meinong's theory of truth and his views upon related matters certainly reveal a lot of affinity to issues discussed currently within the area of philosophical semantics and truth theory. One of the major current positions in the theory of truth, the minimal approach towards the question of defining the nature of truth, finds a realistically-minded representative in Meinong. Like the minimalists, he drops the idea of correspondence and of facts as entities in the world, but unlike many minimalists he is firmly convinced about the absolute and objective character of truth. As an answer to sceptical and relativistic objections concerning the subjective elements in cognition, Meinong offers a suggestion to apply a probabilistic semantic approach in uncertain contexts, and this is the approach which is actually widely applied in the field of empirical sciences at present. Within a philosophical theory of truth there are hardly any final answers to be given, but Meinong's theory indicates a positively realistic direction of thinking.

MEINONG ON MEANING AND TRUTH

255

BIBLIOGRAPHY Albertazzi, Lilianna, "Presentation and Production", The School of Alexius Meinong, eds. L.Albertazzi, D.Jacquette, R.Poli, Western Philosophy Series, Ashgate, Aldershot, 2001. Armstrong, David M., A World of States of Affairs, Cambridge University Press, 1997. Baldwin, Thomas, "The Identity Theory of Truth", Mind 100, 1991. Barwise, Jon & Perry, John, Situations and Attitudes, Massachussets Institute of Technology Press, 1984. Bergmann, Gustav, Realism: A Critique of Brentano and Meinong, The University of Wisconsin Press, Madison and London, 1967. Bolzano, Bernard, Theory of Science, transl. Rolf George, University of California Press, 1972. Brentano, Franz, The True and the Evident, transl. R.Chisholm, I.Politzer, K.Fischer, Routledge, London, 1966. Chisholm, Roderick M., "Homeless Objects" (Revue Internationale de Philosophie 22, 1973, and in: Brentano and Meinong Studies, Rodopi, Amsterdam, 1982. Cochiarella, Nino, "Meinong Reconstructed vs. Early Russell Reconstructed", Journal of Philosophical Logic 11, and in his: Logical Studies in Early Analytic Philosophy, Ohio State University Press, 1987. Davidson, Donald, Essays on Actions and Events, Clarendon Press, Oxford, 1980. Davidson, Donald, Inquiries into Truth and Interpretation, Clarendon Press, Oxford, 1983. Dodd, Julian, An Identity Theory of Truth, Macmillan Press Ltd., Houndmills, 2000. Dölling, Evelyn, "On Alexius Meinong's Theory of Signs", in R.Poli (ed.), The Brentano Puzzle, Ashgate, Aldershot, 1998. Dummett, Michael, Frege: Philosophy of Language, Harvard University Press, 1981.

256

ANNA SIERSZULSKA

Dummett, Michael, The Logical Basis of Metaphysics, Harvard University Press, 1993. Edgington, Dorothy, “Validity, Uncertainty and Vagueness", Analysis, 1992. Engel, Pascal, "The False Modesty of the Identity Theory of Truth", International Journal of Philosophical Studies, vol. 9 (4), 2001. Field, Hartry, "Deflationist Views of Meaning and Content", Mind, vol. 103, 1994. Field, Hartry, "The Deflationary Conception of Truth", in G.McDonald, C.Wright (eds), Fact, Science and Morality: Essays on A.J.Ayer's Language, Truth and Logic, Blackwell, Oxford, 1986. Findlay, John N., Meinong's Theory of Objects and Values, Oxford UP, 1963. Fine, Kit, "Critical Review of Parsons' Nonexistent Objects", Philosophical Studies 45. Føllesdal, Dagfinn, “Husserl's Notion of Noema", The Journal of Philosophy 66/20, 1969. Frege, Gottlob, “What is a Function?", in: Geach, Peter & Black, Max, Translations from the Philosophical Writings of Gotlob Frege, Blackwell, Oxford, 1960. Frege, Gottlob, "Function and Concept", in: Geach, Peter & Black, Max, Translations from the Philosophical Writings of Gotlob Frege, Blackwell, Oxford, 1960. Gardies, Jean-Louis, Rational Grammar, transl. by Kevin Mulligan, München-Wien, Philosophia Verlag, 1985. Grossmann, Reinhardt, Meinong, Routledge and Kegan Paul, London, 1974. Haack, Susan, Philosophy of Logics, Cambridge University Press, 1978. Hintikka, Jaakko, "Meinong in A Long Perspective", Grazer Philosophische Studien, vol. 50, 1995. Hochberg, Herbert, "Abstracts, Functions, Existence and Relations in the Russell-Meinong Dispute, the Bradley Paradox and the RealismNominalism Controversy", Grazer Philosophische Studien vol. 50, 1995. Hornsby, Jennifer, “Truth – The Identity Theory", Proceedings of the Aristotelian Society, vol. 97, 1996.

MEINONG ON MEANING AND TRUTH

257

Horwich, Paul, Meaning, Clarendon Press, Oxford, 1998. Husserl, Edmund, Logical Investigations, Routledge and Kegan Paul, London, 1976. Husserl, Edmund, Ideas. General Introduction to Pure Phenomenology, George Allen and Unwin, London, 1931. Jacquette, Dale, "Aussersein of the Pure Object", The School of Alexius Meinong, eds. L.Albertazzi, D.Jacquette, R.Poli, Western Philosophy Series, Ashgate, Aldershot, 2001 Jacquette, Dale, Meinongian Logic: The Semantics of Existence and Nonexistence, Walter de Gruyter & Co, Berlin-New York, 1996. Kripke, Saul, Naming and Necessity, Blackwell, Oxford, 1993. Künne, Wolfgang, Conceptions of Truth, Clarendon Press, Oxford, 2003. Lambert, Karel, Meinong and the Principle of Independence, Cambridge University Press, 1983. Lewis, David, On the Plurality of Worlds, Blackwell, Oxford, 1986. Lindenfeld, David, The Transformation of Positivism, University of California Press, 1980. Lowe, E. Jonathan, The Possibility of Metaphysics, Clarendon Press, Oxford, 1998. McDowell, John, Mind and World, Harvard University Press, Cambridge, Mass., 1994. Meinong, Alexius, Über die Erfahrungsgrundlagen unseres Wissens, Springer, Berlin, 1906, and in: Gesamtausgabe, vol. V, Graz, 1973. Meinong, Alexius, Über Annahmen, Barth, Leipzig, 1910. Meinong, Alexius, On Assumptions, tr. James Heanue, University of California Press, 1983. Meinong, Alexius, “The Theory of Objects", transl. Issac Levi, D.B.Terrell and R.M.Chisholm, in: R.M.Chisholm, ed., Realism and the Background of Phenomenology, Ridgeview Publishing Company, 1960. Meinong, Alexius, Über die Stellung der Gegenstandstheorie im System der Wissenschaften, Gesamtausgabe, vol. V, Graz, 1973. Meinong, Alexius, On Objects of Higher Order and Their Relationship to Internal Perception, transl. M-L.Schubert Kalsi, Nijhoff, The Hague, 1978.

258

ANNA SIERSZULSKA

Meinong, Alexius, Über Möglichkeit und Warscheinlichkeit, Barth, Leipzig, 1915. Meinong, Alexius, On Emotional Presentation, transl. M-L. Schubert Kalsi, Northwestern University Press, Evanston, 1972. Morscher, E.& Simons, P., "Meinong's Theory of Meaning", The School of Alexius Meinong, eds. L.Albertazzi et al, Ashgate, 2001. Mulligan, K., Simons, P., Smith, B., "Truth Makers", Philosophy and Philosophical Review 44, 1984. Mulligan, Kevin, "Judgings: Their Parts and Counterparts", in: Topoi Supplement, 2: La Scuola di Brentano, 1988. Neale, Stephen, Facing Facts, Oxford University Press, 2001. Parsons, Terence, "Meinongian Semantics Generalized", Grazer Philosophische Studien vol. 50, 1995. Parsons, Terence, Nonexistent Objects, Yale University Press, 1980. Pa niczek, Jacek, The Logic of Intentional Objects. A Meinongian Version of Classical Logic, Kluwer, Dordrecht, 1998. Poli, Roberto, "Truth Theories", The School of Franz Brentano, eds. L.Albertazzi, M.Libardi, R.Poli, Kluwer, 1996. Quine, Willard V.O., The Ways of Paradox and Other Essays, Random House, New York, 1966. Quine, Willard V. O., Ontological Relativity and Other Essays, Columbia University Press, New York, 1969. Reicher, Maria, "Gibt es unvollständige Gegenstände?", Grazer Philosophische Studien vol. 50, 1995. Routley, Richard, Exploring Meinong's Jungle and Beyond, Australian National University, Canberra, 1980. Russell, Bertrand, "Review of Meinong and Others, Untersuchungen zur Gegenstandstheorie und Psychologie,“ Mind 14 (Oct. 1905): 530-8. The Collected Papers of Bertrand Russell, vol. 4, Routledge, London, 1994. Russell, Bertrand, "Meinong's Theory of Complexes and Assumptions", Mind 13, 1904 Russell, Bertrand, "The philosophy of logical atomism" (1918), in: Logic and Knowledge, G. Allen and Unwin, London, 1956. Schubert Kalsi, Marie-Luise, Meinong's Theory of Knowledge, Nijhoff, Dordrecht, 1987.

MEINONG ON MEANING AND TRUTH

259

Simons, Peter, "Vagueness, Many-Valued logic and Probability", in: W.Lenzen (ed.), Das weite Spectrum der analytischen Philosophie, Gruyter, Berlin, 1997. Simons, Peter, "On What There Isn't: the Meinong – Russell Dispute", Philosophy and Logic in Central Europe form Bolzano to Tarski, Kluwer, Dordrecht, 1992. Simons, Peter, "Meinong's Theory of Sense and Reference", Grazer Philosophische Studien vol. 50, 1995. Smith, Barry, Austrian Philosophy, Open Court, Chicago, 1994. Smith, D.W. & McIntyre, R., Husserl and Intentionality: A Study of Mind, Meaning and Language, D.Reidel Publishing Company, Dordrecht, 1984. Tennant, Neil, The Taming of the True, Clarendon Press, Oxford, 1997. Tarski, Alfred, “The Semantic Conception of Truth and the Foundations of Semantics", The Collected Works of Alfred Tarski, vol II, Birkhauser, Basel-Boston-Stuttgart, 1986. Tegtmeier, Erwin, "Meinong's Complexes", The Monist, vol. 83, no. 1, January 2000. Willard, Dallas, “The Paradox of Logical Psychologism: Husserl's Way Out", in: F.A.Elliston and P.McCormick (eds.), Husserl: Expositions and Appraisals, University of Notre Dame Press, 1977. Wittgenstein, Ludwig, Tractatus Logico-Philosophicus, Ruotledge and Kegan Paul, London, 1961. Wittgenstein, Ludwig, Philosophical Investigations, Blackwell, Oxford, 1953. Wole ski, Jan, "Theories of Truth in Austrian Philosophy", Essays in the History of Logic and Logical Philosophy, Jagiellonian University Press, 1999. Wole ski, J. & Simons, P., "De Veritate: Austro-Polish contributions to the theory of truth from Brentano to Tarski", in: K.Szaniawski, The Vienna Circle and the Lvov-Warsaw School, Kluwer, 1989. Wole ski, Jan, Metamatematyka a epistemologia, Wydawnictwo Naukowe PWN, Warszawa, 1993. Wole ski, Jan, "Brentano's Criticism of Correspondence Conception of Truth and Tarski's Semantic Theory", Topoi 8, 1989.

260

ANNA SIERSZULSKA

Zalta, Edward, Intensional Logic and the Metaphysics of Intentionality, Massachussets Institute of Technology Press, 1988. Zalta, Edward, "On Mally's Alledged Paradox: A Reply", History and Philosophy of Logic, 13, 1992. egle , Urszula, "Meinong and Ingarden on Negative Judgments", Axiomathes, No 1-2, September 1996.

MEINONG ON MEANING AND TRUTH

261

INDEX OF NAMES A

Albertazzi, Lilianna 180-181, 255, 257258 Armstrong, David 255 B

Baldwin, Thomas 255 Barwise, Jon 255 Bergmann, Gustav 78, 86-91, 207, 255 Black, Max 83, 256 Bolzano, Bernard 49, 71, 73, 79-83, 114, 255, 259 Bradley, F.C. 208, 256 Brentano, Franz 56, 87, 94-97, 99, 113, 118, 123, 155-156, 172, 176, 255, 258-259 C

Chisholm, Roderick M. 155-158, 188189, 255, 257 Cochiarella, Nino 235, 255 D

Dallas, Willard Davidson, Donald Dodd, Julian Dölling, Evelyn Dummett, Michael

G

Gardies, Jean-Louis 201-203, 256 Geach, Peter 83, 256 Grossmann, Reinhardt 91, 163-166, 256 H

Haack, Susan 245, 256 Heanue, James 257 Hintikka, Jaakko 256 Hochberg, Herbert 208, 256 Hornsby, Jennifer 256 Horwich, Paul 257 Husserl, Edmund 7, 14, 21, 49, 56-61, 67-71, 93, 150, 170-171, 183-187, 191, 201, 256-257, 259 I

Ingarden, R.

172-175, 260

J

259 255 255 62-63, 255 255-256

E

Edgington, Dorothy Elliston, F.A. Engel, Pascal

Frege, Gottlob 7, 49, 59-61, 72-73, 8386, 93, 96, 101, 171, 187, 191, 198199, 205-206, 217, 231, 255-256

256 259 256

F

Field, Hartry 256 Findlay, John N. 52, 60, 93, 147-154, 184, 187, 233, 256 Fine, Kit 159, 237, 256 Fischer, K. 255 Føllesdal, Dagfinn 256

Jacquette, Dale 182-183, 211, 215, 219220, 225, 227, 230-231, 233, 236237, 242-243, 255, 257 K

Kalsi, M-L.Schubert 78, 166-169, 257258 Kripke, Saul 217, 257 Künne, Wolfgang 257 L

Lambert, Karel 158-160, 205, 226-227, 240-241, 257 Lenzen, W. 259 Levi, Issac 257 Lewis, David 229, 257 Libardi, M. 258 Lindenfeld, David 55-56, 93-94, 179, 187-189, 257 Lowe, E.Jonathan 257

262

ANNA SIERSZULSKA

M

McCormick, P. 259 McDonald, G. 256 McDowell, John 257 McIntyre, Ronald 184, 185, 186, 187, 259 Morscher, E. 61, 258 Mulligan, Kevin 4, 170-172, 202, 256, 258 N

Neale, Stephen

258

Russell, Bertrand 7, 79, 89-91, 93, 102, 127-148, 151-162, 170, 177, 191, 208, 211-212, 233, 235, 244-245, 255-256, 258-259 S

Simons, Peter 4, 60-61, 78, 160-162, 176-177, 258-259 Smith, Barry 184-187, 258-259 Smith, DavidWoodruff 184-187, 258259 Szaniawski, K. 176, 259

P

T

Parsons, Terence 210-211, 215, 221223, 225, 229-230, 232-235, 237, 243-247, 256, 258 Pa niczek, Jacek 4, 57-58, 199, 215, 223-225, 227, 230, 232-233, 237, 245-246, 258 Perry, John 255 Poli, Roberto 255, 257-258 Politzer, I. 255

Tarski, Alfred 43, 97, 159, 176, 203, 207, 259 Tegtmeier, Erwin 89-91, 259 Tennant, Neil 259 Terrell, D.B. 257

Q

Wittgenstein, Ludwig 4, 151, 259 Wole ski, Jan 4, 97, 176-178, 259 Wright, C. 256 Z

Quine, Willard V.O. 151, 209, 210, 242, 246, 258 R

Reicher, Maria Rolf, George Routley, Richard

W

258 255 233, 258

Zalta, Edward 57, 205, 212-213, 215219, 225, 227-228, 233, 235-238, 243, 260 egle , Urszula

172-175, 260

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

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]

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

ontos verlag