Meinong's Theory of Objects and Values

Citation preview

MEINONG'S Theory of Objects and Values

Oxford Uniq;ersity Press, Amen House, London E.C.4 GLASGOW NEW YORK TORONTO MELBOURNE WELLINGTON BOMBAY CALCUTTA MADRAS KARACHI LAHORE DACCA CAPE TOWN SALISBURY NAIROBI IBADAN ACCRA KUALA LUMPUR HONG KONG

MEINONG'S Theory of Objects and Values BY

J. N. FINDLAY, F.B.A.

SECOND EDITION

OXFORD AT THE CLARENDON PRESS

© Oxford University Press r963 PRINTED IN GREAT BRITAIN

TO PAUL

PREFACE

A

x1us VON MEINONG, fitly described by a dis­ tinguished English philosopher 1 as the 'infinitely courageous and pertinacious Meinong', was born at Lemberg in Poland in I 8 5 3. 2 We have, for purposes of information, given him the ennobling 'von' of which he never made use, shrinking, as he says, from 'privileges that lack an inner foundation'. He belonged, however, to a noble German family that had moved over to Austria: it was his father's professional duties that took them all to Poland. Meinong's education was in Vienna, first at the Academic Gymnasium, and, after I 8 70, at the University; in his doctorate examination in I 8 7 4 his Hauptrigorosum was History, but Philosophy was his Neben­ rigorosum. For the purposes of the latter, he had studied Kant entirely without the help of commentaries: it would be interesting to be able to hear once again his no doubt audacious, autodidactic sallies. After some further dalliance with history, and some profitable attendance at Carl Men­ ger's lectures on economics, which were to influence his value-theory, Meinong gave himself unreservedly to philo­ sophy. He devoted himself to the study of Hume under the supervision of Brentano, whom he had first encoun­ tered in connexion with his Nebenrigorosum. While much less intimate with Brentano than others were, partly, he says, owing to a deep desire for independence, he was able to write in his last years of the radiant memory-image of his master, in which a 'spiritualized beauty' was 'made golden by the sunshine of his own and my youth'. Of his two profound and sympathetic Hume-Studien, done under Gilbert Ryle in 'Plato's Parmenides', Mind, 1939, p. 328. The biographical facts in this Preface are taken from Meinong's brief autobiography in Philosophie der Gegenwart in Selbstdarstellungen, ed. Raymund Schmidt, series I, 1923. 1

2

VI

PREFACE

Brentano's supervision, the first (1877), on Hume's theory of abstraction, secured his 'habilitation', the second, on Hume's theory of relations, appeared in 1882: both were published in the Proceedings of the Imperial Academy of Sciences at Vienna, of which Meinong was later to be a Fellow. That Meinong should have served his first serious philosophical apprenticeship with Hume, places him in the Anglo-Saxon rather than the Germanic philosophical tradition, and it was in this tradition that he continued mainly to work. It was in the Anglo-Saxon world, like­ wise, that his philosophical reputation and influence were at their greatest. Meinong spent four years (1878-82) as a Privatdozent at Vienna, and then moved on to Graz, where he remained for the rest of his life, first as Professor Extraordinarius (1882-9), and then as Ordinary Professor (1889-1920). Graz is one of those rare places, in the furthermost corner of one world and on the edge of another, where every­ thing seems set in fixed perfection: its river, its plain, its town, its castle, its not too high and not too distant moun­ tains with their many exquisite vantage-points, are all wholly beautiful, whether in sunshine or in snow. The inhabitants share in the grace of the landscape and the architecture, and like these they stay, and do not alter: if they go away for a time, they hardly ever fail to come back. It is agreeable to think of Meinong spending all those years in this delightful place (from which even a call to Vienna did not tempt him), while the theory of objects slowly burgeoned and took shape. Apart from the foun­ dation of an Institute of Experimental Psychology in 1894, the first in Austria, there seem to have been few events during Meinong's professorship. His history was the his­ tory of his publications and of the academic activities of his small school of pupils. Among these publications the most notable were the Psychologisch-ethische Untersuchungen zur Werttheorie(1894),

PREFACE

Vll

which almost succeeds in formalizing ordinary morality; the composite school-publication Untersuchungen zur Gegen­ standstheorie und Psychologie ( I 904), to which Meinong contributed an article 'Ober Gegenstandstheorie'; the valuable but now little accessible epistemological essay Ober die Erfahrungsgrundlagen unseres Wissens (1906); the programmatic Ober die Ste/lung der Gegenstandstheorie im System der Wissenschaften (1906-7); the brilliant Ober Annahmen (191o), with its manifold contributions to psy­ chology, value-theory, &c., and its important introduction of 'objectives', the Satze-an-sich of Bolzano, as peculiar entia rationis; the long treatise Ober Moglichkeit und Wahr­ scheinlichkeit ( I 9 I 5), with its important doctrine of 'in­ complete objects'; the treatise Uber emotionale Prasentation ( I 9 1 7), a uniquely original essay in the epistemology of valuation; and the somewhat unpersuasive Zum Erweise des allgemeinen Kausalgesetzes (19 I 8). Meinong wrote many important articles which were collected by his pupils in the two volumes of Gesammelte Abhandlungen, one vol­ ume devoted to psychology, the other to epistemology and object-theory: a third, to be devoted to value-theory, was never issued. Several important articles on value­ theory, as well as the unreprinted Psychologisch-ethische Untersuchungen, are therefore practically inaccessible. The Grundlegung zur allgemeinen Werttheorie was published posthumously in 192 3, and a work entitled Ethische Bau­ steine is still in manuscript in the Library at Graz. Among Meinong's pupils probably the best known are Ernst Mally, his successor in the Chair at Graz, who died during the second World War; Stefan Witasek (d. 1 9 I 5), author of a Meinongian Grundlinien der Psychologie and of a little known but valuable .Asthetik; and V. Benussi, the psychologist. Christian von Ehrenfels, himself the author of an influential System der Werttheorie (1897-8), and a remarkable metaphysical Kosmogonie, was a pupil of Meinong's at Vienna, and so were Alois Hofler and

PREFACE

Vlll

A. Oelzelt-Newin. Other Meinong-pupils, such as R. Ameseder, E. Martinak, and R. Saxinger, have more or less been forgotten. The school of Graz never achieved the eminence and influence of the phenomenological school of Husserl, though both had their roots in Brentano, and taught simi­ lar doctrines at many points. Husserl's great influence began, however, when he forsook the brilliant dryness and systematic carefulness of his early Philosophie der .Arith­ metik and Logische Untersuchungen, for the exciting pro­ grammatic vistas and systematic transcendentalism of his later phenomenology. After about I 907 he may be said to have become absorbed in the main stream of German philosophical thought, accepting its unqualified idealism, and its somewhat dogmatic way of exploring the a priori, and it is to this change, as much as to his philosophical genius, that his immense later influence is due. Meinong, however, never became part of that stream, and his latest writings show the same spirit of logical dryness and piece­ meal caution as his first. He tells us that 'the continuously increasing preoccupation with Kant has made it into a tradition to give new thoughts, where possible and even beyond this, the form of Kantian conceptions. If one con­ siders, however, how uncertain the interpretation of Kant has become, it seems to me that one can become more than doubtful whether this is a good tradition. I have thus not thought, in any case, that the method of justifying positions by way of Kant was indispensable, and have even thought it detrimental, where it made one's treatment cir­ cuitous. I have therefore avoided it.' 1 It is not remarkable, therefore, that Meinong and his school came to be regarded as travellers on a philosophical by-path or even dead-end, and that they were described as 'scholastic' or merely as 'queer' (fremdartig). 'Scientific research', Meinong re­ marks, 'is generally a lonely business, and one that produces 1

Philosophie der Gegenwart,

p. 55.

PREFACE

lX

loneliness.' It is plain (see Ober Moglichkeit und Wahr­ scheinlichkeit, pp. ix and x) that Meinong rejoiced in the coincidence between many of his independently formed views and those of Husserl, and would have welcomed fruitful co-operation from that quarter had not Husserl, with his jealously proprietary attitude towards certain con­ ceptions, made this impossible. 1 Meinong stoically encountered an expected death to­ wards the end of I 920. The character-profile that emerges from his writings and from report is the somewhat old­ fashioned one of a crusted, formal exterior concealing deep loyalties and attachments to persons and things, many extremely pure and lofty motives and sentiments, great generosity, complete honesty, vast patience and courage, and a somewhat oversensitive, readily injured pride. Philosophically, as well as personally, there is no one Meinong so much resembles as G. E. Moore. They share the same independence of tradition, the same unwillingness to be browbeaten by the 'deep' utterances of philosophers­ Meinong gently sets aside Schopenhauer's explanation of morality by an underlying identity of the 'will', much as Moore sets aside Bradley's assertion of time's un­ reality-the same reluctance to make declarations too firm or too wide, the same cautious quasi-empiricism of approach even in the realm of the a priori, the same sensitiveness and deference to what people actually believe and commonly say, coupled with the same willingness to put forward highly daring analyses of common notions, the same unrepentant, almost obsessive realism, and lastly, though many would question it, the same common sense, in the sense of an unwillingness to abandon what we plainly understand and know, and what forms the firm foundation of our discourse, at the behest of theories much less lucid and indubitable. With this go the typical Mooreian faults of elementarism and atomism, and of a piecemeal 1

Ibid., PP· 55-56.

PREFACE

X

progress so myopic as often to miss the large connexions essential to philosophical insight. A few quotations may here be given as typical. 'It is certainly desirable to find an exact conceptual formulation for the mutual relevance, evident in practice, of all that is traditionally done under the name of philosophy. So far I have not been successful in doing so.' 1 'The theory of objects is no doubt an a priori science, possibly the a priori science, and subsistence and extraexistence (Au}Jersein) must be garnered from the nature of these objects, therefore a priori. None the less this knowledge of being goes back to a direct apprehension of these objects as a sort of quasi­ experience, so that even the theory of objects permits us to tread the path from below upwards, like the empirical sciences.' 2 'Like all apprehension knowledge is also, to a quite particular degree, an entirely peculiar performance, not reducible to anything more elementary. But whoever believes in knowledge will also, wherever he recognizes it, and provided he is consistent, not be able to doubt this performance. It is essential to this performance that it relates to something that in no manner coincides with the knowing experience . . . but always transcends the latter.'3 Meinong's most famous and characteristic doctrine, that of an unbounded realm of objects which are daseinsfrei, indifferent to the antithesis between being and non-being, and his frank espousal of the anti-Parmenidean position that what is not is as much the object of significant refer­ ence and valid examination as what is, might seem to prove Meinong's extravagance and unsoundness, his wide ex­ ceeding of the bounds of common sense. The doctrine, however, is eminently arguable at a common-sense level, and was once even justified by Russell on the basis of 'perception'. 4 Certainly it initially approved itself both to 1 3 4

2 Em. Preis., p. IOJ. Phil. der Gegenwart, p. r2. Phil. der Gegenwart, pp. 42-43. 'Meinong's Theory of Complexes and Assumptions,' Mind, r904, p. 2r7.

PREFACE

xi

Russell and Moore, and might have continued to deserve their approbation, had they sufficiently considered the qualifications with which Meinong held it. It is strange, further, that Meinong's object-theory should have been regarded by some as a bewildering and tangled 'jungle' : 1 it resembles rather an old formal garden containing some beautiful and difficult mazes. Meinong's thought had a relatively poor and rapidly dwindling reception in the German philosophical world. In the obliteration of practically all philosophical land­ marks by the blanketing snowfall of Heidegger-one of those great natural cataclysms to which even philosophy seems liable, and to which recent British philosophy offers analogies-the work of Meinong was plunged into neglect. It was in the Anglo-Saxon world that his thought, that certainly belongs to the tradition of Ockham and Hume, achieved both a temporary influence, and a lasting, if equi­ vocal, respect. Meinong's teachings certainly did not pene­ trate to the rank and file of British philosophers, but they impinged on the most eminent and understanding, and through them affected the rest. Russell was early aware of Meinong's great merits, and devoted the generous, brilliant 'Meinong's Theory of Complexes and Assumptions' (Mind, I 904) to their ex­ position and partial defence. It is well known that Russell's Theory of Descriptions, that famous and historically im­ portant piece of logical analysis, was, in part at least, an attempt to provide an alternative to Meinong's doctrine of Au}Jersein, of extraexistential objectivity. Russell also dis­ cussed Meinong's theory of content in the first chapter of The Analysis of Mind. Unfortunately Russell was far too concerned to advance from Meinong to his own notions and conclusions to bother to get Meinong quite straight, and the accounts he put into circulation of Meinongian contents as consisting of sense-data and images, and of 1

W. Kneale, Probability and Induction, p.

r2.

Xll

PREFACE

Meinong's non-existent objects as 'subsistent', are simpli­ fying travesties of Mdnong's complex opinions. In Moore, Meinong met with a far more careful and receptive student: the doctrine of content is correctly ex­ pounded and valuably criticized in 'The Subject-Matter of Psychology' (Aristotelian Society Proceedings, 1909-10, pp. 36-62), and Moore's remarkable Morley College Lec­ tures, delivered in 1910-11 and published as Some Main Problems of Philosophy in 19 5 3, deal mainly with Meinong­ ian problems in an almost wholly Meinongian manner.We may note that Mooreian analysis found itself unable to deal satisfactorily with the problem of meaningful reference to the false or non-existent, the precise problem with which Meinong struggled so valiantly, and that Moore never thought that the Theory of Descriptions had solved all the philosophical difficulties connected with this sort of refer­ ence, nor imagined that Russell had worsted Meinong in the manner attested by the prevailing legend. Through Russell and Moore the echoes of Meinong were carried to Wittgenstein : no one who considers the stress on facts and being-the-case in the Tractatus Logico­ philosophicus, or the frequent references to the puzzling character of thought about the non-existent in The Blue Book and elsewhere, can doubt that Meinong is here exer­ cising a remote influence. Meinong also had a considerable influence on C. D. Broad, who reviewed the second edition of Ober Annahmen in Mind, 1913, and on G. Dawes-Hicks, who contributed a valuable article entitled 'The Philo­ sophical Researches of Meinong' to Mind, 1922. In the United States the direct influence of Meinong extended more widely. The writers of the co-operative volume called The New Realism (1912) explicitly associated them­ selves with Russell, Moore, and Meinong as their 'big brothers overseas'. Edwin Holt's theory-stated in The Concept of Consciousness (191 3)--of a world of 'neutral en­ tities', variously divided up by 'conscious cross-sections',

PREFACE

Xlll

represents a greatly simplified, behaviouristically modified version of Meinong's doctrine of AuJJersein, and the same applies to Santayana's beautifully phrased, but less exactly thought-out, doctrine of a 'Realm of Essence'. It is not widely known that T.S. Eliot wrote a doctoral dissertation on Meinong which still reposes in the Library at Harvard: it seems a pity that no reference to AuJJersein has strayed into The Waste Land. There was, however, a second season of Meinongian influence in Britain, the time at the end of the twenties and the beginning of the thirties when the Cam overflowed into the Isis, and Oxford for the first time took serious cognizance of the thought of Russell, Moore, and Witt­ genstein, and of those who had influenced them.Among these last Meinong held an honourable place.The author of the present book belongs to this period, and so do his contemporaries or near-contemporaries William Kneale and Gilbert Ryle.Ryle's estimate of Meinong at that time may be gauged from the following quotation: 'Meinong was the sort of reformer who makes revolutions inevitable, yet himself stops short of seeing that they are even possible. He was a philosophical Kerensky.Both in epistemology and in logic, with terrifying assiduity and remarkable rigor­ ousness of reasoning, he carried to their extreme con­ clusions the implications of presuppositions which no one had yet questioned, and which he himself did not question. But of these conclusions he never said what has to be said, "By God, this is impossible." He was perhaps the supreme entity-multiplier in the history of philosophy, and yet, I suppose, the main service which he really rendered philo­ sophy was to force logicians to see that "wherever possible logical constructions are to be substituted for inferred en­ tities " ... Meinong is in my opinion important for ... three main reasons.First, he was very largely responsible for the de-psychologizing of logic. Next he raised, and raised together, the logical and metaphysical questions of 824181

b

PREFACE

XIV

the nature and status of relations, numbers, facts, univer­ sals, negation, possibility, probability, necessity. He also forced philosophers to investigate the whole problem of what is meant by such predicates as "exists", "subsists", "is an object", &c. But thirdly, and chiefly, he accepted the traditional doctrine of Terms in logic, and by genera­ lizing the issues and remorselessly drawing the conclusions from the premisses, he showed, though he did not see, that the whole structure was rotten. If the orthodox theories of terms were true almost the whole fantastic hierarchy of Meinong's non-actual entities would have to be accepted.' 1 It will be seen from the above that the role of Meinong in modern British philosophy has been largely that of a philosophical Aunt Sally, who is honoured in being struck down. Except to a very few like Ryle, he was known only by hearsay. Possibly his fate is happier than that of Moore who, imperfectly read, has been treated as the mere 'fore­ runner' of another philosopher, and whose most character­ istic and valuable doctrines have been interpreted in a manner quite at variance with their obvious sense. Recently there have been signs that Meinong may come to be more understandingly treated: the formal discipline of semantics has found it profitable to revive something like his object­ theory, and a philosopher like Chisholm has translated some of his work. Meinong's importance as a thinker ought, in our day, to be no longer a matter for debate or doubt. The present work was first published in 1933, and is being reissued now, since some have found it useful in understanding Meinong's work, and since it seemed proper that an easy access to Meinong should be possible for English-speaking readers. I have made only trivial changes in its eight original chapters. If I rewrote them now, I should no doubt write them rather differently, but I should not write them better, and my present ideas might obtrude 1

Oxford Magazine,

26

October

r933.

PREFACE

xv

on a body of doctrine with which I was then, to a much greater degree, identified. I have, however, added a ninth and a tenth chapter on Meinong's extremely important value-theory, which has incidentally left more mark on my thought than any other part of Meinong's teaching. Others have discussed the value-theory (for example, H. 0. Eaton in The Austrian Philosophy of Values, I 930), but not, to my knowledge, in close connexion with the theory of objects, of which it forms the completion. And in a final eleventh chapter entitled 'Appraisal of Meinong' I have tried to fulfil an injunction given me by Ryle in I 9 3 3, that I should write one more chapter in which I stood back from my canvas and asked: 'What does it all amount to ? What is Meinong for ? Why is it useful and not only hard work for philosophers to discover what Meinong thought ?' I do not imagine that my answers to these questions would have satisfied him then, nor that my present answers will be satisfactory now, but I give them for what they are worth.

J.

King's College, London, ]une 1 9 6 2

N. F .

CONTENTS I . THE DOCTRINE OF CONTENT AND OBJECT I . The doctrine of content in relation to modern English realism

r

I I . Brentano's doctrine of intentionality. The distinction of the idea, the judgement and the phenomena of love and hate

3

I I I . The use of the word 'content' in Brentano and in the early writings of Meinong. The 'immanent' and the 'transcendent' object

6

I V . Twardowski's account of the content of ideas. His arguments for the existence of contents

8

V . The content in Twardowski does not resemble the object nor any part of it

12

V I . Meinong's arguments to prove that the object of an experience is not a part of it. The pseudo-existence of objects

I7

V I I . Argument for the existence of contents. Contents and acts

22

V I I I . Contents are not mental images or sense data, but are often wholly unlike their objects

26

IX. Can contents be perceived or are they merely inferred ? The pseudo-obj ect. Reasons for thinking that contents can be perceived

28

X. Examples from Meinong to prove that contents are in some cases immediately perceived

32

XI . The relation of content to object XII . Meinong's theory fails to explain completely how states of mind pass beyond themselves, but is the basis of an important type of research

35

37

I I . THE PURE OBJECT AND ITS INDIFFERENCE TO BEING I. The pure object lies outside of the antithesis of being and non-being. Whether it is or not, makes no difference to what it is. Comparison with Russell's theory

42

I I . Non-existent objects appear to be involved in negative facts. Arguments to prove (a) that there are genuine negative facts, (b) that they do involve non-existent objects . Reasons for our prejudice against non-existent objects

50

CO N T E N T S

XVIU

I I I . THE THEORY OF OBJECTIVES I . Objectives and facts I I . The expressive and significant use of words. Sentences express j udgements or assumptions, but they mean objectives. Objectives and objecta

59 60

I I I . Ameseder's characterization of obj ectives

69

IV. Objectives are objects of higher order

7I

V. Objectives are incapable of existence

73

V I . Objectives d o not depend for their being o n superordinate objectives VII. The relation of objectives to time

VIII. Negativity a characteristic of objectives, not of objecta

75 77 8I

IX. Comparison of the theory of objectives with the theory of propositions

83

X. Objectives cannot be reduced to characteristics, relations, or concrete objects

89

XI . Objectives are not complexes. Meinong's account of the relation of objectives to complexes

94

XII. Criticism of Russell's theory of j udgement as a many-termed relation

1 00

IV. THE MODAL MOMENT I . The relation of an objective to its factuality

102

I I . ' Watered down' and 'full-strength' factuality and existence. The function of the modal moment

1 03

I I I . 'Full-strength' factuality or existence cannot be assumed to be present where it is not present

1 06

IV. How it is possible to think of the factuality of the unfactual, or the existence of the non-existent

1 09

V Mally's theory of determinates

1 10

V. OBJECTS OF H I G HER ORDER I . The distinction between existence and subsistence

113

I I . Meinong's defence of analysis

1 16

I I I . Meinong's criticism of the reduction of characteristics to relations of similarity

118

CO N T E N T S

IV. Development ofMeinong's theory of the relation ofan object to its characteristics V. Relations and their fundamenta V I . Meinong's original psychological theory of relations. The distinction between ideal and real relations. His abandonment of this theory VII. The Principle of Coincidence. Real and ideal complexes

XIX

123 128 1 32 1 37

VII I . Meinong's later theory of the distinction between ideal and real relations IX. A relation is not a constituent of the complex it generates

1 45

X. Comparison of Meinong's conception of complexes with the theories of Russell and Wittgenstein

1 46

XI . Criticism of Meinong's conception of relations and complexes as 'obj ects of higher order' . Meinong's views on continua

1 48

1 42

VI. THE THEORY OF INCOMPLETE OBJECTS I. All objects which have being are completely determined I I . D iscussion of Meinong's views on relational properties

1 52 1 53

I l l . Introduction of incomplete objects

1 56

IV. Incomplete objects and the Law of Excluded Middle

1 59

V . Reasons for our inattention to incomplete objects. An ambiguity in the word 'universal' VI. Implexive being and so-being VII. The reference by way of being and the reference by way of so-being VIII. The completed obj ect. Difficu lties of Meinong's doctrine IX. Analytic and synthetic j udgements X. Treatment of Meinong's problem in the light of Mally's theory of determinates

1 62 1 66 1 70 1 74 1 80 1 82

VII. THE MODAL PROPERTIES OF OBJECTIVE S I . The modal properties of objectives are wrongly attributed to j udgements I I . Factuality and truth

I 86

I I I . Necessity and the experience o f understanding. The two species of necessity

l 87

IV. Inhesive and adhesive factuality

I 92

l85

xx

CO N TEN T S V . Possibility and probability. Degrees of possibility V I . The upper and lower limits of possibility

V I I . The modal level of objectives V I I I . Internal and external possibility IX. Possibilities and the Law of Excluded Middle

202

204 205

X. The function of incomplete obj ects in possibilities

209

XI. The application of possibility to complete objects

214

VIII. THE APPRE HENSION OF OBJECTS I. The relation of the theory of apprehension and knowledge to the theory of objects I I . Active and passive experiences. The function of ideas

2I8 2 I9

I I I . The ideas of production and their functions

222

IV. Experiences of thinking. Moments involved in the j udgement, assumption, and surmise

225

V . M einong' s theory of presentation VI. Meinong's theory of our reference to objects V I I . Meinong's theory of our awareness of complexes V I I I . Implicit and explicit apprehension

23o 238

24 5

IX. Meinong's theory ofknowledge. The experience of 'evidence'

24 8 25 I

X. The evidence of surmises. The epistemology of perception, memory, introspection, and induction

2 56

IX. VALUATION AND VALUES I . Meinong's main writings on value-theory I I . Doctrine of the Psychologisch-ethische Untersuchungen.Valuations as existence-feelings

2 64

266

III. The 'psychological presuppositions' of valuation . Opposition of value-feelings to sense-feelings, aesthetic feelings, and know­ ledge-feelings. Relation of value-feelings to judgements of instrumentality, of non-existence, and to modal distinctions. Relation to the distinction of Ego and Alter. Errors in valuation

269

IV. Analysis of moral value : distinctions of the meritorious, the correct, the allowable, and the censurable. Types of egoistic and altruistic motivation . Use of binomial formulae to express principles of moral valuation

27 5

CONTENTS

XXI

V. Moral valuation measures good-\vill, ill-\\·ill, and indifference. Relation to valuations of justice . The 'su bject' of moral valuation as the 'environing collectivity' of unconcerned persons. Connexion with obligation and moral ascription

28 1

V I . Doctrine of the Grund!egung. Feelings as the main, and desires as subsidiary value-experiences. Valuations as being­ feelings. Relation to knowledge-feelings, aesthetic feelings, and sense-feelings

289

V I I . The counter-feelings : j oy-in-being and sorrow-in-non-being, sorrow-in-being and j oy-in-non-being. Meinong holds that such counter-feelings ought rationally to be of equal intensity whatever they in fact are. This view of Meinong's has absurd consequences . 'Potentialization' of the concept of value

29 5

X . DIGNITATIVE S AND DESIDERATIVES I . The notion of emotional presentation and its relation to the modern 'emotive' theory. The doctrine of emotional presen­ tation an unexpected by-product of Meinong's doctrine of content. Reconsideration of this doctrine 303 I I . Emotional presentation may b e self-presentation o r other­ presentation, whole-presentation or part-presentation : only emotional part-presentation can introduce us to peculiar ob­ j ects. Instances of such emotional part-presentation. Corresponding instances of desiderative part-presentation

30 7

I I I . Dignitatives and desideratives and their relation to objectives and obj ecta

3I3

IV. That dignitatives and desideratives are objects does not prove that they have being. That they have being can be known only by an evident j udgement, not by an emotional experience. The evidence for the being of dignitatives and desideratives is a priori, but, since we only possess it feebly, may be eked out by a quasi-empirical approach

3I5

V . Meinong's views regarding absolute dignitative3 (dignities) and absolute desideratives ( desiderata)

319

XI. APPRAI SAL OF ME INONG I . Contemporary difficulties in being interested in the problems and answers of Meinong

322

I I . Revival of Plato's cave-image to assist in the appraisal of Meinong's philosophical contributions

328

CONTENTS

xxii

Ill. The main merit of Meinong is that he is a thoroughgoing and impartial empiricist and phenomenologist : he describes the world as we actually experience it, and not as distorted by logical assumptions or doctrines of origin

33I

IV. Meinong is great in recognizing that the non-existent, the non-factual, and the absurd are essential elements in des­ cribing the experienced world . He is also great in recognizing a necessary connexion between featu res attri buted to objects and interior phases of experience

3 37

V. Meinong's realistic approach to all the entities of reason and unreason cannot however be sustained.

3 40

V I . 'Thinking' and its cognates are not relational expressions nor can they be said to express relations, since relations only obtain when all their terms exist. The non-existent and the false are not constituents or terms of relations in our thoughts or beliefs, nor in anything else, though they enter indispen­ sably into the description of our thoughts and beliefs and of objects in the world 343 V I I . Meinong's faults are those of the elementaristic psychology of his day. Merits of Meinong as a philosopher of the 'understanding'. His resemblance to G. E. Moore 34 5 INDEX

3 49

NOTE T HE following is a list of the publications of Meinong which have been referred to in the text:

Hume Studien I. Zur Geschichte und Kritik des modernen Nominalismus, 1 8 77, reprinted in Gesammelte Abhandlungen, vol. i (Barth, Leipzig), referred to as Krit. Nom. Gs. Abh. I . Hume Studien II. Zur Relationstheorie, 1 8 8 2, reprinted i n Gesammelte Abhandlungen, vol. ii, referred to as Reith. Gs. Abh. I I . Zur erkenntnistheoretischen Wiirdigung des Gediichtnisses, 1 8 86, reprinted in Gesammelte Abhandlungen, vol. ii, referred to as Gediichtnis Gs. Abh. I I . Ober Begriff und Eigenschaften der Empfindung, 1 8 89, reprinted i n Gesam­ melte Abhandlungen, vol. i, referred to as Empfindung Gs. Abh. I . 0ber P hantasievorstellung und Phantasie, 1 8 89, reprinted i n Gesammelte Abhand!ungen, vol. i, referred to as Phantasie Gs. Abh. I . Zur Psychologie der Komplexionen und Relationen, 1 8 9 1 , reprinted i n Gesam­ melte Abhand!ungen, vol. i, referred to as Psy. Kom. Gs. Abh . I . Beitriige zur Theorie der psychischen Analyse, 1 8 9 3 , reprinted i n Gesammelte Abhandlungen, vol. i, referred to as Psy. An. Gs. Abh. I . Psychologisch-ethische Untersuchungen zur Werttheorie, 1 89 4 (Leuschner and Lubensky, Graz, out of print), referred to as Psy .-eth. Unt. Ober Werthaltung und Wert, 1 89 5 , in Archiv fur systematische Philosophie. Ober die Bedeutung des Weberschen Gesetzes, 1 896, reprinted in Gesam­ melte Abhandlungen, vol. ii, referred to as Web. Ges. Gs. Abh. I I . Ober Gegenstiinde hiiherer Ordnung und deren /7erhiiltnis zur inneren Wahr­ nehmung, 1 8 99, reprinted in Gesammelte Abhand!ungen, vol. ii, referred to as Geg. hoh. Ord. Gs. Abh. I I . Abstrahieren und /7erg!eichen, 1 900, reprinted in Gesammelte Abhandlungen, vol. i, referred to as Abst. u. /7ergl. Gs. Abh. I . Ober Annahmen, fi rst edition, 1 902 (Barth, Leipzig, o u t o f print), referred to as 0. A., ed. 1 . Bemerkungen iiber den Farbenkb'rper und das Mischungsgesetz, 1 90 3 , re­ printed in Gesammelte Abhandlungen, vol. i, referred to as Farbenko"rper Gs. Abh. I . Ober Gegenstandstheorie, 1 904, reprinted in Gesammelte Abhandlungen, vol. ii, referred to as 0. Gegth. Gs. Abh. I I . Untersuchungen zur Gegenstandstheorie und Psychologie, 1 904 (Barth, Leip­ zig, out of print), written in collaboration with others, referred to as Untersuchungen.

XXlV

N O TE

0ber die E,jahrungsgrund!agen unseres f//issens, 1 9 06 ( Springer, Berlin, out of print), referred to as Erfg!. Ober die Ste/lung der Gegenstandstheorie im System der Wissenschaften, 1 906-7 (Voigtlander, Leipzig, out of print), referred to as Ste/lung. 0ber Annahmen, second edition, I 9 I o (Barth, Leipzig), referred to as 0. A. Fur die Psycho!ogie und gegen den Psycho!ogismus in der a!!gemeinen Wert­ theori e, 1 9 1 2, published in Logos. Ober Mog!ichkeit und Wahrschein!ichkeit, 1 9 1 5 (Barth, Leipzig), referred to as Mbg. 0ber emotiona!e Priisentation, I 9 1 7 (report of the Kaiserliche Akademie der Wissenschaften, Vienna), referred to as Em. Priis. Zum Erweise des al!gemeinen Kausa!gesetzes, I 9 I 8 (report of the Kaiserliche Akademie der Wissenschaften, Vienna), referred to as Erw. Kaus. Zur Grund/egung der al!gemeinen Werttheorie, 1 9 2 3 ( Leuschner and Luben­ sky, Graz), referred to as Grund!egung. Ethische Bausteine, an unpublished posthumous fragment in the University Library at Graz.

I THE DOCTRINE OF CONTENT AND OBJECT I

o D E R N realism has, in the course of its de­ velopment, proved itself quite as difficult and unstable as any other philosophical theory; it may be claimed, however, that it has left certain distinctions clearer than they were before. In the case of intelligences as confused as ours, which cannot hold more than three points together in thought, which cannot think of B without forgetting A, and which cannot see the most obvious implications of a view unless made to do so by the grossest absurdities, this is quite a considerable achievement for thirty years of discussion and polemic. Realistic theory has strengthened our insight in one most important respect: we are no longer tempted to con­ fuse the experiences which we live through with the obj ects, physical, mental, or ideal, which are presented to us by their means. We are no longer able to profit by the ambiguities of language in order to pass from those pecu­ liar modifications of our inner life which we call sensations and emotions, or those peculiar inner activities which we call judging and willing, to those wholly different entities whose nature it is merely to be there, to exemplify or to be an essence, to stand in certain relations to each other, but not to act or suffer. An ' idea' as an inner process, as something which happens to us, has once and for all been distinguished from an 'idea' as the term which stands before our thought. A judgement with its peculiar colour of belief, and its reaching out beyond the subject and his states, will never be confused with the colourless piece of 824187

B

2

T H E D O C T R I N E OF C O N T E N T A N D O B J E C T

fact which is, in favourable circumstances, apprehended by its means. Whether we should be justified if we went farther and distinguished between a beauty which consists in the fulfilment of certain objective requirements, and the pleasure by which we are aware of such a fulfilment, or between a wholly objective goodness inherent in certain states of affairs, and a felt impulsion to produce such states, is a harder question, with regard to which it seems quite clear that agreement will never be reached. But the distinction between an experience, as something which is a real moment in our mental life, and an object which we can handle and look at, or apprehend in a less concrete way, but which we can never really live through, remains an ultimate one. This does not prevent us from returning to idealistic explanations if we wish to; we are quite free to hold that the independence of objects is an independence of abstrac­ tion, that they have no being unless there are experiences in which they are presented; this also is an issue into whose insolubility we have almost an a priori insight. But, how­ ever we decide such difficult and unprofitable questions, it seems clear that the absolute distinctness of experiences and their objects must remain unaffected. Higher unities may be introduced, but the terms which are unified in them will not fuse into a featureless identity. Realism in English-speaking countries has in general contented itself with exposing the fallacies of idealistic arguments, with showing that it is perfectly possible for our minds to gain a foothold in a world of things to which reflection is alien; it has then proceeded to deal with the world of objects, among which minds are included as an eccentric species, without troubling itself much further about epistemology. The theory of Twardowski, Meinong, and Husserl, which we propose to study in this chapter, holds that the relation between an experience and the object to which it points is not really the simple, straight-

T H E D O C TRI N E O F C O N T E N T A N D O B J E C T

3

forward thing which it appe�rs to be, and that we can only understand how an experience can point beyond itself, if we suppose that there is in that experience a peculiar element which makes such a self-transcendent reference possible. This element was called the content (lnhalt) of an experience by Twardowski and Meinong, while Hus­ serl, partly because this name was unsuitable and partly because he could not make use of names which other people had invented, called it the matter (Mate rie) of a mental act. 1 It is only because we live through contents that we can refer to objects ; these two sorts of things stand in a necessary relation to each other, which does not, how­ ever, prevent them from existing quite independently in totally different contexts, and having no vestige of mutual resemblance. II

In approaching this theory, it will be well to begin by turning to a well-known passage from Brentano's Psycho­ logy from the Empi rical Standpoint, a passage which almost attained the position of a Credo, inasmuch as all the philosophers who made Brentano's remarkable work the starting-point of their investigations begin their treatises by repeating it or something very like it. Brentano had attempted to discover some characteristic which would distinguish mental states, or, as he called them, ' psychical phenomena', from all other types of phenomena. This, to quote his own words, he believed himself to have found in a property ' which the Schoolmen of the Middle Ages called the intentional (or mental) inexistence of an object, and which we, although such expressions are not free from ambiguity, should describe as the relation to a con­ tent, or the direction to an object (by which we need not understand a reality), or an immanent objectivity. Every mental state possesses in itself something which serves as 1

See Logische Untersuchungen, vol. ii, part i, p. 4 1 5.

4

T H E D OCTR I N E O F CO N T E N T A N D O B J ECT

object, although not all possess their objects i n the same way.' 1 In the states of mind known as presentations (Vo r­ stellungen) something is presented to us, in those states known as judgements we either accept or reject something that is presented, in the emotional states of mind we either desire or feel averse from objects and states of affairs. The ability to direct itself to something different from itself is, according to Brentano, altogether peculiar to mental phenomena; it is evident that no purely physical pheno­ menon could exhibit such a power. The subject-matter of psychology becomes therefore the study of the various ways in which a state of mind can direct itself to an object, of the various fundamental types of ' intention': of these Brentano held that there could only be three. The most fundamental of them was the Vo r­ stellung, an untranslatable word, for which ' idea' or ' pre­ sentation' may serve as inadequate equivalents. Whenever anything stands before consciousness, whenever we see a colour or hear a tone, construct an image in our fancy, or understand the meaning of a word, we are living through a Vo rstellung in Brentano's sense. A judgement distin­ guishes itself from a Vo rstellung in that, when we judge, we accept something as true or reject something as false ; such judgements are present even in simple cases of per­ ception and memory where we trust our experiences too implicitly to express such trust in words. Finally we have 'the phenomena of love and hate', in which something is accepted as good or rejected as bad; the pleasure and dis­ pleasure which obj ects awaken in us, the desires and hopes which spring from such pleasure and displeasure, and finally the voluntary decisions in which desires terminate : all these are regarded by Brentano as variants of one fundamental type of attitude, whose two opposed forms are loving and hating.2 1 2

P sychologie 'Vom empirischen Standpunkt, See Psychologie, I I . vi . 3 ; I I . viii. r .

II.

i. 5.

T H E D O C TR I N E O F C O N T E N T A N D O B J E C T

5

These three types of intention stand to each other in relations of one-sided dependence. A Vo rstellung is the logical p rius of all other types of experience, in that we can perfectly well conceive of a being who had ideas of obj ects but who never passed any judgements on them or had any emotional attitudes towards them; on the other hand, a person must have ideas of obj ects in order to judge or feel about them. The judgement is built on the //o rstellung, but is independent of the phenomena of love and hate; the lumen siccum of scientific thought proves that this is so. On the other hand, Brentano holds that the phenomena of love and hate presuppose both the judgement and the //o rstellung; j oy, grief, hope, and fear are states of mind which stand in the most intimate and necessary relation not only to our ideas but also to our beliefs. 1 Brentano's threefold division of intentional experiences has not found general acceptance; the distinction between the idea and the judgement maintains itself in his succes­ sors with various modifications, but the phenomena of love and hate tend to be divided in the traditional way into experiences of feeling and experiences of desire or will . It is also far from clear that Brentano has really discovered the distinguishing characteristic of the mental when he points to the way in which states of mind are directed to obj ects; it is, in fact, doubtful whether any such single distinguishing characteristic exists. The characteristic in question seems to be much more strongly present in those states of mind, such as judgement or desire, which we class as activities, than in more passive states, in which our experiences have a certain quality or a certain 'colour', but in which hardly any reference to an obj ect is discernible. In Meinong's later theory, he holds that a //o rstellung, so far from actually setting an obj ect before the mind, only provides the necessary basis for such an explicit apprehen­ sion. The //o rstellung in itself is a wholly passive experience, 1

Ibid., II. ix.

2.

6

T H E D O C TR I N E O F C O N T E N T A N D O B J E C T

to which we surrender ourselves without endeavouring to make anything out of it; such experiences are, in their pure form, infrequent in adult mental life, but of their occasional occurrence there can be no doubt. 1 If some one were to look at a coloured pattern in a wholly passive frame of mind, he would presumably live through or enjoy certain mental modifications or Vorstellungen, but he would only be aware of the pattern and its properties if he abandoned this passivity. The potential direction to an object which we find in the Vorstellung becomes a com­ plete and explicit apprehension of something when the active experience of judgement or assumption supervenes, when the presence of an object is acknowledged or its nature is recognized. It is therefore doubtful whether the direction to objects is as fundamental a property of mental states as Brentano supposed, but it seems to be evident that it is a property which no non-mental thing could possibly possess. Whether stones refer to objects or not, it is impossible to say, but, if they do, they undoubtedly have, or are, minds. This rules out as a priori unthinkable all those modern theories in which something analogous to a mental reference is attributed to entities which are nevertheless not held to have minds. The events of Professor White­ head are held to enjoy 'uncognitive apprehensions' of other events, and to 'mirror' the modes of their pre­ decessors and successors; 2 if Brentano is right, there is no good sense which can be given to such statements. I II

It will be noted that Brentano makes no distinction between the content (Inhalt) of a mental state and its object ( Geienstand). Both words mean for him the 'some­ thing' which has intentional inexistence in the mental I

0. A.,

p p . 2 2 7, 2 2 8 .

2

Science and the Modern World, p. 8 6 .

T HE D OCTR I N E O F CO N TEN T A N D O B JECT

7

state, to which the state is directed. Meinong's original use of the words Inhalt and Gegenstand is quite as undis­ criminating. If I think of white or black, round or oval, these things are said by Meinong to be contents, and if I emphasize their relation to my thought, I may call them objects. They are in any case purely mental, and are each an integral part of an idea. 1 The notion that such entities as white or black, round or oval, could in some way find lodgement in a mind, gave rise to a manner of speaking in which they were described as immanent objects; we have seen Brentano using such language and Meinong adopted it without question in his earlier phases. 2 We may quote a typical passage from the Logic of Hofler, published in I 890, in whose production Meinong assisted : ( I ) What we have called the 'content of the presentation and the j udgement' lies entirely within the subject, like the presenting or the j udging act itself. (2) The word 'Obj ect' (Gegenstand or Objekt) is used in two senses : on the one hand it means the indepen­ dently existing thing . . . to which our presentation or j udgement directs itself, on the other hand it means the more or less accurate image of such a reality, which exists in our mind ; this quasi-image, or rather sign, is identical with the content which we mentioned above. I n order to distinguish it from the obj ect which we take to be independent of our thinking we call the content of a presentation or j udgement (or of a feeling or act of will) the 'immanent or inten­ tional obj ect' of these mental phenomena. 3

Behind the immanent object lies the transcendent object, the real thing in itself; of this we can know nothing except that it is the cause, or part of the cause, of the presentation of the immanent object. When a relation such as causality is in some manner given to our thought, and one of its terms, the presentation of the immanent object, is also 1 2

J

See Reith. Gs. Abh. I I, p. 1 40. See Geg. hoh. Ord. Gs. Abh. I I, p. 3 8 2 ; Erfgl., p. 5 6 . Hofler, Logik, § 6, q uoted from Twardowski, p. 4.

8

T HE D O C TRI N E O F C O N TEN T A N D O B JE C T

given, we are able to describe indirectly the other term of this relation, that is, the transcendent object or thing in itself. 1 Meinong's original doctrine was therefore entirely on orthodox representational lines; he started with a theory as to whose complete impossibility most schools of philo­ sophy are in agreement, but which will maintain its hold on our minds till the end of time, because all its consistent alternatives are difficult and doubtful. The absurdity of locating the mountain-ranges of Asia, my neighbour's state of envy, my own experiences of yesterday, in my thoughts about such objects, or, on the other hand, of holding that I am not really thinking about these things, but about other, wholly private objects, will always seem slight in comparison with the extravagance and strange­ ness of a fully developed realism or idealism. IV

Meinong acknowledges that he owes his recognition of the distinction between content and object to Twardow­ ski's study on this topic.2 This is unquestionably one of the most interesting treatises in the whole range of modern philosophy; it is clear, concentrated, and amazingly rich in ideas. It represents a transitional phase between the image-theory of knowledge exemplified in the passage quoted above and the fully developed content-theory of Meinong. According to Twardowski both judgements and ideas direct themselves to objects which are taken to be inde­ pendent of our thinking. He agrees with Mill that a person who uses the word 'sun' is thinking primarily of a mind-independent physical thing, and not of an idea or of any part of an idea which is passing through his mind.3 1 2

3

Reith. Gs. Abh. II, p. 1 2 3 . Zur Lehre cvom Inhalt und Gegenstand der Vorstellungen, 1 8 94. Ibid. § 3 .

T HE D O C TR I N E O F C O N TEN T A N D O B JE C T

9

The true object of an idea or a judgement is not the mental image of that object, but the object itself. There is there­ fore in his thought no tendency to confine the mind to the narrow circle of its own creations, or to deny that we can think about realities. But he holds that in order to pass beyond ourselves in this way we have to build up within our minds an image or sign of the object to which we are going to refer; 1 we require some link ( Bindeglied) which will make it possible for an idea to refer to one definite object and no other. 2 We therefore require three terms, an idea (Vorstellung) as a mental act, an object which is presented to us by means of this idea, and a content which exists in our idea, and through which, as intermediary, the reference to the object takes place. 3 There are passages in the earlier parts of Twardowski's treatise in which the content presents itself altogether in the familiar guise of a mental image. He writes: If we compare the idea as an act with the act of painting, the content with the picture, and the obj ect with the subj ect, a land­ scape, for instance, depicted on the canvas, we shall have expressed approximately the relation between the idea as act, its content and its obj ect. For the painter the picture is a means by which he depicts the landscape ; he wishes to represent, paint, a real or imaginary landscape, and he does this inasmuch as he paints a picture. He paints a landscape, inasmuch as he prepares, paints, a picture of this landscape. The landscape is the 'primary' obj ect of his painting activity, the picture the 'secondary' obj ect. The case of an idea is similar. The thinking subj ect has an idea of some obj ect, e.g. a horse. I n having this idea, he sets a mental content before his mind. The content is an image of the horse in much the same way as the picture is an image of the landscape. If the thinker has an idea of an obj ect, he has at the same time an idea of a content which refers itself to that obj ect. The obj ect of the idea as an act is its primary obj ect; the content through which we have an idea of the obj ect is the secondary obj ect of the idea as act. 4 1

3

Ibid., P· 9 · Ibid., P · I 8 .

2

4

I bid ., p . 3 r . Ibid., pp . I 7 , I 8 .

ro

T HE D OCT R I NE O F CO N T E N T A N D O B J ECT

If we take this passage at its surface-value, it expresses a theory of knowledge which it is by no means easy to accept. Orthodox representational theories make the object of an idea something wholly immanent in that idea; if a transcendent real object is introduced at all, this is only possible if the theorist forgets that he can neither think nor speak about such a thing. But on Twardowski's theory we genuinely think about transcendent real objects, only we think of them by means of other immanent objects. There ought therefore in some way to be two objects before our minds when we think of a horse, the mental picture and the real horse. It is not easy to maintain that we can observe such a duality. No doubt there is a stream of more or less inaccurate imagery, verbal and visual, which accompanies any idea, but it is clear that we are not thinking about this, and that, if we require to think by means of such irrelevant material, it is possibly because of some mental weakness on our part. But we shall find that Twardowski is very far from holding the rather naive theory which his more metaphorical statements suggest. Twardowski gives four important reasons for distin­ guishing the content of an idea from its object, of which the first two recur in the exposition of Meinong. The first is that the content of an idea necessarily exists as an integral part of that idea, whereas the object need not exist at all. We can only think of a golden mountain by having an idea with a certain content; the content exists in the idea, whereas the object has no existence in the idea or anywhere else. 1 Twardowski discusses with great subtlety Bolzano's doctrine that ideas in whose content incompatible attri­ butes are united have no objects. Balzano had held that, when I think of a round square, I am having an idea to which no object corresponds. 2 If this view is correct it would follow that the content of my idea had contradictory 1

Twardowski, Zur Lehre, &Jc., p. 30.

2

Wissenschaftslehre, vol. i, p. 304.

T H E D O CT R I N E O F C O N T E N T A N D OBJECT

II

attributes, since there is nothing else to have them; since, however, something with contradictory attributes cannot exist, the content of my idea is also non-existent, and I am therefore not really thinking of anything, which is absurd. Twardowski deduces from this the consequence, that it is not the content of my idea which is round and square, since this content undoubtedly exists, and is therefore free from contradiction. It is only the object which exhibits a contradiction, for which reason it is excluded from the realm of existence. Our idea of a round square has there­ fore an object, as much as any other idea; this object is distinct from the content, since the latter exists, whereas the former, by virtue of its possession of incompatible properties, is incapable of existence. 1 We may note in passing the important point that an object's non-existence does not make it in the least impossible for it to be an object of thought. Many of Meinong's most characteristic doctrines spring from this consideration. The second reason given by Twardowski for distin­ guishing the content of an idea from its object lies in the fact that the content is necessarily mental, whereas the object may possess properties which no mental existent can possibly have. A golden mountain is an extended object, it is made of gold, it is greater or smaller than other mountains. None of these properties can be attributed to our ideas, or to any component of those ideas. 2 This is probably the most important and permanently valuable of the reasons which he gives; no one who has a clear idea of what a mind is, that it is made up of experiences, which cannot be divided into anything but experiences, will ever entertain the grotesque notion that a golden mountain can find a habitation in it. The two remaining reasons adduced by Twardowski are of less importance, because they involve the difficult problems of an indirect or descriptive reference to objects. 1

Zur Lehre, &Jc., p .

24.

2

Ibid., PP· 30, 3 I .

T H E D O C TR I N E O F C O N T E N T A N D O B J E C T

12

The same object can be referred to in many ways ; we can think of the same city as the birthplace of Mozart or as the city which stands on the site of ancient J uvavum ; the difference between the two ideas, which we thus form of one and the same object, must lie in the content of those ideas. 1 The weakness of this argument lies in the fact that it seems fairly clear that there is some difference even in the objects of the two ideas ; if they concern the same city, they certainly also introduce us to other objects and rela­ tions which would suffice to distinguish them. For Mozart and ancient Juvavum are as much objects as the modern city of Salzburg. The fourth reason given by Twardowski is that a single idea may have many objects ; the idea of a triangle cannot be held to have more than one content, but it has as its objects all actual and possible triangles. 2 Here again, as Twardowski himself admits, the argument is not of great worth. He comes to the conclusion, which we shall find fully developed in Meinong, that 'a triangle' in abstracto is as much a genuine object of thought as any particular triangle, though it is of course incapable of existence. 3 V

When we consider Twardowski's further statements about the content of an idea, it becomes less and less like any entity which is familiar to popular thought. An idea divides itself into two elements, an act and a content, and these form together a single mental reality ; at the same time, if we consider these elements separately, we shall see that the act is always something real, whereas the con­ tent is never real. 4 It might seem very curious that Twar­ dowski denies the reality of the content, after having affirmed its existence with great vehemence, but we must Zur Lehre, &c., p. 32. 2 Ibid., p . 34. Meinong also makes use of these last two arguments in a footnote. See Geg. hoh. Ord. Gs. Abh. I I, p. 384. 4 Zur Lehre, &c., p . 3 r . 1

3

T H E D O C TR I N E O F C O N T E N T A N D O B J E C T

13

understand that both he and Meinong use the word 'reality' in a peculiar sense. The real is that which is capable of concrete existence; a tree, a note, a state of grief, a movement, are given by Twardowski as instances of real objects. On the other hand we have objects such as lack, absence, possibility, which, in the language of Twardowski, exist but are not real, in the language of Meinong subsist but cannot exist. 1 The content of an idea is genuinely 'there', it is an element which can be distin­ guished in the idea, but it has not the same crass reality as the idea or its object. In this doctrine Twardowski's theory differs from Meinong's, for Meinong holds that the act­ element and the content-element in an idea exist in pre­ cisely the same way. 2 When we come to Twardowski's account of the rela­ tion between the content and the object of an act, the last vestiges of the image-theory disappear. That an idea con­ tains within itself an image of the obj ect, to which it has some sort of photographic resemblance, is rejected as the fancy of a primitive psychology. 3 This conclusion is im­ plied in what has been already said : the content of the idea of a golden mountain was, as we saw, something mental, and, being mental, it could not have any size or be made of any sort of physical material. In these circum­ stances it is quite clear that there can be no question of photographic resemblance between content and object. Similarly in the case of the idea of the round square, the object is round and square, and is incapable of existence because it is infected with contradiction; the content, on the other hand, is neither round nor square, since it exists, and nothing can exist which involves a contradiction. Here again the possibility of a resemblance is excluded. It might be imagined that the content-theory of Twar­ dowski is a modern version of the Aristotelian theory of 1 2

Ibid., p. 3 6 ; 0. A., P · 73. Geg. hoh. Ord. Gs. Abh. I I, pp. 3 82, 3 84.

3

Zur Lehre, &c., p. 6 7.

r4

T H E D O CT R I N E O F C O N T E N T A N D O B J ECT

knowledge, according to which the knowing mind receives in itself the pure form, but not the matter of the objects which it knows. Thus it might be held that when I think of an apple, I do so by receiving into my mind its con­ stitutive properties, a certain shape, colour, size, and so on, bound together in a certain characteristic way. The content of my idea would then be made up of those properties (Me rkmale) by means of which I refer to the physical thing, which, as a concrete physical existent, can of course not enter into my mind. Such a conception of content would recommend itself to many philosophers, and Twardowski points to Sigwart and others who have adopted it. 1 But it requires very little thought to perceive that such a view is open to precisely the same objections as the image-theory which was rejected above. The size, shape, and colour of an apple are quite as much objects of our idea as the apple itself; the apple is the whole object of our idea, the properties mentioned are parts of that object. 2 That, when we think of an apple, they are specially emphasized in comparison with less known properties, cannot take them out of the realm of objects into the realm of ideas. Moreover, it is as impossible for such properties as extension, shape, and colour to characterize a mind, or any part of a mind, as it is for the concrete apple to be absorbed into an idea. Twardowski considers therefore that no part of an idea is made up of any properties of an object, but only of ideas of those properties; an idea cannot resolve itself into any­ thing that is not itself mental. He remarks with regard to the idea of gold: I t is not the content of the idea of gold, but gold itself, i .e. the obj ect of the idea, to which the determinations heavy, yellow, shining, metallic, &c., belong. These determinations are given by means of the idea of gold, but the sum of these determinations does not constitute the content of the idea of gold. Rather must we say I

Zur Lehre, &Jc., p. 4 1 .

2

Ibid ., p . 42 .

T HE D O C TR I N E O F C O N T E N T A N D O B J E C T

15

that the latter is made up out of j ust as many ( or more) parts as the determinations which are distinguished in the gold ; these deter­ minations are presented by means of the various parts of the idea, i .e. also by means of ideas. The content of the idea of gold is therefore not made up out of the sum of its properties, but out of the sum of the ideas of its properties. 1

The whole conception of a mental picture therefore falls away : the content of the idea of gold is not yellow or shining, but we can distinguish in it other ideas which present these properties. In the case of an idea of a simple object, or of some object which we apprehend as simple, the relation between content and object is primary, origi­ nal, and irreducible, and it is not a relation of resemblance.2 The content of the idea of a certain shade of red is nothing but that content which, in conjunction with a certain act, constitutes the idea of that shade of red. Whenever we have an idea with that content, that shade of red will be given to us as object; nothing more can be said. The only case in which there is any shadow of resem­ blance between content and object is where the object is thought of as complex and as built up out of many con­ stituent objects. In this case there will be a resemblance between content and object inasmuch as they will exhibit a similar structure; 3 the elements of the content will be totally unlike the elements given in the object, but to each element in the content an element in the object will corre­ spond, and each set of elements will enter into its own systematic unity. Thus if I have an idea of an apple, the content of my idea will be composed of contents which are correlated with the various moments distinguished in the apple; these partial contents will be fused in the unity of the whole content, just as the moments of the apple are fused in the unity of the object. This structural resem­ blance will therefore be like the purely structural resem­ blance which is believed to hold between our sense-data 1

Ibid., p. 44 .

z Ibid., pp. 68, 8 1 .

3

Ibid., PP· 6 9, 8 I .

16

T H E D O C T R I N E OF C O N T E N T AN D O B J E C T

and their physical causes ; common sense is so little in­ clined to regard such similarities as genuine resemblances that it would probably say that content and object, like sense-datum and physical stimulus, were in all cases 'wholly dissimilar'. We may notice in passing one curious notion of Twar­ dowski's. He thinks that the content of an idea will exhibit a higher order of complexity than the obj ect, since there will be material elements of the content which correspond to formal or structural components of the object. In my idea of the apple are not only ideas of its colour, shape, &c., but also of the peculiar unity which these aspects constitute. The structure of the elements of the content cannot present this structural unity of the obj ect because a relation between contents is not itself a content. 1 The mental synthesis of my ideas, not being an idea, cannot set before me the non-mental synthesis of the moments of the object. A diagram may make this complicated situation a little clearer. I is an idea of an object 0, which is made up out of A and B bound together in a characteristic manner R. The content of

--r -------0 -oO I

A'

R'

B' Content

A

R

B

□-----□

Obj ect

I, has as parts A', R', and B' which present A, R, and B respectively. The manner, however, in which A', R', and B' are bound together, is not a material constituent of the

content, but a purely formal one, and does not therefore present anything. This curious doctrine leads inevitably to the most hopeless difficulties ; to have an idea of 0, it is surely not enough to have ideas merely of A, R, and B 1

Zur Lehre, &c., p. 93 .

THE D O CT R I N E O F CONTENT A N D OBJECT

17

as three separate entities, one must also have an idea of them as bound together in the unity of the object. On Twardowski's theory, it is hard to see how we ever cognize more than a set of independent moments, which cannot possibly constitute a unitary object. We shall see that Meinong suffers from the same atomistic prepossessions; he devotes a whole chapter in Ober Annahmen to over­ coming, by heroic means, the difficulties involved in our apprehension of complex objects. 1 VI

We have sketched Twardowski's theory to a sufficient extent to be able to understand Meinong's brief and rather puzzling introduction of the notion of content in his treatise entitled Ober Gegenstande hoherer Ordnung. 2 In reality, Twardowski merits a much more thorough study, but, as the work of Meinong has carried investigations much further and rectified many of Twardowski's mis­ takes, it will be better to leave aside a detailed criticism of his views. We may sum up Twardowski's doctrine in the following way: ideas are directed to objects which lie beyond those ideas, but we cannot understand such self­ transcendence unless we discover, in the idea itself, an element which corresponds to the object and to each of its presented nuances. This element is mental, its corre­ spondence with the object is not one of resemblance in its individual constituents, but involves, in the case of com­ plex ideas, a resemblance of structure. Lest it should be imagined that a content is something entirely hypothetical, we may note that Twardowski holds that it can become the object of an introspective idea which is directed to the idea of which it is the content. 3 But whether it is easy to distinguish the content from the act-component of the 1 3

0. A., ch. viii.

2

Zur Lehre, &c., p. 6 3 .

824187

C

1 8 99. Gs. Abh. I I, p . 377 et se q .

T H E D OCTR I N E O F CO N T E N T A N D O B J ECT

r8

idea, or whether we merely have an idea of an idea as a whole, in which we believe the content to be an element, is far from clear. We shall find similar obscurities in Meinong's account of our knowledge of contents. Like Twardowski, Meinong bases his distinction be­ tween the content and the object of an idea in the first place on the fact that we are able to think of objects which do not exist. Of these non-existent obj ects he distinguishes three types : we have such objects as golden mountains, whose non-existence is merely a matter of brute empirical fact ; then we have such objects as round squares, which cannot exist because they involve a contradiction; finally we have such entities as the equality between three and three, or the diversity between red and green-even in ordinary language we recognize that such things may subsist (bestehen), but not exist 'like a house or a tree' . 1 According to Meinong existence and subsistence are two forms of being whose distinctness is indescribable but immediately evident: 2 we shall discuss this doctrine later, but may observe here that it is in accord with ordinary notions. Equality and diversity have undoubtedly some sort of being, but they do not fulfil the same function in the universe as the existents which are its elements. Hence it is not unreasonable to hold that they are, in a sense quite different from the round square, incapable of existence. We are able to think of objects which do not exist, and this proves that these objects, at least, are not literally contained in the ideas which refer to them. If they were, they would necessarily exist, as much as the idea of which they formed a part. If the round square or the relation of diversity is really the content of my idea of either of these objects, then it must share in the existence of the complex which it helps to constitute. Hence a suit­ able mental state can give existence to an absurdity or to 1

Geg. hoh. Ord. Gs. Abh. I I, p. 3 8 2 .

2

0. A., p . 7 3 .

T H E D O C TR I N E O F C O N T E N T A N D O B J E C T

19

something purely ideal; this i s nonsense, and we must draw the conclusion that these objects are not parts of the corresponding ideas . Once this has been established in the case of ideas of non-existent objects, Meinong's conclusion would natur­ ally extend itself to ideas of objects which really exist. Whatever be the correct analysis of the experience which takes place when we are said to be thinking about some­ thing, it is clear that it is in every case qualitatively the same, whether the object to which it is directed actually exists or not. We should gain no deeper insight into a certain idea, e .g. that of a material object, if we knew that there were material objects, or that there were no such objects. The attitudes which we have to real entities need not differ in any respect from those which we have to wholly fanciful creations. Even the intuitive element, which might seem to be peculiar to ideas of real objects, can be present in its full vividness in hallucinations. To assume that the real object X is a constituent of the idea although an idea qualitatively indistinguishable from could exist even if X had no existence whatever, is an astonishingly futile piece of thinking. It seems therefore likely, if we adopt the point of view of this argument, that an object is not a part of the idea which presents it. Similar arguments apply to those cases in which we have ideas about objects which belong to the past or the future. The difficulty is not, on Meinong's view, that such objects are not yet in existence, or have lost the existence which they once enjoyed, and are therefore un­ able to be a part of something that exists at the present moment. If this were the argument, it would only be a special case of the previous one. But Meinong holds that there is a purified concept of existence in which all refer­ ence to the position of a judging observer, and hence all distinctions of past, present, and future, fall away; from this point of view last year's melted snows, or the snows

r,

r

2 :>

T HE D OCTR I N E O F CO N TEN T A N D O B JECT

that will fall next year have a timeless existence (or per­ sistence, to use Meinong's technical term), at a certain position in the time-series, which is for them present. 1 But, even if this is so, we cannot hold that last year's melted snow, which stands at one point in the time-series, is also a part of my idea which stands at another point in the time-series. If such multiple ingression of an object into the history of the universe were possible, there would be no sense in locating an object, of which I am thinking now, at some date before the date of my idea: it would be as much present as my idea, and this is plainly not the case. The next reason for holding that objects of ideas are not parts of them lies in the fact that we can have ideas of physical objects and their properties, while no one would contend that extended and ponderous objects, and pro­ perties which presuppose such objects, can be literally parts of something which consists of acts and experiences. 2 If we had ideas of mental states, which Meinong, like Berkeley, was afterwards led to deny, it might be con­ ceivable that they could be a part of the ideas which were directed to them, provided they occurred at the same time as those ideas. In the case of physical objects, such a con­ tention would be nonsense. The mountain-ranges of Asia cannot be an element in any experience: vast masses of matter cannot possibly be ' lived through'. They cannot enter into the stream of private, inalienable happenings which constitutes a mind. The same argument might have been applied to purely ideal objects such as the diversity and equality which were mentioned above; we may think of such objects as often as we like, and such thought may alter the course or the ' colour' of our experiences, but we cannot ' live through' them in the same way that we 'live through' a judgement, a decision, or a state of grief. 1 See 0. A., p. 77.

2

Geg. ho"h . Ord. Gs. Abh. II, p. 384.

T H E D O CT R I N E O F C O N T E N T A N D O BJECT

:u

Meinong considers the attempt to get rid of the diffi­ culties involved in thinking of what is non-existent by saying that golden mountains and round squares exist 'in my thought'. He thinks it clear that this is a subterfuge. The predicate 'existence' has a totally different meaning when it is predicated of things simpliciter, and when it forms a part of the expression 'existence in thought'. For we do not mean to affirm that a golden mountain really exists, still less do we wish to affirm the even stranger pro­ position that a mountain, which is an extended object, is a part of my thought. It is clear that the misleading sentence ' A golden mountain exists in thought' is really equivalent to the proposition: ' There exists a thought about a golden mountain.' Meinong wishes us to distinguish clearly between the primary sense of existence, and all those cases in which we say that there is (es gibt) such and such an entity, with­ out being willing to say that it exists. Like Aristotle he believes in the fundamental ambiguity of the word 'being'. In the next chapter we shall deal with the widest sense of the words es gibt; here we may consider a rather limited sense. If I say ' There are many unpleasant characters in this novel', I may merely mean that there exist in myself and others certain mental processes in which those charac­ ters are contemplated with pain, not that such characters have existence in the ordinary sense. If we must attribute some sort of being to them we may say that they have pseudo-existence; this means that there exists, in the pri­ mary sense, a state of mind in which they are presented. 1 The gulf between existence and pseudo-existence becomes very clear when we consider that any sort of entity may have pseudo-existence: not only the round square, but also purely ideal entities such as relations, numbers, or facts, may be the obj ects of appropriate states of mind. 2 Whether Meinong uses the word 'pseudo-existence' merely as a 1

Ibid . , P · 3 8 3 ; Erfgl., p. 5 6 .

2

Ibid ., p. 62 ; Mag., p. 40.

22

T H E D O C T R I N E O F CO N T E N T A N D O B J ECT

verbal equivalent for ' being present to thought', or whether he means to indicate by it a genuine variety of being, is not entirely clear. Quite possibly the latter is his intention, for even the pseudo-existent comes into relation with things that genuinely exist, and may, by such indirect partici pa­ tion in the actual world, acquire some simulacrum of being. VI I

We have given a number of cases in which it is im pos­ sible that the object of an idea should be a part of it. On this fact Meinong bases his doctrine of contents; if the object is not a part of an idea, there must be something in that idea which gives it direction to a given object. Meinong writes: Whether I have an idea of a church-steeple or a mountain-peak, a feeling or a desire, a relation of diversity or causality or any other thing whatsoever, I am in every case having an idea. In spite, therefore, of the unlimited variety of their obj ects, all these mental processes manifest a common feature, which makes them ideas, and this is the act of having an idea. On the other hand ideas, in so far as they are ideas of distinct obj ects, cannot be altogether alike ; however we may conceive the relation of the idea to its obj ect, diversity of obj ect must in some way go back to diversity of idea. That element, therefore, in which ideas of distinct obj ects differ, in spite of their agreement in the act, may be properly called the content of the idea. This exists, is therefore real and present, and is of course mental, even when the obj ect presented by its aid does not exist, is not present, and is not mental. 1

The argument as it is here stated has an unconvincing, a priori sound, but it rests on a very important considera­ tion. The object has been shown in many cases to lie out­ side and to be indifferent to the idea which is directed to it. It might now occur to us to suppose that the idea is purely a mental act, in which nothing corresponding to the object is discernible, and that this act stands in some T

Geg. hoh. Ord. Gs. Abh. I I, p. 3 84.

T HE D OCTR I N E O F CO N T E N T A N D O B J ECT

23

relation to its object, a relation which may be necessary but which nevertheless falls outside of the act. Theories of this crude type have often been held by realist philoso­ phers. But they ignore one important and extraordinary fact, that our introspection reveals to us not only that we have ideas but also that they are ideas of such and such objects. 1 If I contemplate my ideas, I do not discover in my mind a set of naked acts, qualitatively indistinguishable from each other. The notion that a mind must be made up of such entities has caused many philosophers to cast doubt on the existence of mental acts, or the possibility of intro­ spection. But what I do perceive are ideas of church­ towers or mountains or equality. In fact, to isolate the act-element in any experience always requires a certain amount of abstraction: in the case of strong emotions or powerful desires such an abstraction comes easily and naturally, whereas it is almost impossible to isolate the act-element in the 'seeing' of a church-tower. 2 The ordi­ nary man is quite sure that he is going through the whole experience denoted by these words, but he would find it hard to fix his attention on the mere 'seeing'. We do not look into our minds and perceive a certain naked idea, then look outwards and see a certain object, and then finally perceive that they are related in a certain way, such that the one is the idea of the other. If this were our procedure we might quite conceivably connect an idea with the wrong object. If we were 'living through' two ideas .A and B of two objects A' and B', there would on the theory be nothing in the ideas as pure acts to distin­ guish them from each other; hence if we did not carefully observe their relation to the objects A and B, we might easily imagine that A was the idea of B' and B of A' . That such errors in connecting ideas with objects not only do 1

2

See ibid., p. 404; E rfgl., P· 5 7• Ibid., p . 6 r; Geg. ho"h. Ord. Gs. Abh. I I, p. 404.

24

T HE D OCTR I NE O F CO N TEN T AN D O B JECT

not occur, but are a priori unthinkable, p roves that the act-element in an idea is only an abstract moment, and that every idea, as a concrete experience, bears on its face a reference to a certain object, which reference is immedi­ ately accessible to introspection. We do not, in ordinary introspection, see ideas and obj ects, which we then proceed to relate; we simply see ideas of obj ects, beside which we may set the obj ect if we wish. To consider an analogous case, it would be an absurd account of our perception of a red obj ect to say that we saw the obj ect as a naked thing together with the characteristic red, and perceived also that the latter inhered in the former; a red thing is a concrete whole, in which no naked substrate can be dis­ cerned. It is just as absurd to attempt to see an idea per se without its reference to an obj ect, as to see a red thing without its redness. It is also very difficult to understand how, if the mind consisted merely of acts, we could have any particular obj ect before our minds. The idea of the Himalayas and the idea of j ealousy would be nothing but transparent intentions; their obj ects would lie outside of them, and nothing in our experience would tell us to what they were directed. We should have as little conception of the obj ects of our ideas as we have of the cerebral changes that accompany them. It is impossible to see how any philo­ sophy which builds the mind out of mere acts can avoid this consequence. But we find that even when A and B are obj ects which cannot by their nature be 'lived through', the reference to A is nevertheless a different experience from the reference to B ; es ist uns anders zu Mute, as Meinong says : 1 we can only translate this by saying that 'it feels different' . The reference to one definite obj ect and no other is therefore one of the most intimate properties of an idea ; if it is not the whole nature of the idea, it is certainly involved in that nature. The surprising fact that, I

Erfgl., P · 59•

T HE D OCTR I N E O F CO N TEN T AN D O B JECT

25

in the very bosom of an idea, we can find a clear reference to something which is not in any way a part of that idea, but is entirely beyond and indifferent to it, demands some elucidation, and this Meinong provides when he holds that an idea must have a content as well as an object. Let us dwell for the moment on the other element in an idea which Meinong has called the 'act'. It would be natural to apply the word 'act' to a whole experience, the act of judging that there is snow, for instance, or of desir­ ing the end of a journey. Meinong, however, means by the 'act' in an experience that element which exhibits a variability independent of the reference to a given object. So defined, the 'act' involved in an idea is the function of presenting which is common to all ideas, the 'act' involved in a judgement is the function of judging which is present in all judgements, the 'act' involved in a desire is a function present in all desires. Meinong conceives of an 'act' as a qualitative moment in a complete experience, the way in which the mind directs itself to an object; it is not itself a complete experience. It would perhaps have been better if, like Husserl, he had spoken of an 'act-quality' instead of an 'act'; 1 we shall, to avoid confusion, speak of his 'act' as the act-element in a mental state. The act-element may of course vary while the content remains unaltered. Meinong gives as an instance of this the way in which the hearing of a tone passes over into an imaginative reproduction: the 'act' alters, the experience changes in quality, but the same object is presented.2 This is even clearer if we consider our attitudes to those objects which are usually called propositions; we may believe them, or we may merely assume them, or we may surmise their truth with varying degrees of certainty. All such variations in the quality of our experiences have nothing to do with the object which is in every case set before the 1 2

See Logische Untersuchungen, vol. ii, part i, p. 4 1 3 . Em. Preis., p . 5 7 .

26

T HE D OCT R I NE O F CO N TEN T A N D O B JECT

mind. 1 The act-element is therefore that moment in our experience which varies in a wholly subj ective way, and which does not by itself present anything. It will be seen that Meinong's 'act' has nothing to do with activity as opposed to passivity; Meinong believes that ideas and feelings are passive experiences, as opposed to judgements, assumptions, and desires which involve an active 'doing' on the part of the subject, but this does not prevent him from speaking of the act-element in the former as much as in the latter. 2 VI I I

We now come up against two difficult questions. Are contents things which are genuinely given to us in inner experience, or are they merely hypothetical entities, like the microscopic objects postulated by physical science as the causes of certain appearances ? Can we 'live through' contents in full intuitive consciousness of the fact that they are there, or do we merely infer their existence as the result of arguments like the ones which we have given above ? The second question we must ask is whether the concep­ tion of a content, or the existence of a content, if that can be proved, really makes the self-transcendence of the mental more intelligible. As regards the first question, if we were to identify con­ tents with the fragmentary sense-data and images which illustrate our referential experiences, their existence and accessibility to introspection would be beyond all doubt. Professor Stout, in his 'Fundamental Points in the Theory of Knowledge', interprets Inhalt in this way. He writes: Sensuous experiences fulfil a peculiar function i n our mental life which requires to be explicitly recognized in our terminology. They constitute a link between mental acts and obj ects which are not themselves present contents of immediate experience. Thus sense 1

Em. Preis., p. 6 2 .

2

Ibid., p. 56.

T HE D OCT R I N E O F C O N TEN T A N D O B JECT

27

impressions and images are means by which we perceive o r imagine material things and their qualities, states and processes. A special term is required to designate contents of immediate experience which thus fulfil, or are capable of fulfi l ling, the function of presenting or introducing obj ects which are not themselves contents of imme­ diate experience. The term selected for this purpose by the group of writers I am dealing with (Meinong, Husserl, &c.) is 'lnhalt'. 1

There can be no doubt that the conception to which Professor Stout refers is a very important one, and it is clear that the entities of which he speaks are open to our observation. Unfortunately, however, they do not corre­ spond to Meinong's 'contents'. For sense-data and images are objects quite as much as the remoter objects which they i ntroduce ; they are not the sort of thing a person could 'live through' . 2 In pointing to them we only shift the problem of the reference to objects a stage farther back; our objects are simpler, and their presentation in­ volves intuition: that is all. They still have extension and shape, in common with physical objects, as well as other properties, such as blueness, warmth, heaviness, which no state of mind can possibly have. 3 In Ober Annahmen Meinong discusses the whole ques­ tion of what he calls the adeq uateness of states of mind. An idea is adequate to a given entity when it presents that entity as it really is. Meinong asks whether such adequate­ ness involves any resemblance between idea and object, and comes to the conclusion that this is quite impossible in many cases. To have an idea of a square table, I do not require to have an idea which is square in any part, but merely an idea of a square thing. The content of my idea, being mental, can be neither round, nor oval, nor square.4 This excludes the notion that contents are illustrative sense-data and images, since these can undoubtedly be I 2 J

Studies in Philosophy and Psychology, p. 355. For the whole position of sense-data see Ste/lung, § 2. Geg. hb"h. Ord. Gs. Abh. I I, p. 384.

4

0. A., p. 263.

28

T HE D OCTR I NE O F CO N TEN T AN D O B JECT

round, oval, or square. It is possible that Meinong has not dealt sufficiently with the part played by sense-data and images in referential thought; even his conception of the Hilfsgegenstand, or auxiliary object, does not tally with that of Professor Stout. 1 But the whole question is not of fundamental importance; if a philosopher can explain how we can see coloured patches or feel square surfaces, there is no great difficulty in explaining how we can think of the Taj Mahal or the Milky Way. We must therefore look for our contents in some other direction. IX

Meinong admits that the contents of our ideas are not easy to discern. One of the reasons for this difficulty is that we have no suitable words to describe them by. Lan­ guage serves a double function: it exp resses our states of mind and it means or refers to the objects of those states of mind.2 A man who utters the words 'red' or 'blue' gives expression to a peculiar inner experience through which he is living, but he is not meaning or referring to this experience. He is talking about certain properties which can only be manifested in extended objects. If he wishes to compare red and blue, he can only do so by experiencing the transition from one content to another, but he will not be thinking about this transition, but about the difference between red and blue. The fact that, when we are living through contents, our attention is usually not directed to them at all, has as a consequence our total lack of words to describe their qualitative nuances. We have accustomed ourselves to carry their peculiarities over into the object, so that, if we are to refer to them at all, we must speak of 'the content of the idea of red' or 'the content of the idea of blue'. 3 As a result of this absorption in objects, we have 1 See Mag., p. 1 96. 2

3

Geg. hb"h. Ord. Gs. Abh. II, p. 3 8 5 ; 0. A., p. 24 et seq. Geg. hb"h. Ord. Gs. Abh. II, p. 38 5 ; Eifgl. , p. 5 9.

T H E D O C TR I N E O F C O N T E N T A N D O B J E C T

29

to describe our experiences i n a roundabout way by means of their objects, and the most indubitable data of our inner life assume the appearance of hypothetical entities. .,. \\ e must admit that Meinong is right, and that lan­ guage fails us almost entirely when we attempt to describe the quality of inner experiences . The Lange-James theory of emotion, in which a unitary state of mind is resolved in to a set of sense-data, and other similar theories, are a consequence of this difficulty. They all ignore the impor­ tant fact that, at the moment when our heart is beating hard or our hair is rising, our experience is something p urely q ualitative in which localized sense-data have melted away ; these can only present themselves when the flood of emotion has subsided, and we can then say, not that they were present previously, but that if attention to objects were compatible with strong feeling, our experiences would have presented us with certain sense-data which they did not, in fact, present. If our vocabulary for qualities of mind were as precise as our vocabulary for colours, we should not be exposed to the fallacy. We may note here how, in the case of smells, where our terminology is inadequate, we also speak of ' the smell of sweet-peas' or ' the smell of tar', describing them by means of their relation to other objects instead of treating them as objects in their own right. Meinong maintains, as we saw, that our inner percep­ tion revealed to us not only the act-element in our mental states but also their objects. I perceive that I am hearing a note, or that I am judging that it is four o'clock. It is very hard to see why we should perceive the object and not the content. Meinong admits that, owing to the diffi­ culties of attending to our mental state, contents retreat behind the objects which they present ; we perceive that we are thinking of such and such an object, but the con­ tent is obscured by the pseudo-existent object. It is much easier to hold before our mind a red square, or a green

30

T H E D O C TRI N E O F C O N T E N T A N D O B J E C T

circle, than to maintain unchanged before the introspec­ tive judgement the elusive modification of our inner life by which we are aware of these obj ects . An obj ect which is used in introspection as a substitute for the correspond­ ing content is called by Meinong a pseudo-object 1 because we are not treating it as an obj ect in its own right, but as a mere means towards the apprehension of something else. The relation between content and obj ect shows a curious reciprocity ; obj ects cannot enter into the mind but are made accessible to thought by means of contents ; con­ tents, though they are in the mind, are difficult to appre­ hend directly, and so are generally grasped by means of obj ects. Nevertheless, if the theory is not to be a mere construc­ tion, there must be some occasions on which contents are directly observable, even if they are elusive. Can we find anything in our inner life which corresponds to M einong's contents ? It seems to be quite clear that we can . If I com­ pare the experiences of looking at a red square and a green circle, there can be no doubt that these experiences differ in a wholly indescribable qualitative way . It 'feels different' to be considering the obj ect given in each case ; the 'colour' of the experience is different. This 'feeling' is not an emotional feeling, whose connexion with the obj ects pre­ sented would be rather fortuitous . We seem to see some­ thing necessary in the way in which we are affected by the smooth roundness of a circle and by the straight lines and pointed corners of the square. The green similarly 'feels' different from the red, and by this we do not mean that the one is restful while the other is stimulating, but some­ thing far more elementary, which such qualities presup­ pose. It is not that our experiences are red or green, smooth or with edges and corners, round or straight, but such language would be a good way of indicating the intimacy and necessity of the relation between them and their obj ects . 1

Etfgl., P · 57·

T H E D O CT R I N E OF C O N T E N T A N D OBJECT

31

I t seems to be a self-evident proposition that, if any one had that curious experience which is involved in look­ ing at a red square, he would, if he chose to refer his experience to any object, necessarily be aware of a red square and nothing else. Visual experiences are, however, rather unfavourable to the presentation of contents; we can observe them much more readily in such experiences as auditory sensations, or sensations of movement, or pains. If we 'live through' an auditory sensation or a sensation of movement, we need not hear a sound or perceive a movement. It is easy not to pass beyond the experience as a peculiar change in our inner life. The contents of our experience may then be revealed to our introspection as purely mental qualities. It is a sheer fallacy to suppose that there are full-blown sense-objects given in such experiences, to which we hap­ pen not to attend. 1 It would be like thinking that we could see equally well with all parts of our retina because we can turn our fovea in the direction of any object we wish to observe. It is only when, instead of merely living through the passive sensuous experiences in question, we pass to an active apprehension of something outside ourselves, that the qualitative peculiarities of the auditory sensation will present to us a tone of a certain pitch, loudness, and timbre, the sensation of movement will present an interest­ ing phenomenon located in some part of the body. It is because certain experiences, e.g. headaches, are not gener­ ally used to apprehend objects, that we are in doubt whether to place them among objects or experiences. The truth is that there are two distinct headaches, an un­ extended, unpleasant headache through which we can live, and a headache which consists in the intermittent pervasion of certain parts of our head by a certain peculiar quality; this second headache is as much an object as a flash of lightning. 1

See 0. A., p.

236.

32

T H E D O C T R I N E O F C O N T E N T A N D O B J E CT

It seems then to be possible, if we resolutely banish any tendency to think of a non-mental object, even a sense­ object (which psychologists examine under the absurd impression that they are introspecting), to discover in any experience a set of simple subjective phases which under­ lie all ideas and judgements.To every nuance of quality, every configuration, every relation in the outer world, correspond indefinable variations of our state of mind.In some spheres, such as the visual, where the trend towards the object is dominant, they are not easy to seize; in others, such as the sphere of organic sensations, they offer them­ selves readily to our inspection.There seems little doubt that, if no such mental variations existed, none of the higher intellectual references would be possible.1 X

That contents are in some cases perfectly accessible to observation, and that their relation to objects is intelligible, can be shown by considering a few other instances given by Meinong.We choose them from fields that are widely removed from that of simple sensation.We may first con­ sider the property of being a fact, which distinguishes all facts from all fictions. That there is an integer between 3 and 5 is a fact, that there is one between 3 and 4 is not a fact, and this is so whether these circumstances are con­ ceived by any one or not.We can consider such circum­ stances without in any way deciding whether they are facts or not; we can also, in certain experiences, see that the first is a fact and the second not. By what moment in our experience is this purely objective element, which we may call 'factuality' (to translate Meinong's Tatsachlichkeit), 1 It is interesting to inquire whether the distinction between the content and the object of an idea corresponds to the distinction between the objective and the formal essence of an entity which we find in Descartes and Spinoza. Spinoza, in the Tractatus de Intellectus Emendatione, § 3 5, identifies essentia objecti'Va with the modus quo sentimus essentiam formalem; the content would certainly be des­ cribed as the mode in which we 'felt' the object.

THE D O CTRI N E OF CONTENT A N D O BJ ECT

33

apprehended ? Meinong points to a quality which he cal ls evidence (Evidenz). A j udgement unaccompanied by this inward evidence may as easily get hold of something which is not a fact as a fact, whereas a j udgement that has evidence is the adequate mode of apprehending that quality in facts which makes them facts. 1 It cannot be doubted that in this case Meinong has succeeded in pointing to two things, one mental and one non-mental, between which we can nevertheless see a necessary connexion. The evidence which accompanies certain j udgements, e.g. the j udgements passed on our own contemporary mental states, or on certain ideal ob­ j ects, such as numbers, is clearly a property of experiences, not of their obj ects. To describe this property is very difficult; one may regard it, rather unsatisfactorily, as a sort of compulsion which certain circumstances exercise upon our j udgements, or we may say that something leuchtet uns ein, 'bursts on us with a flood of light'. In whatever the essence of evidence may lie, it is clearly something which we live through, and which is ultimately the only testimony by which a fact can show us that it is a fact. It is the factual hardness of the circumstances which we are considering, which appears in our experience as the penetrative power of evidence. Our second example is chosen from our consciousness of the experiences of other minds, or of our own past states of mind. According to Meinong this can only take place in so far as we live through imaginary experiences, which resemble the experiences we are thinking of. 2 I can only present to my thought some belief which I held or which other people hold, by assuming for the time being the proposition believed; I can only think of feelings and desires of e.g. characters in a novel, by having similar feelings in which that peculiar element which we call 1 2

0. A., P· 88. Eifgl., p . 7 6 ; Mog., p .

824187

2 5 1. ;

Em. Preis., p . 28. See below, Ch. V I I I, v. D

34

T H E D OCT R I N E O F CO N T E N T A N D O B J ECT

seriousness, is lacking. I have to 'put myself in the place' of the subject whose states I wish to apprehend; such a putting myself in his place need not involve real belief or serious feeling. It is very probable that this is the right theory of our awareness of foreign mental states; how we could have them before our minds without in any way living through them, or anything like them, is quite un­ intelligible. Here I can clearly distinguish the content, an imaginary experience of my own, to which I do not ordinarily attend, but which I can observe if I wish to, and the object, a serious experience of some other mind, or of my own mind in the past. It is true that Meinong is unwilling to regard the state which functions as a content in this case as a genuine content, because the whole experi­ ence, inclusive of the act, is used in presentation, and not merely a portion of it. From this point of view he prefers to speak of it as a quasi-content. 1 But the example shows that we can sometimes definitely observe a subjective phase which helps to present an object different from itself. We may give a third rather dubious example from the world of values. There are certain experiences of approba­ tion or disapprobation in which an object is given to us as beautiful or ugly. That this beauty or ugliness is an object is clear when we consider that it may be seen in a thing by some one who is not in the least considering his mental states, or passing self-conscious judgements. Yet such an object is given to us by means of a peculiar sort of approbation or disapprobation, of which it is possible to be introspectively aware; these emotions must therefore function as contents. 2 We have a similar case when some one perceives that something ought to be. This 'ought' is also an object, because it can stand before us when we are not considering our states of mind at all. That the last war ought not to have taken place, or that the oceans ought 2

Em. Preis., p. 3 7.

THE D O CTR I N E O F C O NTENT A N D OBJECT

35

not to be so large and unvariegated : these circumstances seem to have the same evident truth and the same inde­ pendence of our minds as the truths of mathematics or biology. Yet nobody could possibly think of such a mind­ independent 'ought' if he had never demanded anything of the universe, if he had never had the experience of desire. An 'ought' is the appropriate object of a purified desire, and cannot be apprehended unless this desire is present in some degree. The desire therefore functions as the content by which the 'ought' is grasped, 1 and this desire is of course an observable entity. The three instances we have given indicate sufficiently that, in spite of the obscurity and haste with which Meinong treats the whole conception of content, he never­ theless supposes that contents, or entities that function like contents, are observable in certain cases. XI

The relation of content to object is, according to Meinong, an ideal relation. 2 We cannot deal fully here with the important distinction which he draws between ideal and real relations, but we may say briefly that if an ideal relation holds between two existents, it binds them into a unity, but that this unity is not a complex existent of which the two existents are constituents. 3 The similarity between two peas certainly binds them together, but the result is not a complex existent like a house or a tree. Similarity is called an ideal relation because it does not bring its terms into any real touch : the two peas 'have nothing to do with each other'. A second characteristic of ideal relations is that their subsistence is always neces­ sary.4 It is not merely an accident that this pea is like that pea, that this point is between these two points, that this I

3

Ibid., p. 42. Geg. hoh. Ord. Gs. Abh. I I, p. 3 95.

2

4

0. A., p. 26 5 . Ibid., p. 3 99 .

36

T H E D O C TRI N E OF C O N T E N T A N D O B J E C T

moment is later than that moment : that they stand i n these relationships follows with absolute necessity from the fact that they are the entities they are. Once two obj ects are given their degree of similarity is fixed, once two points or moments are picked out their relative positions can­ not be otherwise. The relation between the content of an idea and its obj ect is an ideal relation because it fulfils both these con­ ditions. It does not generate any real complex, and its subsistence between its terms is always necessary, never contingent. As regards the first point, it is clear that the obj ect known does not coalesce with the knowing act to form any complex existent. 1 It is as unaffected by my knowing it as one pea is unaffected by the existence of another similar pea. If I have an idea of the Himalayas, my idea and the Himalayas are not compounded in some monstrous symphysis of the mental and the material . The stream of my experiences and the stream of events known as the Himalayas each go their own way, but there is for the moment an ideal ade q uateness between them. The con­ tent of my idea ' fits' the Himalayas and nothing else : I must be living through the sort of content that can present a snowy range of mountains in a country called Asia. There are peculiar modifications of my mind which occur when I think of a many-sided obj ect like snow, similar modifications are necessary to apprehend the meaning of a word like 'mountainous' . All the aspects of the Hima­ layas which are actually brought to mind presuppose such inner qualitative changes ; in experiencing them all, and in experiencing their synthesis by means of appropriate judgements, I am at last aware of those snowy ranges which are so utterly unlike anything that I can ever live through. In making my ideas of the Himalayas adequate I receive no assistance from the Himalayas ; their existence is irrelevant to the whole process. I have, by my own I 0. A., p . 266 .

T H E D OCTR I N E O F CO N T E N T A N D O B J ECT

37

activity, constructed a reference which points i n one un­ ambiguous direction, and in that direction the Himalayas happen to lie. In all this there is no real touch, much less a mystical fusion of knower and known. If the relation of adequateness between a certain content and a certain object is an ideal relation, it must also subsist in the necessary, a p riori, intelligible way in which a rela­ tion like similarity or distance subsists. Meinong does not think that there is only one content which is adequate to one object and vice versa. 1 A single content, according as it is used, will enable us to refer either to a certain pro­ perty or to a thing which has that property. 2 We may also refer to one object by means of different contents according as we think of it as having this property or that. 3 But all such complexities are subsidiary : given a content which is correlated to redness, it may enable us to think of all red things, but it will not assist us to think of things that are blue or green. We must be able to see, by a mere examination of the content, that redness and red things are its appropriate objects; if we could not do this, the adequateness would be a contingent, and therefore not an ideal relation. These considerations dispose of the view that a content is merely a hypothetical entity invented to account for our reference to definite objects. Contents must be observable entities, if we are to see that a certain relation is grounded in their natures. XI I

It throws a certain amount of light on the theory of content and object to compare it with Aristotle's picture of sensation. The sense-organ in Aristotle receives the forms of the objects perceived without their matter, just as wax receives the form of a ring without its matter. The content, as a purely mental modification, is vaguely 2 I bid . , p . 2 76. 3 Mag., p. 1 8 7. I Ibid., p. 2 3 7 .

38

T HE D OCTR I NE O F CO N TEN T AN D O B JECT

analogous to the form left in the wax, the object to the ring which fits that form. We may notice that, in this case too, there is no exact similarity between the shape of the ring and the shape left in the wax ; where the former is convex the latter is concave, and vice versa. Spatial shapes fit each other, as content and object are adequate to each other, without needing to be similar. We might further imagine the wax capable of looking out of itself and making judge­ ments, when the shape in it will permit it to think of one object only, namely the ring which fits it. This will re­ semble the way in which a content, by its presence in the mind, enables it to 'aim' at one object and no other. The analogy is also apt in that the wax may retain the impres­ sion left by the ring, or might, by some internal activity, assume a shape that would fit a ring, although no such ring existed. In this way it would resemble a mind which thinks of an object which has ceased to exist or which has never existed. The comparison is of course only a pictorial analogy; there is infinitely more resemblance between the shape of the ring and the shape of the impression that it leaves, than there is between, say, a physical object and a mental content. And it would be wrong to imagine that objects are capable of any direct impact or impression on the mind. But the analogy has the advantage of showing up the whole nature of the theory, as well as one of its weak­ nesses, or rather one point which it fails to explain. This point concerns the fundamental nature of intentionality. In the case of the wax, we cannot see how it can pass beyond the form which it receives to the form of a totally different entity which fits the impression in itself. That one term of a relation should reveal or present another, even when the relation in question is necessary and intelli­ gible, still remains mysterious. In the case of intentionality Meinong believes that it is the activity of a judgement or assumption which makes use of the content of an idea in

T HE D OCTR I N E OF CO N TEN T AN D O B JECT

39

order to pass beyond the circle of inner experiences, and to 'hit' a certain object. The pure idea is a passive experi­ ence ; to live through its content is not to have an object actually presented to one, but only to have the power of apprehending it. 1 To have an auditory sensation or fancy is not to hear a sound, but only to have the power of hear­ ing it, should we judge: 'There is a sound', or, in the case of imagination, merely assume it. How the judgement or assumption makes use of the content to 'hit' the object is not a fact which the theory completely explains. If we found a paralytic seated on a roof, the discovery of a ladder at the side of the house would give us some assistance towards understanding his presence there, but it would hardly be a sufficient explanation. The same is true with regard to contents; their presence in the mind helps us to understand the way in which it 'aims' beyond itself, but fails to make this completely intelligible. With such aid we do not yet altogether see how an experience, in which an object O is neither a moment nor a part, can neverthe­ less be the experience of that object and no other. It may throw some light on the extraordinarily difficult problem of intentionality if we compare the reference to an object, which is peculiar to mental states, with other relational properties. The Philippine Islands are to the east of China; the subsistence of this relation between them endows the Philippine Islands with what is called a relational property, which we may write as the 'being-to­ the-east-of-China'. Is it possible to regard this property, as its verbal symbolization might suggest, as a complex of which the country China and the relation 'to-the-east-of' are constituents ? The grotesque absurdity of such a notion is evident. How one could make anything out of a relation and a single term it is impossible to see, still less how such a monstrous combination could be a characteristic of any­ thing. Characteristics may certainly be complex, but it is I

V. A., p.

22

5 ; Eifgl. , p. 5 8 .

40

T H E D OCTR I N E O F CO N T E N T A N D O B J ECT

incredible that any analysis of them should reveal con­ stituents which are not themselves characteristics. If A is a constituent of B, and B is a characteristic of C, then it seems evident that A must also be a characteristic of C. Consequently if China is really a constituent of the rela­ tional property 'to-the-east-of-China', then China must inhere, in all its solid immensity, in the Philippine Islands. By similar arguments one could prove that the Stock Exchange inhered in this teapot, or that mind was a pro­ perty of matter. These absurdities force us to believe that China is not a constituent of the property 'being-to-the­ east-of-China', but that this latter, in spite of its com­ plicated descriptive name, is probably a perfectly simple characteristic, or, if it is in any way complex, is only capable of analysis into characteristics. We call such pro­ perties as 'being-to-the-east-of-China' relational properties, not because they involve relations as a part of themselves, but because they are generated by the subsistence of rela­ tions, presuppose such relations, whereas other properties, such as green or blue, do not. Relational proper­ ties seem to be a wholly unique class of object, and it is extremely odd that they have never been properly studied. If we now turn to intentionality, the facts about the Philippine Islands will be helpful. If I think of X or think of r, my thinking-of-X or thinking-of-Y are mental relational properties, of which X and r are not consti­ tuents. The direction-to-X, if it permits an analysis at all, can only be analysed into such terms as direction-to-A, direction-to-B, &c., where A, B, &c. are the distinguish­ able moments of X; X itself, or its moments, cannot be discovered in it. But the oddity of this relational property is that it is not generated by any relation of adequateness between the state of mind and its object, but that this relation of adequateness presupposes the relational pro­ perty. It is not because my state of mind is adequate to X

T HE D O C T R I N E OF C O N TEN T A N D O B JECT

41

that it has the relational property of being-directed-to-X, but it is adequate to X because it has the relational pro­ perty of being-directed-to-X. Thus we see that not only are relational properties a unique class of objects, but that relational properties involving mental reference are unique among relational properties. 1 We have shown that Meinong's theory of content and object does not altogether explain away the difficulties to overcome which it was constructed. It remains, how­ ever, a most illuminating doctrine in epistemology. That, wherever any sort of object is apprehended, there are characteristic experiences correlated to that object, enables us to use experiences to throw light on objects and objects to throw light on experiences. It is the very nerve of the type of research which Meinong pursues. We shall, in the following chapters, which are for the most part con­ cerned with objects, point to many interesting cases where it is used. 1 These suggestions owe a great deal to Brentano's article ' Die psychische Beziehung im Unterschied von der Relation im eigentlichen Sinne.' See Psycho­ logie, Appendix i, 1 9 1 r edition .

II THE PURE OBJECT AND ITS INDIFFERENCE TO BEING

T

I

the realm of objects is far wider than the realm of existents is a paradoxical thesis, whose develop­ ment and application is one of Meinong's most interesting contributions to philosophy. The con­ siderations which led Meinong to adopt this view are to be found in the treatise entitled On the Theory of Objects. 1 Brentano had laid it down that all mental states are directed to an object, and that such direction is ultimately the distinguishing feature of the mental. We have now, says Meinong, to discover some science which will treat of all the differences and varieties of the object as such, without regard to the nature of our apprehending experiences. If we restrict ourselves to the objects of knowledge, and ignore the objects of fancy or hypothesis, it might appear that metaphysics is the science in question. According to Meinong, such a view would be incorrect, because meta­ physics has an essential limitation in its treatment of ob­ jects : it is only the science of reality, it can only include in its survey everything that exists, has existed, or will exist. Vast as such a realm of objects undoubtedly is, it is, on Meinong's view, infinitesimally small in comparison with the totality of the objects of knowledge. The world of actual existents is only a poor selection out of an in­ finitely rich and various range of possible objects ; these objects are in some obscure and very difficult sense always 'there', and form the raw material of any actual world. H AT

1

Uber Gegenstands theorie, Gs. Abh. I I, p. 48 3 .

P U RE O B JECT A N D I N D I FFEREN CE T O BEI N G 43

The study of such objects is in many cases as interesting and full of discovery as the study of actual objects; we need only think of the excursions into elliptical or para­ bolic space of the first metageometers to be assured of this. It is only our practical interests that combine to create in us what Meinong calls the prejudice in favour of the actual, 1 which makes us ignore the unreal and treat it as a mere nothing. Of one group of unreal objects we shall say nothing in this chapter. According to Meinong there are certain entities, such as relations, numbers, and facts, which un­ doubtedly go to the building up of the real world, but of which it would be absurd to say that they existed. The similarity of two pictures, or the existence of the antipodes, are not concrete reals like a knife or a state of anger. Such entities are said to subsist or to be ideal objects; we shall leave aside, for the time being, the whole theory of such objects, and pass on to consider those objects which neither subsist nor exist. It is clear that we are able to think of objects which do not exist. We can have both an intuitive and a non­ intuitive idea of a golden mountain, and we can have an idea of a round square, though it is not possible for us to work this out in intuitive imagery. The objects that are before us are undoubtedly something, they are distinct from the experiences by whose means they are given to us, they lie beyond us and are apparently indifferent to our thinking. Meinong also holds the view that there are many true statements that we can make about them. Though it is not a fact that the golden mountain or the round square exists, he thinks that it is unquestionably a fact that the golden mountain is golden and mountainous, and that the round square is both round and square. We shall see reason later on to question this assumption, on grounds different from those of Mr. Russell; for the time 1

0. Gegth. Gs. Abh. II, pp. 48 5, 486.

44

T H E P U RE O B J EC T A N D

being, we may allow it to enjoy the plausibility it un­ doubtedly possesses. That an object can have definite properties, a definite nature, although there is no such object, is the highly paradoxical principle which these considerations imply. This principle was called by E. Mally the independence of so-being (Sosein) from being: 1 an object can still be 'so', i.e. such and such, even though it has no being of any sort. Meinong admits that the principle is very diffi­ cult to stomach, but thinks that this is solely due to our prejudice in favour of the actual. There is very little sense in saying that a house is big or small, or that a region is fertile or unfertile, until we know whether there really is such a house or such a region. But that a fact is trivial or practically unimportant does not make it less of a fact. Even if some one were to doubt whether a golden mountain was really golden, he might be led to see that it was a genuine object by another route. It is a fact that the golden mountain does not exist, and this fact is as 'hard' as any other fact, yet it is about an object which has no existence. It follows that, even in the case of a non­ existent entity, there are definite facts which concern it, and that it is out of the question to treat it as a mere nothing. Meinong formulates this point in an explicit con­ tradiction: There are objects concerning which it is the case that there are no such objects (es gibt Gegenstande, von denen gilt, daJ3 es dergleichen Gegenstande nicht gibt). There are several ways in which this contradiction may be avoided. The first is the obvious way, always resorted to on such occasions, of saying that non-existent objects have 'existence in thought' or 'existence for thought' ; in this sense alone will it be true that there are such objects. Untersuchungen, p. 1 2 7 ; 0. Gegth. Gs. Abh. II, p. 489. Professor Mally has long since abandoned the point of vi ew of this principle ; his later views, which remove many of the difficulties of Meinong's theory, will be dealt with in Chapter IV. 1

I T S I N D I FF E R E N C E T O B E I N G

45

As we saw in the previous chapter, Meinong does not believe in an 'existence for thought'; the thought of a round square may exist, but the round square cannot exist either in or for thought. We saw that Meinong allows us to say of the object of an idea, that it has 'pseudo-exis­ tence', a rather obscure conception of which several inter­ pretations are possible.' X has pseudo-existence' may mean simply that an idea of X exists, in which case the falsehood lies in the transference of the existence from the idea to its object, or it may mean that X is believed or supposed to exist, or, finally, it may mean that, by virtue of its relation to an existent, X acquires some sort of being which cannot be identified with existence.This last would be the being­ for-thought 1 of Professor Stout, where the hyphenation indicates that we are dealing with a sort of being, and not merely with being in a particular situation. It is not important for us to decide which of these con­ ceptions is identical with Meinong's pseudo-existence, though all his statements tell against the last.In any case, he rejects the view that every non-existent object neces­ sarily has pseudo-existence, that is, must be presented to some mind or other. With regard to an innumerable multi­ tude of non-existent objects it may be the case that no one thinks of them or needs to think of them. That there were no motor-cars in I 700 did not become a fact when some one first conceived the idea of a motor-car; it was a fact in I 7 00, though the minds of that period were unable to apprehend it. A deficiency or lack in a certain direction does not become less of a fact because we are unable to discover that anything is missing. And every deficiency or lack is a fact about a non-existent object. If, further, we take the case of a person who investigates the properties of a space which is not the space of our actual worId, it would be absurd to maintain that the facts which he dis­ covers would not have been facts if he had not conceived 1

See Studies in Philosophy and Psychology, p. 3 3 5 et seqq.

46

T H E PURE O B J E CT A N D

the notion of the space in question. Here, too, we have mind-independent facts about the non-existent. Meinong closes the discussion by saying that there would be a faint shadow of reason in maintaining that every object must be thought of in order to exist, but that it is utterly absurd to hold that an object must be thought of in order not to exist. 1 Another way of dealing with the problem of non-exis­ tent objects would be to extend the notion of being to cover their case. This was the solution of Mr. Russell when he wrote The Principles of Mathematics. There he says : Being is that which belongs to every conceivable term, to every possible obj ect of thought-in short to everything that can possibly occur in any propositions true or false, and to all such propositions themselves . . . . '.A is not' implies that there is a term .A whose being is denied, and hence that .A is . . . . Numbers, the Homeric gods, relations, chimeras and four-dimensional spaces all have being, for if they were not entities of a kind, we could make no propositions about them. Thus being is a general attribute of everything, and to mention anything is to show that it is. Existence, on the contrary, is the prerogative of some only among beings. 2

It would not be correct to deny that Meinong, even in his later phases, sometimes inclines to conceptions which resemble those of Mr. Russell. At no time would he have set relations, which have subsistence, side by side with chimeras, which neither exist nor subsist ; but there are times when he seems to believe that even a chimera is in some sense really there. Thus he says that even a non­ existent object presents itself to his thought with a posi­ tiveness that cannot be argued away, and that he is therefore unwilling to exclude the possibility of a third sort of being beside existence and subsistence. 3 But it would be quite wrong to take such a tentative statement as an 1

J

0. Gegth. Gs. Abh. I I, p. 49 1 . 0. A., p. 80.

2

Prine. of Maths., p. 449 .

ITS I N D IF F E R E N C E TO B E I N G

47

expression of Meinong's settled belief ; in his treatment of the subject in Ober Gegenstandstheorie he allows not a vestige of being to chimeras and four-dimensional spaces. It is therefore unfortunate that it should be generally believed that Meinong and Russell mean the same thing by the word 'subsistence', and that Meinong attributed subsis­ tence to chimeras. Meinong tells us that he originally believed in a variety of being possessed even by chimeras, to which he gave the name of Quasisein (quasi-being). This sort of being, like the being of Russell, pertained to everything ; it was dis­ tinguished from other varieties of being by the fact that it had no contrary. For if it had a contrary, the entities which lacked Quasisein would have to possess Quasisein of a higher order, since they would certainly not be nothing. And so we should be drawn into an infinite series of orders of Quasisein ; which is not an impossible but a very un­ plausible assumption. Quasisein had therefore no contrary, but belonged to all entities whether they existed or not. Meinong rejected the doctrine of Quasisein, because he could not conceive that there should be a variety of being to which no corresponding non-being was opposed. If being means anything at all, the statement that X is must contribute something to our knowledge, and this will only be so if it is conceivable that X is not. A being which automatically belongs to every entity and whose contrary is inconceivable is really nothing at all. 1 We may note here that Meinong subsequently believed himself to have found a non-existence more profound than that of the round square, and this might have functioned as the required contrary to Quasisein. In the conception of a v6'Y} aL� vo�aEw� there lies, according to him, a senseless­ ness with which even the round square is not infected ; a person who tried to think about the very thought he was thinking and nothing else, would be thinking a thought 1

0. Gegth. Gs. Abh. I I, p. 492.

48

T H E P U R E O B J ECT A N D

so nonsensical that beside it the thought of the round square would be as sensible as a dictionary. 1 But this state­ ment of Meinong's is so tentative, and the analysis of the situation involved so difficult, that we had better ignore it. Meinong considers two arguments which might be brought forward in favour of Quasisein. The first is based on the fact that, in order to refer to a non-existent object, we must for the time being suppose, though we need not believe, that some such entity exists. 2 In reading a novel we have to assume that certain persons and objects exist, we have to pretend to ourselves that they are real, in order to be able to think of them at all . 3 It might be argued that this involves that an object without any being is no object at all, and that non-existent objects must therefore at least possess Quasisein. M einong makes the somewhat unsatis­ factory reply that, even if we can only refer to non-existent objects by assuming them to have being, this is only a fact about our apprehension, and not about the objects them­ selves. Meinong does not explain why, if a chimera or a golden mountain are genuine but non-existent objects, we have to assume that they exist in order to have them before our thought. At a subsequent period he suggests, as a possible alternative, an immediate apprehension of non­ existent objects, by means of appropriate judgements, in which the non-existent object is treated as a pure object, without any assumption of its existence; he is not, how­ ever, inclined to treat this possibility very seriously. 4 The second argument for Quasisein is as follows: the fact that there are no golden mountains has being, and since golden mountains are a constituent of this fact, they too must necessarily have being. Meinong considers that this is an unjustifiable application to the entities which we call facts of a principle which is only evident in the case of other objects. A house can certainly not exist unless all 1 See Em. Preis., p . 2 3 . 3 See 0. A., p p . 2 2 7, 24r .

2 4

0. Gegth. Gs. Abh. I I, p. 49 3 . Ibid., p. 242 .

I T S I N D I F F ERE N C E T O B E I N G

49

its constituent bricks exist. But Meinong is very far from holding the crude view that the fact that there are no golden mountains is literally made up out of golden mountains, existence, and negativity. That the so-called constituents of a fact, i.e.the entities with which it is con­ cerned, are literally parts of it in the same way that a brick is part of a house, is a view that could only be entertained by a person who had never considered the utterly peculiar nature of the entities known as facts.Hence there is no reason to hold that, because a fact has being, the object which it concerns must have some sort of being as well. 1 Meinong then formulates his own doctrine on the sub­ ject: the pure object stands beyond being and non-being; both alike are external to it.Whether an object is or not, makes no difference to what the object is.The pure object is said to be auJJerseiend or to have AuJJersein : it lies 'out­ side'. What the object is, its real essence, consists in a number of determinations of so-being; the object 'elephant' for instance is determined by the determinations of being an animal, having a thick hide, having a trunk, and so on. Meinong believes that such determinations are genuinely possessed by an object whether it exists or not: the round­ ness of the round square is a fact about it which is unaffected by its non-existence.In contrast with such in­ timate determinations, the fact of existence seems wholly alien and extrinsic, which is the point stressed by Kant when he maintained that existence was not a genuine predicate of things. Meinong is careful to point out that this does not mean that any objects are exemp t from being and non-being; the law of excluded middle lays it down that every object necessarily stands in a fact of being or a fact of non-being. He maintains that being and non­ being have nothing to do with the object as object. In some cases the so-being of an object implies its non-being; the so-being of the round square implies its non-existence, 1

824187

0. Gegth. Gs. Abh. I I, p. 4 93 . E

T HE P U RE O B JECT A N D

50

but even this does not enter into the nature of the round square. 1 So far is it from being the case that a thing can only have properties if it is there to have them, that a thing can only be there if it has a definite nature which can be brought into existence. II

We must now consider the various attempts which might be made to rob Meinong's non-existent obj ects of their status as genuine objects. As we have seen, his doc­ trine of AujJe rsein rests on the grounds: (a) that there are facts of non-existence; (b) that these have a being indepen­ dent of the mind; (c) that facts of non-existence concern non-existent objects. We may first consider the view of those people who think that there are no genuine negative facts: to them the fact that there are no phoenixes, or that animals do not speak, is somehow less of a fact than that there are lions or that men speak. They would not, however, doubt that it is right to deny that there are phoenixes or that animals speak, and we may then ask how, if such a denial is not the apprehension of some fact about the universe, it can pos­ sibly be right or wrong, or anything more than a curious mental experience. When I deny that Herbert Spencer was witty, I am not merely expressing a subjective experi­ ence of rejection, and I am certainly not thinking about such an experience. Nor am I addressing a precept to all other minds that may think about Herbert Spencer. I am endeavouring to express something which is a fact or which I believe to be a fact. It is clear that in any universe a great number of things are absent, and that in any object in that universe a great many properties are lacking. Thus in our universe a per­ petuum mobile, a process travelling faster than light, an 1

See U. Gegth. Gs. Abh. I I, pp. 493, 494 ·

I T S I N D I FFERE N C E T O B E I N G

immortal organism are absent, and the absence of these entities is as interesting, curious, and ultimate a fact as any other. Similarly water lacks odour and colour and vis­ cosity, and the exclusion of such properties from its nature is undoubtedly a fact. There are people who regard such facts as merely subsidiary and derivative; there must, they maintain, be some positive fact about the universe which prevents a p erp etuum mobile from existing, or which pre­ vents water from having viscosity, and it is this positive fact which really j ustifies our denial in each case. It is not in the least evident that this need be so. It is quite con­ ceivable, for instance, that a certain liquid should merely lack odour, although it possessed no positive property i ncompatible with odour, or that certain objects should simply not be found in a universe though there was no positive circumstance which made their absence necessary. But even if we admit the principle of the positive ground, it is clear that we cannot identify an absence or lack with the positive fact on which it is grounded. If a liquid lacks odour because it possesses an unknown property P, we are still confronted by two distinct facts; their distinctness becomes very evident when we consider that it would be quite possible to know one of these facts without knowing the other, and that we frequently know that something is missing without knowing why it is missing. There seems then to be no possibility of a general exorcism of negative facts from the universe. We may point out in passing that the function of nega­ tion in knowledge is not merely the subsidiary one of guarding against possible errors. If all the concrete objects in the universe were revealed to our gaze, or if all the posi­ tive first-order properties of an object were known to us, we should not yet know that universe or that object com­ pletely, unless we knew that these were all the objects in that universe, or all the properties of that object. And this would involve knowing the negative fact that there were

THE P U R E OBJ E CT A N D

no other objects in the universe, or positive first-order properties in the thing. Negative facts are the bounding lines of the actual, and unless we know where these boundaries lie, we cannot know anything, however simple, with completeness. It is, however, possible to admit that there are negative facts, and yet to deny that there are genuine facts of non­ existence. The statement ' X does not exist' is not really about X, we might say, but about all the actual objects in the universe, and we are saying of them that none of them have the properties which would justify us in applying the name 'X' to them. Professor Moore writes in his article entitled The Conception of Reality : We have said that what 'Lions are real' means is that some particular property or other-I will say, for the sake of brevity, the property of being a lion, though that is not strictly accu rate, does in fact belong t o something-that there are things which have it, or, to put it another way, that the conception of being a lion is a conception which does apply to some things-that there are things which fall under it. And similarly what 'Unicorns are unreal' means is that the property of being a unicorn belongs to nothing. Now, if this is so, then it seems to me, in a very important sense, 'real' and 'unreal' do not in this usage stand for any conceptions at all. The only conceptions which occur in the proposition 'Lions are real' are, on this interpretation, plainly ( r) the conception of being a lion, and ( 2) the conception of belonging to something, and perfectly obviously 'real' does not stand for either of these. 1

On this view, which is also the view of Principia Mathe­ matica, 2 no statement of existence really concerns the objects which it appears to concern. In the case of an object which lay before us in immediate perception, it would be significant to ask what its colour or its dimen­ sions were, but not whether it existed or not.The state­ ment that it existed would not really be about it at all, but 1 2

Phil. Studies, p . 2 1 2 . See Principia Mathematica,

vol .

i, p. 1 74

(ed .

2) .

I T S I N D I FF E R E N C E T O B E I N G

53

i t would b e a general statement about all the obj ects in the universe, to the effect that one of them had the properties of the thing in question. That an obj ect is an obj ect at all, suffices; it does not require any addition of existence. All statements of existence or non-existence really point to general facts of so-being ; we are asking whether anything in the universe has a certain set of characteristics. 'Does Vulcan exist ?,' means 'Has anything the property of being a solar planet within the orbit of Mercury ?'; 'Vulcan does not exist' means that nothing has this property. In this analysis of the real meaning of statements of non-existence, all reference to non-existent obj ects has been eliminated; if economy of hypothesis were a valuable principle even in regions where direct insight is possible, there can be no doubt that such an elimination would constitute a very important advance. There are, however, insuperable obj ections to this re­ duction of facts of being to facts of so-being. Let us consider the case of a person who wishes that there were such a thing as the philosopher's stone. If the fact of the existence of the philosopher's stone is identical with the fact of the possession by some obj ect in the universe of the properties of the philosopher's stone, then we must suppose that what the person in question is really wishing is that one or other of the obj ects in the universe should possess the properties of (a) being a stone, (b) turning baser metal into gold. But it is perfectly clear that the man might wish nothing of the kind. He might be perfectly satisfied with all the obj ects in the universe, and, according to the theory, these are all the obj ects with which we are concerned. But he might wish, not that any of the obj ects in existence should be other than it is, but that some other object, some obj ect not comprised among the obj ects of our universe, but whose nature is nevertheless determinate in various ways, should be comprised in that universe, that is, should exist. If a childless woman wishes for a child, she

54

T H E P U R E O B J ECT A N D

does not wish to stand in the relation of motherhood to one of the children actually in the world, but to another child, which is not in the circumstances an actual child at all, but which would become an actual child if her wish were fulfilled. It is no doubt true that, if her wish is ful­ filled, one of the things actually in the world wil l be her child, but it is equally true that, while she is wishing, she is directing her mind to something which lies beyond the actual world, and which wil l possibly always lie beyond it. Meinong thinks that many distinctions which belong to the theory of objects are obscured if we insist on regard­ ing logical equivalences as identities. Such a procedure may be justified in a symbolic system, where we are attempting to deduce all our propositions from the mini­ mum number of postulates, but it cannot be justified in philosophical theory. He criticizes the view of Heymans, who, like Professor Moore and Mr. Russell, believes that the statement ' Ghosts do not exist' really means that no actual thing is a ghost. Meinong says in this connexion that we need no theory of obj ects to recognize that, if on one occasion some one thinks about ghosts and denies their existence, and on another occasion thinks about something actual whether this be vaguely or precisely determined, and recognizes that such an obj ect is not a ghost, he is in each case thinking two totally different thoughts. And that the second thought amounts in practice to much the same as the first does not entitle us to say that the first is the second, or that the first is not as significant as the second.I

We may now point out that, even if it were possible to replace facts of existence by facts of so-being, we should still be faced with very similar problems, since every negative fact is the absence or lack of something. There is no escape from this, unless we substitute for negation something which is not negation. If we consider only 1

See Stellung I, pp. 8 3 , 84 .

I T S I N D I FFERE N C E T O B E I N G

5S

actual objects, and their positive relations to each other, we shall nowhere meet with a negation; a negation is essentially an absence, an exclusion, and what it excludes can have no place in the actual world. We may then ask what precisely is excluded by the negative fact that nothing in the world has the properties of a ghost, which is the suggested substitute for ' Ghosts do not exist'. It is clear that it does not exclude the real objects in the world, nor does it exclude the properties of a ghost, provided these are taken separately. But what it does exclude is a possible combination of these properties, and their embodiment in a concrete case, in short the circumstance that something should have the properties of a ghost. This is presupposed in the negative fact, as its negatum, just as ghosts appear to be presupposed in the fact of their non-existence. If, then, we can admit such non-existent entities as the pos­ sible circumstance that something should be a ghost, there seems to be no reason why we should not admit ghosts themselves as genuine objects. The upshot of these arguments has been to show that, unless we are willing to abandon negative facts altogether, we must admit that in some sense non-existent objects are still objects. The absence of the philosopher's stone from our universe is not the same as the absence of the islands of the blessed, the darkness which is the lack of light is not the same as the silence which is the lack of sound. Absences and lacks are all perfectly specific and distinct, just as the holes in a piece of lace have shapes as characteristic as the actual embroidery. We do not create such absences or lacks by our thinking, but discover them in the same way that we discover the Milky Way, only with much greater difficulty, since something has first of all to suggest to our mind the possibility of entities which would fill up the gaps in question. Unless we wish to accept the unplausible theory that the difference between these gaps in the uni­ verse is something irreducible and purely qualitative, we

56

T HE P U RE O B JECT AN D

must find their distinguishing feature in the obj ect which is in each case absent. Non-existent obj ects are therefore by no means confounded in the grave of a common nothingness. In spite of these arguments, the theory of a vast realm of objects beyond the confines of being retains a certain immense unplausibility. It is hard to believe that the wildest fancy and the most arbitrary definition, which appear to evoke an object out of nothingness, are in reality only selecting something out of the infinite abundance of AujJe rsein. 1 Meinong is right, of course, in maintaining that neither fancy nor arbitrary definition c reate their objects, in the sense of causing them to exist, for the simple reason that these objects exist as little when we are imagin­ ing them as at any other time. Yet it is hard to see why, if the world of AujJe rsein is so much richer than the actual world, our prejudice in favour of the actual is so unshake­ able. Why do men of science devote themselves to the study of horses and cows and not to that of dragons or unicorns, or why are the ways of chlorine more interesting than those of phlogiston ? The answer to this question throws a great deal of light on the nature of AujJe rsein, in so far as this lies beyond the boundaries of the actual. We speak of the wo rld of AujJe r­ sein, but in reality the objects which have no being do not constitute a world. They are a chaos of incoherent frag­ ments, and the only relations that subsist between them are those of similarity and diversity. Between actual exis­ tents we can always discover a large number of intimate connexions; they people a single space-time and influence each other profoundly. Beyond the boundaries of being no such connexions subsist: there is no fixed distance between Valhalla and the islands of the blessed, nor does the planet Vulcan exert a gravitational pull on the star of Bethlehem. It is obvious that such isolated objects cannot 1

See 0. A., p. 2 74.

I T S I N D I FF E R E NCE T O B E I N G

57

interest a science which is always striving to discover law and system in its material. Another fatal weakness in the objects which have no being is that some of them are not fully determined, and about such objects few questions can be significantly asked. All the objects in the actual world are fully deter­ mined, and we can pass beyond the determinations which we know to others which we do not know. We can ask about a certain mountain, for instance, how high it is, and what its geological strata are like, and be sure that such questions have definite answers. But in the case of non­ existent objects these questions do not always have answers; we cannot say how high a golden mountain is, because it is simply indeterminate in respect of height. This explains the folly of the problems which, according to Suetonius, perplexed the senile mind of Tiberius: what song did the sirens sing or who was the mother of Hecuba ? These are insoluble problems because they involve the assumption that Hecuba had a definite mother, or that the sirens sang a perfectly determinate song. It is clear that in dealing with some non-existent objects we must com­ pletely alter our ordinary habits of thought and, after a certain point, relinquish all desire for further informa­ tion. Again, if some non-existent objects are indeterminate, there are others which are impossible, such as the round square, and we can hardly hope to find in them a fruitful field for scientific investigation. From another point of view Auj]e rsein is incapable of scientific treatment because of its excessive richness. In the case of the actual world we can always ask whether a certain object is comprised in it or not; the question is interesting, because some things are excluded from it. The realm of Auj]e rsein, however, has no such exclusiveness; every possibility or impossi­ bility is comprised in it, and this fact silences a multitude of questions.

58

PURE OBJ ECT AND IND IFFE R E N C E TO B E I N G

Auj}ersein is a strange sort of desert in which no mental progress is possible, but the desert has many oases, as no one who has read a fine novel, or a treatise on meta­ geometry, can possibly doubt. Such oases are also the infinite number of possible worlds which, according to Leibniz, were presented to the choice of God; all of these are as interesting and as highly organized as our own universe, though we have neither the time nor the wit to 'think them out'. Auj}ersein comprises these articulated fragments, and our own universe, as a pure object, is one of them, but it remains, as a whole, too chaotic to be studied scientifically. It is only interesting because, like the Boundless of Anaximander, it is the inexhaustible source of all existence, 'that becoming may not fail'. We have therefore explained our prejudice in favour of the actual, without allowing that non-existent objects are simply nothing at all.

III THE THEORY O F OBJECTIVES

S

I

o far we have frequently spoken of 'facts' as a peculiar type of entity, distinct from the objects about which they are. While every scientific work professes to set forth certain facts, and while nothing is more common than to make an appeal to the facts, it is not usual to consider what sort of thing a fact is. Here, as in other cases, it seems that a type of object may be involved in our simplest experiences, that it may have been familiar to men for centuries, and that it may even have received various inappropriate names, before its nature is properly discerned and its peculiar status acknowledged. Primitive thought is almost exclusively occupied with concrete existents ; the discovery of charac­ teristics and incomplete objects goes back to the time of Plato, who acquainted philosophy with objects like Man and Goodness ; relations and facts have only been recog­ nized as distinct varieties of object in very modern times. In the case of the last two types of object Meinong made some of the earliest and most valuable investigations. His theory of relations we postpone to another chapter ; the present chapter will deal with his work on facts. The anatomy of the fact is in one sense thoroughly familiar to philosophy and grammar ; under the heading of 'the judgement', or 'the proposition', or 'the sentence', most of the peculiarities of the fact have been noted, and its more important kinds have been distinguished. But such knowledge has remained confused and unsatisfactory, because it has been mixed up with irrelevant psychological and linguistic material.

60

T H E T H E O R Y OF O B J E C T I V E S

We saw that Meinong widened the conception of an obj ect to include objects which were not nothing, but which nevertheless did not exist ; in the same way he has added to the realm of facts a set of entities which resemble facts in every essential respect, but differ from them in so far as they are not the case, that is, in so far as they lack the kind of being which is appropriate to facts. To the whole class of entities of which some are, and others are not, the case, he has given the name of objectives. Objec­ tives are such things as 'that China is a Republic', 'that there is an integer between 3 and 4', 'that j ealousy is a good emotion'. Meinong tries to prove that they are a unique and irreducible sort of entity, indispensable to our knowledge of reality and to reality itself.

II

We are introduced to objectives in the first instance by a long discussion of sentences and judgements, which Meinong afterwards admits to be rather a misleading ap­ proach, because it makes no difference to the nature of any entity how we apprehend it or in what words we express our apprehension. He remarks: We must make special and emphatic mention of the fact that the obj ective, like every other obj ect (in the wider sense), allows of non­ psychological treatment and accordingly demands such treatment . . . . Without considering an apprehending subj ect and its experiences we may say that every obj ect (in the wider sense), has both its generic and its specific nature, and, on the basis of these natural distinctions, obj ects fall apart into the two big groups of obj ects (in the narrower sense), and obj ectives. 1

He confesses, however, that this purely objective approach is difficult ; it is certainly the case that in starting with questions of thought and language he is following the line I

0. A., pp. 60, 6 r .

THE THEORY OF OBJECTIVES

61

of development that logical discussion has taken. Logic existed as a study of verbal argument and correct thought for centuries before its objective implications were made clear. According to Meinong words serve a double function, they express our inner experiences and they mean or refer to the objects of those experiences. Thus, if I make use of the word 'sun', I am, whether I wish it or not, giving expression (Ausdruck) to the particular mental process called an idea, to the fact that, either in perception or imagination, something is being set before my mind. But at the same time, in so far as I express this idea, I also refer to a certain physical object, namely the sun, and this reference to the sun is the meaning (Bedeutung) of the word. 1 If a person hears me use the word 'sun' he can take this word as a sign of a certain idea in my mind, whose existence he can infer with high probability. But it is perfectly plain, as Meinong points out, that this idea is not what I mean when I speak of the sun; unless I am intro­ specting, and attempting to examine the mental states which accompany my use of words, nothing can be farther from my thoughts than my own ideas. What I am con­ cerned with, what I am referring to, is an extended physi­ cal object millions of miles away, which does not resemble my mental processes and stands in no real2 relation to any­ thing in my mind. Meinong thinks that the tendency to confuse the ex­ pressive and the significant function of words goes back to the more fundamental confusion of the content and object of mental states. Once it has been supposed that the sun is either identical with the idea of the sun, or is in some way a part of it, there is no reason why a person who expresses his idea of the sun should not at the same time Ibid . , p. 25. To Meinong the relation of a mental state to its object is not a real, but a purely ideal relation, like similarity. I

2

T HE T HEOR Y O F O B JEC T I VES

give expression to the sun itself. But, if content and obj ect are clearly distinguished, and it is realized that no physical object is ever a constituent of any mental state, then it is plain that the sense in which a word is a sign of a mental process in the user of the word, and the sense in which it is a sign of something he is thinking of, are totally different. Meinong now turns to consider the sentence, and applies to its interpretation his distinction between the expressive and the significant function of words. That there is some very fundamental difference between those words or complexes of words which constitute sentences, and those which do not, has been quite clear since the time of Aristotle.He pointed out the difference between a verbal combination like 'goat-stag' which means some­ thing but is not a sentence, and a form like 'the goat-stag exists' which is a sentence. But in what objective feature such a difference lies he does not suggest, though he tells us that we require both a noun and a verb to make a sentence.1 We may look for the characteristic function of a sen­ tence either in what it means, or in what it expresses. If w e pursue the former course, we might believe that the meaning of a sentence was a complex whole whose con­ stituents were the meanings of its several parts. But it is clear, if this is so, that a sentence and a form of words that is not a sentence can have the same meaning. 'This metal is light' is a sentence, 'this light metal' is merely a phrase, but, if their meaning is merely a complex of the meanings of the words which compose them, they mean exactly the same thing.In both cases we have a particular thing (this metal) and a characteristic (light), and the latter is given as characterizing the former.If we are merely to consider the constituents of the complex whole before us and the 1 That mere verbal complexity has little to do with the characteristic functions of the sentence becomes clear when we remember, as Meinong points out, that a sentence need not be a complex of words (Credo is a complete sentence), and that a complex of words need not be a sentence. See 0. A., p . 30.

T H E T H E O R Y O F O B J ECT I V E S

63

mode of their synthesis, it is hard to see in what respect the meaning of the two phrases could differ. We are therefore led to consider the possibility that the verbal forms 'This metal is light' and 'this light metal' do not differ in meaning at all, but that the difference is solely one of expression. The inner experiences expressed by the sentence differ from those expressed by the corre­ sponding phrase; a certain objective situation is set before the mind in both cases, but the way in which this situation is apprehended is different in each case. What this differ­ ence is, seems clear enough: 'this light metal' expresses a mental attitude of pure contemplation, we merely set an object before the mind and look at it, without deciding anything affirmatively or negatively about it, and without having any feeling of conviction. On the other hand, the sentence 'This metal is light' expresses, not merely our apprehension of a situation, but a definite conviction with regard to it, in this case an affirmative one. We assent to, or accept, a certain suggested connexion. Now these characteristics are the marks which distin­ guish two fundamental types of experience, the type known as the //orstellung (the pure presentation or idea), and the type known as the Urteil or judgement. Meinong remarks that there are two 'moments' which distinguish a judge­ ment from a mere idea; a judgement adds to the pure idea on which it is built a moment of conviction ( Oberzeugt­ heit), and, secondly, in this conviction it adopts one of two attitudes to the pure idea, the attitude of acceptance, or that of rejection, that is, it has a determinate position between affirmation and negation. 1 In the pure idea some­ thing is given to us, but we have no beliefs about it, either affirmative or negative. There need then be no difference in meaning between a sentence and a phrase, but they express different types of experience, a sentence expressing a judgement, whereas a phrase expresses a mere idea. I

0. A., p. 2 .

T H E T H E ORY O F O B J E C T I V E S

That the traditional account is correct, and that sen­ tences do in a large number of cases express judgements, is admitted by Meinong. 1 He believes, however, that there exists a peculiar class of experiences, called by him supposals or assumptions (Annahmen), which are inter­ mediate between pure ideas and judgements. The dis­ covery and analysis of these experiences is one of his most valuable contributions to philosophical psychology; we shall postpone a full consideration of them to a later chapter. The peculiarity of an assumption is that it pos­ sesses one of the distinguishing 'moments' of a judgement, but not the other. It lacks entirely any trace of the conviction which is essential to judgement; on the other hand it retains a determinate position between affirma­ tion and rejection which distinguishes it from a mere idea. 2 Meinong restricts the application of the word 'assump­ tion' to those experiences in which we 'take' something to be the case, quite regardless as to whether there are grounds for believing it to be the case or not. His assump­ tions are therefore totally distinct from some of the 'as­ sumptions' of ordinary speech which are really surmises or judgements. Thus Meinong gives as an example of an assumption the assumption that the Boers were victorious in the war of I 8 99- 1 902; we can suppose this to be so for the purpose of a discussion, though we know that it is not so. If an assumption is completely lacking in con­ viction it is nevertheless distinct from a pure idea. The man who supposes the Boers to have won the Boer War, is saying 'yes' to a certain something which lies before his mind even if he is only saying 'yes' in play. To make a negative assumption, such as assuming that the Boers were not defeated in the war in question, involves, in the same way, a definite 'no'. In making assumptions we are definitely active, whereas pure ideas involve no more than I

0. A. , p. 32.

2

Ibid., p . 4.

T H E T H E O R Y O F O B J E C T IV E S

65

a passive self-abandonment to objects, without any attempt to make anything out of them. 1 Admitting then the existence of these judgement-like experiences which Meinong calls assumptions, we may reformulate the proposed theory of the characteristic function of the sentence, and say that a sentence is dis­ tinguished from all other linguistic forms in that it gives expression to a judgement or an assumption, that is, to some experience in which the mind fulfils an active accep­ tance or rejection, whether this acceptance or rejection be accompanied by conviction or not. Meinong now shows that this subjective theory of the sentence, which has been thought good enough by most philosophers since Aristotle, will not hold water for a moment. In the first place it is quite clear that sentences do mean something; they are not merely expressions of inner experiences, but they also refer to the objects of those inner experiences. We are therefore led to consider what the objects of judgements and assumptions can pos­ sibly be. Meinong takes the case of some one who judges, in a rather turbulent election, that there has been no disturbance of the peace. If this denial is an act of know­ ledge, then there must be something that is known by its means. To deny is not merely to experience a personal reaction towards something, as a feeling of dislike is purely personal; it is an act of mind which makes a claim to know something as it really is, and if the denial is a valid one, there must be some object which is apprehended by its means. It is quite clear, however, that a disturbance of the peace, which is the sort of thing which might be an event in the real world, is not the object that is apprehended by the negative judgement, for the simple reason that such an obj ect does not exist. That there is knowledge of the properties of non-existent objects would of course be I

824187

Ibid., P· F

227.

66

T H E T H E O R Y O F O B J ECT I V E S

admitted by Meinong, but when we pass the judgement in question we are not attempting the analysis of a pure object regardless of its existence. We are trying to get a better understanding of a certain section of the history of the actual world. It is clear that, if we attempt to state what is meant by the judgement in question, no form of words less simple than 'the non-existence of a disturbance of the peace', or 'that there has been no disturbance of the peace' is at our disposal. Meinong points out two curious things about the object which these words mean. In the first place it is perfectly senseless to say that it exists, as a definite piece of reality. A disturbance of the peace might exist, but its non­ existence could hardly be said to exist. On the other hand, there are things which language does permit us to say of the non-existence of such an entity as the disturbance of the peace; we can say that such a non-existence is the case, or is so, or is a fact, or simply that it is. 1 The result of our analysis is, therefore, that the judgement in question does apprehend some object, or else it would not be an act of knowing at all, but that this object is of such a sort that it is wholly unmeaning to say that it exists; the only sort of being of which it is capable is 'being the case'. It is clear that a similar analysis will apply in those cases where we make affirmative judgements or assumptions . If we judge that there is snow, we have judged something which is not simply identical with the object 'snow'. For snow is something which may exist or not exist; if the latter possibility is distinct from the object 'snow', the former must be so too. Detailed arguments of thissort might be used in every case to distinguish between the mean­ ings of sentences and the objects which they concern. But the ultimate justification for Meinong's view must be our simple insight into the fact that when I judge e.g. that grass is green or a mountain high, the being-green-of-the-grass I

V. A., p. 43 ·

T H E T H E O RY O F O B J ECT I V E S

67

or the being-high-of-the-mountain is a definite and distinct something, which is nevertheless not an existent as the grass, the mountain, or their properties are. 1 At most we can call such entities facts, but to be facts is a privilege which only some of them possess. False beliefs and false assumptions also set something before our mind, and that this something is nowhere to be found in the real world does not make it nothing. Just as the philosopher's stone is a genuine object, though it does not exist, so the existence or the discovery of the philosopher' s stone is a genuine object, though it is not a fact. Meinong proposes to use the word 'objective' for these entities which can be judged and assumed, and which are in some cases facts. He sets them over against objects in the narrower sense, which can be given to us by mere ideas (Forstellungen), and which are never the case. Vv e may use the word 'object' to translate Meinong's Gegenstand; an object is anything to which a mental process may be directed. For Meinong's word Objekt, which applies only to objects in the narrower sense, i.e. to those which are not objectives, we shall in this chapter use the word 'objectum'. Objects therefore divide into the two classes of objectives and objecta. Meinong now points out that every judgement or as­ sumption has at least two objects; each appears to be the object of the judgement or assumption from one point of view. These two objects are the objective itself and the objectum which the objective concerns. If I ask simply what I am judging or what I am assuming, in the sense that I ask for the object of an idea, it is clear that the objectum is not the entity required. For it is impossible to judge or assume an entity like snow; we can only assume or judge that it exists or is white, and this is an objective. r

1 According to Meinong, properties of existents and some relations between existents (the 'real' relations) themselves exist : Geg. htJh. Ord. Gs. Abh. II, p. 395 ; Mog., p . 1 69.

68

T H E T H EO R Y O F O B J ECT I VES

Snow can only be apprehended by a judgement or assump­ tion in so far as it is one of the constituents of an objective. From this point of view the objective is the primary object of judgement or assumption, and the objectum is only indirectly given as that about which something is thought. If I know that there has been no disturbance of the peace, the objectum 'disturbance of the peace' is only given to my knowing as something about or concerning which something is the case. 1 On the other hand, if we consider the matter from the point of view of our natural interests, the obj ective sinks into the background in favour of the objectum, much in the same way as a content sinks behind an object. If I make judgements or assumptions about a certain objec­ tum, it is not the objectives, but the objectum whose nature becomes clearer by their means. To illustrate Meinong's meaning, the description of a certain plant is not interesting as a mere description; the various facts into which it resolves itself are not severally interesting in their own right, they are only lines of approach by which we gain a more and more lucid and detailed mental picture of the plant and its properties. No one cares to know isolated facts about people, places, or objects otherwise unknown; it is only if a fact is 'about' some obj ectum to which we have been introduced by other facts that it has any significance for us. We have a prejudice in favour of objecta as opposed to objectives, and this prejudice is assisted by the fact that judgements and assumptions are a variety of experiences which, unlike ideas, do not naturally present their objects. 2 If I judge that there is snow, the natural focus of my interest is on the phenomenon 'snow'; the objective that there is snow, which is what the judgement really 'gives' 0. A., p. 52. Ibid., p. 29. Meinong really means that they do not apprehend their objects in the full sense. In Mog., p. 249, he holds that they always present their objects. I

2

T H E T H E O RY O F O B J E C T IV E S

to my mind, is not presented to me in the same startling fashion as the snow, and requires a special act to bring it from the margin into the focus of attention. This act does not, however, present the difficulties, amounting to impos­ sibility, which attend the effort to be aware of contents as opposed to objects. Normally an objective is judged (geurteilt) or assumed (angenommen) ; while the objecta in­ volved in it are judged about (beurteilt) or assumptions are made about them (they are beannahmt). 1 Now it is perfectly possible for an objective to play the part of an objectum in the sense that we no longer judge or assume it, but make judgements or assumptions about it. To do this has the additional advantage of proving, with the greatest possible clearness, that objectives belong wholly and solely to the realm of obj ects, and are utterly indif­ ferent to the workings of our minds. If I say 'It is estab­ lished that the files have not yet been closed', it is quite clear that what has been established is not any judgement or assumption of mine, but an obj ective which concerns the files. 2 Similarly if I say 'It is not the case, that A does not exist', it is clear that what is not the case cannot be anybody's judgements or assumptions about A. 3 Of these it is senseless to say that they are or are not the case, since a judgement or an assumption is an experience capable of existence, that is an objectum; it is quite impossible to identify an experience which exists, with an objective which cannot by its very nature exist, and which has the further peculiarity of being an objective of non-existence. I II

So far Meinong has established the being of objectives indirectly, by showing that judgements and assumptions do have objects, and that it is impossible to identify these I

3

0.A., PP· 44, I 32. Ibid., p. 50.

:1.

Ibid., pp. 4 8, 49·

T HE T HEORY OF O B JECT I VE S

70

objects with any of the things about which the judgement or assumption is made. He follows this up by a sketch of a non-psychological characterization, 1 which he admits to be very inadequate. That the division of objects into objectives and objecta is exhaustive he sees no method of proving. 2 He then quotes a rather cryptic statement of his pupil Ameseder which runs as follows: Every obj ect has being (or non-being). There are, however, obj ects which not only have being (in this widest sense), but also are being, and these obj ects are the obj ectives, while those obj ects that have being but are not being, are thereby characterized as obj ecta.3

This statement becomes clearer when we remember that Meinong divided all objectives into Seinsobjektive (objectives of being) and Soseinsobjektive (objectives of so­ being or inherence). 4 'There is snow' is an objective of being, while 'Snow is white' is an objective of so-being. 'Being' simp!iciter plays the part of predicate in the first objective, whereas 'being-white' plays the same role in the second objective. If we extend the notion of being to include that of inherence, as Aristotle for instance did, 5 then every objective is either the being or the so-being of some entity. That Meinong could say this, shows how utterly remote his objectives are from anything like a proposition or a judgement. For we could never say that the judgement that snow was white was the being-white-of-snow ; at most we could say that the judgement predicated whiteness of snow. Just as little could we say of the proposition 'Snow is white' (if such an entity exists at all) that it was the being-white-of-snow. At most we could say that the pro­ position said this, or meant this, or corresponded to this (or in whatever other way the relation of a proposition to a fact might be conceived). But it is clear that the objec1 4

0. A., p. 59, § r r . Ibid ., p. 7 2 .

Ibid., p. 6 r . s Metaphysics,

2

1017

a.

3

Ibid.

T HE T HEORY O F O B JE C T I VES

7I

tive 'Snow is white' neither says anything, nor means anything, nor corresponds to anything, nor predicates any­ thing of anything. It is not a mental activity which can do something; it simply is something, namely, the being­ white-of-snow. IV

A second fundamental characteristic of objectives is their dependence on objecta. An objective is an object of higher order, 1 which is 'built upon' other objects. Every­ thing which is the case, is the case with regard to or con­ cerning something else; and those objects about which an objective is, may be called its material. 2 The objective that snow is white is built upon or presupposes the object snow, which is therefore its material. The 'material of an objective' plays much the same part as 'the constituents of a proposition' or 'the constituents of a fact' in the terminology of modern logic. The immediate material of one objective may of course be another objective; an objective is an object, and there is no reason why there should not be objectives about it. Thus the objective ' 0 is a fact' has as its material the objective 0. 0 might itself have another objective P as its material, and P might be about Q. There is no limit to the complexity of such a Chinese-box system; however complex 0 may be we can always turn to 0', which con­ cerns 0, and which is therefore of a higher order of com­ plexity. But though there are always objectives which are more complex than any given objective, there is no objec­ tive of infinite complexity; we could not have an objective which concerned an objective which concerned an objec­ tive and so on ad infinitum. In every case we should neces­ sarily arrive, after a finite number of steps, at an objective whose material was not an objective but an objectum. In this way, though the proximate material of an objective I

Em. Preis., p. ro5.

2

0. A., P· 63.

T HE T HEO R Y O F O B JECT I VE S

may be another obj ective, the ultimate material of all obj ectives are obj ecta; they are the foundation on which the whole structure of obj ectives reposes. This argument is an application to obj ectives of a prin­ ciple which Meinong elsewhere establishes for other ob­ j ects of higher order. He says of relations that they can never serve as an absolute starting-point; they can only subsist if there are terms between which they can hold. A relation may hold between relations, but ultimately we must come down to relations which hold between terms which are not themselves relations. Without such an abso­ lute basis, a relation would be like a comparison in which nothing was compared. 1 The same principle holds for all sorts of complexes, since all of these are dependent on relations. 2 In the case of obj ectives we may similarly say that, if we attempted to build obj ectives on nothing but obj ectives, we should have before us entities whose nature it was to be about something, and which were neverthe­ less ultimately about nothing. In this dependence of the obj ective on the obj ectum we may therefore find a charac­ teristic distinction between them; an obj ective must be about something, and it must have some finite power of the relation 'about' to an obj ectum; on the other hand, no obj ectum is ever about anything else. We may note, however, that the dependence of the obj ective on the obj ectum is not a dependence as regards being. An obj ective which is a fact (and which therefore has the being peculiar to obj ectives) need not be about an obj ectum which has being. 'There is no perpetuum mobile' is an obj ective which is a fact, whereas the objectum which it concerns, the perpetuum mobile, has no being. Again an obj ective may concern an obj ectum which has being and yet fail to be a fact; thus 'Rome is in France' is about something which exists, yet it is not the case. As far as being goes, the sphere of obj ectives seems to be wider 1

Reith. Gs. Abh. II, p. 44 .

2

Geg. Mh. Ord. Gs. Abh. I I, p. 3 8 9.

T HE T HEORY O F O B JECT I VES

73

than the sphere of objecta, because there are all sorts of interesting facts about objecta which do not exist. To interpret such facts solely in the light of their implications for actual existents, and thus to force them into connexion with the actual world, is very unnatural. The non-existence of the phoenix is surely a fact in its own right which con­ tributes to the total system of facts ; it is not merely a fact, because actual existents think of phoenixes or miss them, or because their absence has such and such effects in the real world. Thus the dependence of objectives on objecta will not prove all that is desired by people obsessed by the prejudice in favour of the actual ; it will certainly not satisfy those who believe mystically that all facts somehow qualify or describe a great objectum called Reality. We may also note that an objectum, though it may be the basis of objectives, could certainly not exist unless there were objectives about it, facts in which its being and nature lay, and in this sense, though Meinong does not consider it, an objectum is as much dependent on the objectives which concern it, as they are dependent on it. V

Meinong then indicates as a third distinguishing mark of objectives their incapacity for existence ; 1 the Antipodes exist but their existence cannot be said to exist. In this doctrine, as in his whole conception of objects of higher order, Meinong is keeping close to the intuitions of the ordinary man. The ordinary man feels that similarities, patterns, relations, numbers, are in some sense 'really there' ; at the same time they do not exist in their own right alongside of existents. If we treated the existence of the Antipodes as an existent, the existence of its existence would be another existent, and so we might spin an in­ finite number of existents out of the Antipodes. We feel, 1

0. A., p. 6 3 ; 0. Gegth. Gs. Abh. I I, p. 48 7.

74

T HE T HEORY OF O B JECT I VES

however, that all this is merely 'multiplying variety in a wilderness of mirrors'; we cannot by such means conjure anything new out of the concrete material before us. There is a necessary sterility in all abstract discussions, even if it is in some cases a noble and profitable sterility; it may deepen our understanding, but it cannot add to our in­ formation. If the existence of the Antipodes does not itself exist, its non-existence is nevertheless not on a level with the non-existence of the phoenix; it has a certain sort of being, though it may not have the reality of houses and trees. For this type of being Meinong employs the word sub­ sistence (Bestand). The difference between existence and subsistence is one that we apprehend immediately like the difference between blue and yellow; it neither admits nor requires analysis or justification. 1 Objectives are not the only subsistents: a relation such as diversity subsists be­ tween two entities, and the number of a group of existents subsists, but cannot exist as they do. In the case of objec­ tives, if they subsist they are known as facts (Tatsachen), or are said to befactual (tatsachlich). 2 If they do not subsist, they are unfactual (untatsachlich). Meinong points out several confusions which surround the word 'fact'. It is sometimes applied not to objectives but to objecta, as whenever we say that muscle-contraction is a physical fact, or that will is a mental fact. 3 Carlyle was guilty of a similar confusion when he spoke of the great fact of the universe, which, in the midst of a poor paper-age, revealed itself to the mind of Dr. Johnson. Obviously a muscle-contraction, an act of will, or the universe are not facts at all, any more than Mussolini or London are facts. That there are such entities as the above-named, or that they have such and such properties, or stand in such and such relationships, these things may 1 2

0. A., p. 7 3 ; 0. Gegth. Gs. Abh. II, p. 48 7. 3 Ibid. 0. A., p. 6 9 .

T HE T HEO R Y O F O B JECT I VES

75

very well be facts, but no objectum can itself be a fact. It is rather pleasant, incidentally, to consider that those people who make a frequent appeal to 'hard facts' are making their appeal to something which does not and cannot exist. The second unjustifiable limitation of the word 'fact' is that we apply it more readily to those facts which are about existents, and which are discovered by empirical means, than to those facts which concern subsistents and which are known a priori. It is not usual to call the subsistent objectives of logic and mathematics 'facts'; we generally take refuge in non-committal words such as propositions, theorems, principles, or truths. But it is clear that the implication which holds between equilateralness and equi­ angularity in triangles, is not a form of words, but a fact just as the hotness of certain summers is a fact; the only difference between the two sorts of fact is that in the latter factuality merely adheres to the objective in an external and contingent fashion, whereas in the former it inheres in it intrinsically and necessarily. 1 Further, on Meinong's view, it is a fact that pink elephants are pink, though there are no pink elephants. Meinong therefore uses the word fact to apply to everything that is the case, whether it concerns the existent, or the subsistent, or entities which neither exist nor subsist. VI

As a fourth characteristic of objectives we may note that their relation to their own sort of being is distinct from that of objecta to theirs. All objecta are, as we saw, indifferent to being; whether they are or are not, makes no difference to what they are. It is only by virtue of their inclusion in objectives that they come to be or not to be. 2 The case of an objective is utterly different. No objective 1 For adhesive and inhesive factuality see Mog., P · 1 4 2 , 2 2 1 . P 2

V. Gegth. Gs. Abh. II, p . 493·

76

T HE T HEORY O F O B J ECT I VES

acquires factuality or unfactuality merely because it serves as the material for superordinate obj ectives. The truth of the principle will become plain if we tem­ porarily suppose it to be false. On such an assumption, there will be a sense in which unfactual obj ectives are factual, and factual objectives unfactual, a conclusion which makes nonsense of the whole theory of objects. If we take an unfactual objective 0, we have the obj ective ' 0 is a fact', which will guarantee its factuality, if any superordinate objective can do so; if we protest that this guarantee is spurious, since the guaranteeing obj ective is itself a lying witness, we can find another objective 'It is a fact that O is a fact' which guarantees the second objective. And so we can produce an infinite series of objectives, all of which will endow an objective with factuality, if this lies in the power of superordinate obj ec­ tives. There will then be no difference in the pedigrees of factual and unfactual objectives, which will therefore be indistinguishable. Meinong meets this difficulty by holding that no obj ective merely becomes factual through, or is factual by virtue of superordinate obj ectives ; it must carry factuality in itself, indeed, as far as I can see, it must carry it as a fundamental property which admits of no definition, and, for the present at least, of no description. This property, whenever it adheres to an obj ective, is not only, so to say, complete in itself, but it also implies the factu­ ality of all positive superordinate obj ectives of being. 1

This statement involves the difficult and almost unintelli­ gible contention that it is 0, the factual obj ective, which makes the factuality of O a fact, and not the latter which gives factuality to 0. It is clear that there is something in a factual objective which our thought cannot give or take away; we may suppose that what is not a fact is a fact, but the factuality that can be acquired in this way is deficient in some vital element, an element which in I

0.

A.,

PP · 70, 7 1 .

T HE T H EORY O F O B JECT I VES

77

another treatise Meinong has called the modal moment. 1 We shall discuss the doctrine of the modal moment in a later chapter; it is quite probably the most vulnerable point in the theory of objects. VII

Meinong mentions, as a fifth distinguishing mark of objectives, their peculiar relationship to time. This rela­ tionship they share with all entities whose being, if they have any being, takes the form of subsistence. 'Subsistents distinguish themselves from existents in several respects, among which is the circumstance that they are not bound down to any definite time, and are in this sense eternal or rather timeless.' 2 Since objectives are incapable of any sort of being other than subsistence, it follows that all objectives are timeless. This statement admits of no doubt in the case of those objectives which concern timeless subsistents: thus the diversity of red from green, or the equality of two and two, are facts whose factuality is unaffected by the course of events in time. There are also facts of so-being about certain objects which have no relation to time; the yellow colour of the proverbial golden mountain or the roundness of the round square are cases in point. But, if we turn to objectives concerning entities which exist in time, Meinong's position becomes more questionable. Con­ sider the fact that my writing-table exists; this objective is certainly a fact at the present time, but it seems as certain that it was not always a fact and that it will not always be a fact. During the greater part of the past no such object as my writing-table existed, and in the greater part of the future no such object will exist. It looks, there­ fore, as if certain objectives go through a period of un­ factuality which has no beginning though it has an end; 1

Mog., p .

266.

2

U. A., p.

64 .

78

THE THEO RY O F O B J E CT IV E &

this is succeeded by a period which has both a beginning and an end, in which they enjoy factuality ; then a second period of unfactuality commences to which there is no end. These facts seem fairly obvious, but Meinong attempts to circumvent them by various devices which have been frequently used since the time of Aristotle. The statement 'My writing-table exists' is not, according to him, a com­ plete verbal formulation of the objective it refers to ; our meaning would be more satisfactorily formulated if we said : 'My writing-table exists at the time t'. Temporal dates are, according to Meinong, a set of absolute deter­ minations, though we usually apprehend them in their relation to the date at which our apprehension occurs. We regard this date as specially privileged because we happen to be there ; it is dignified with the title of 'present', while all later and earlier dates are referred to as 'future' or 'past' respectively. There is, however, nothing objective in such distinctions : we are intruding into a statement about non­ mental objects certain irrelevant statements about our­ selves. Meinong writes : I t is of course absolutely indifferent to a certain astronomical constellation, whether any one apprehends it, and whether his appre­ hension takes place simultaneously with the constellation or pre­ viously or subsequently . . . . The constellation certainly exists at a definite time t, and at another time t' it does not exist ; but none of these times is in any way present, past or future ; the one is simply t, the other t', however easy or difficult it may be for us to apprehend them in their intrinsic nature ( Eigenhestimmtheit). 1

In another passage he also speaks of the unj ustifiable intrusion of a completely subjective moment into our notion of existence, which reveals itself as past, present or future ; such a determination is nothing but a relation between the date of the j udgement and the date of its obj ect, which is j ust as 1

0. A., pp . 76, 77.

T HE T H EORY O F O B JEC T I VES

79

accidental to the reality, as it is accidental to it whether it is known by any one at any given time. 1

The theory outlined here is identical with that of Mr. Russell in his article entitled 'On theExperience of Time'. 2 Mr. Russell also holds that distinctions of present, past, and future are merely relative to the date of the judging observer. Meinong thinks that the meaning of the word 'exis­ tence' is too deeply tainted with a reference to the stand­ point of the judging subject; we should, therefore, in strict discussions, employ a more neutral word. He sug­ gests that we should say that an obj ect persists, or that it persists at a certain time t ; whatever the temporal point of view from which we approach such an obj ect, it will still be the case that it persists at that time. Thus Pheidias's statue of Athene persists at 400 B.c.; its persistence then is a fact now and at all other moments in the history of the Universe. vVhen we say that its persistence at 400 B.c. is a fact at other times, we mean that it is a fact for any experient who apprehends it by a judgement at those other times. Meinong says: I f one takes the trouble to think out clearly the notion of an obj ective, it seems to me that it becomes absolutely evident that temporal dates have throughout the character of obj ecta and not of obj ectives, and can therefore by their very nature never be predicated of obj ectives. Moreover there are adverbs of place, and yet hardly any one would be doubtful that existence or subsistence cannot have a place.3

Meinong means that, if it is a fact that snow is cold, then snow will be cold in the fires of Etna or on the molten plains of Jupiter. Wherever we transport the judging subj ect, it will make no difference to the facts that he apprehends by his judgement. A fact seems to be a fact 1

J

Geg. hoh. Ord. Gs. Abh. I I, 0. A., p. 6 7 .

p . 457.

2

Monist,

1 9 1 5.

80

T HE T HEORY O F O B JECT I VES

in every place, because it is, strictly speaking, in no place; similarly, a fact seems to be a fact at every time, because it is, strictly speaking, at no time. The objecta which facts concern may have places and dates, and the facts may con­ cern those places and dates, but, as facts are never identical with their material, they are without place and date. In one of his latest treatises, Zum Erweise des allgemeinen Kausalgesetzes, Meinong modifies this rather unplausible view of the relation of objectives to time, and concedes that there is an absolute sense in which the present is privileged above the future and the past. Whether an object or event is present, past, or future has nothing to do with any one's judgement on the matter. We call objects which exist now actual (wirklich), because they are able to act (wirken), whereas objects that persist at dates in the future or the past have ceased to be, or are not yet actual. Temporal determinations, with which certain events and objects are . associated, persist timelessly, but they do not always enjoy the privilege of existence. 1 There must therefore, on this later view of Meinong's, be two varieties of objective, those that are and those that are not affected by the passage of time. That the sun is going down, or was going down, or will be going down at a certain time, are not always facts, but it is a fact at all times that a particular sunset persists at a particular date. For every fact in whose material a tense is included, and which is therefore affected by time, and need not always be a fact, there is another tenseless fact of persis­ tence which is always a fact. Meinong therefore gets rid of the specious, but intolerably superficial view of time to which most scientific thinkers become victims. As no objecta known to us are timeless, and as some objectives which concern temporal existents have a timeless factu­ ality, there is still a very great gulf between objectives and objecta. 1

Erq)J. Kaus., pp.

60,

77, 7 8 .

THE THEORY OF OBJECTIVES

81

VII I

There remains a sixth feature of obj ectives which distinguishes them from all objecta. This is not clearly stated by Meinong, but seems to be implied by what he says. According to Meinong, objecta can only be given to our judgements or assumptions by means of obj ectives which concern them. 1 True, objecta can be presented to us directly by Vorstellungen (ideas), but in such presenta­ tion we are wholly passive, and do not 'make' anything out of the presented objecta. 2 We then come to an impor­ tant point, concerning which Meinong is very positive. 'Negation is never a matter for the mere idea, although of course it will never occur without an idea; it follows that wherever we find a negation we have passed beyond the boundaries of the mere Vorstellung. ' 3 Meinong is here basing himself on a very profound instinctive conviction which every one shares; one cannot see a negative in the way that one can see a positive, it does not force itself upon us in the same startling and im­ mediate fashion. We have to exert a certain amount of intellectual activity before it will reveal itself. It is this conviction which has led some people to adopt the view, which we saw to be mistaken, that there cannot really be negative facts. In the case of intuitive ideas, i.e. ideas whose obj ects are given 'sensuously' or 'imaginatively', it is clear that we cannot see absences. If we attempt to picture a knife without a blade, we may set before our minds a picture of a knife-handle, but there is no visible object in the picture which corresponds to the lack of a blade. 4 If we try to form an image of a musical tone which is not loud, we may set before our minds a very soft tone, but its falling-short­ of-loudness is not represented in the image. We may even 1

3

824187

0. A., p . 47· I bid ., pp. 9 , 2 5 6 .

2

4

G

Ibid., p . Ibid ., P ·

227. II.

82

THE TH E O RY O F OBJE CTIVES

set a loud tone and a soft tone side by side, and i n the act of comparison become aware of their diversity; this in­ volves an active attitude of the mind, but not the attitude which apprehends a negative. The diversity between the soft note and the loud note is a wholly positive circum­ stance which concerns both notes; it is quite different from the lack of loudness which concerns the soft note alone. 1 Meinong considers that we are only aware of negatives because at some time in our life we have actively denied something, that is, experienced a negative judgement or assumption. As long as we neither judge nor assume, the most incompatible obj ects will lie peacefully side by side. When we actively rej ect a certain suggested alternative, our rej ection, being an act of apprehension, and not merely an emotional reaction, reveals to us an obj ective of absence, whether such absence be a fact or not. It follows that all apprehension of negatives is via obj ectives; all negative properties of obj ecta, e.g. deafness, blindness, colourless­ ness, &c., can only be given to us in so far as we become aware of the fact that these obj ecta are not able to hear, or see, or have no colour. 2 It seems clear that the facts adduced by Meinong are of importance, not only for psychology, but also for the theory of obj ects. If there really were negative obj ecta, there could be no good reason why they should not be given to us immediately by simple ideas. The fact that we can only arrive at them by means of obj ectives seems to indicate that negativity is entirely a matter of the ob­ j ective. We cannot regard the absence of an obj ectum as if this were another obj ectum. This is quite clear in the case of existents; my present state of mind exists, but if it did not exist, we could not regard the void created by its absence as another existent. There is, however, a tendency to treat the lack of a property as if this were a particular sort of property; thus we attribute deafness or colourlessness I

0. A., p.

14.

2

Ibid . , p . 2 7 5 .

THE THEORY OF OBJECTIVES

83

to an object in the same way that we attribute loudness or redness to it. On the view that we have suggested, deafness or colourlessness are not objecta like loudness and redness. The property of deafness is really a relational property possessed by an objectum because it is the material of a negative objective; inasmuch as X is the material of the negative objective ' X cannot hear', we say that X has the relational property of deafness. This rela­ tional property is not negative ; it is a positive relationship to the only sort of entity that can be negative, i.e. an objective. We conclude therefore that objectives are differentiated from objecta in that the former admit of the positive-negative antithesis, whereas the latter do not. That objecta are never negative is also clearly implied by a passage in Ober Moglichkeit und Wahrscheinlichkeit. 1 Here Meinong is considering the zero of various sorts of intensive magnitudes, e.g. the intensity of a sound. He says : 'If zero here were a negative object, then we must in that case be dealing with an objective. ' But he says that he cannot believe that the zero of a scale of intensities can be an objective while all other degrees on the scale are objecta. He therefore comes to the conclusion that there must be some positive objectum at zero, 'that is by its nature inaccessible to our perception, but is indirectly characterized by the fact that it necessarily involves an objective of negative quality'. Whether there is such an objectum as Meining supposes is irrelevant ; the impor­ tant point is that he maintains that what we apprehend must be an objective because it is negative, whereas the hypothetical objectum that is inaccessible to our percep­ tion is held to be positive because it is an objectum. IX

It will be convenient at this point to consider a rather misleading conception of Meinong's doctrine which is i Mag . , P · 9 4 ·

84

T H E T H E O R Y O F O B J E C T IV E S

due to Mr. Russell. In his article entitled 'Meinong's Theory of Complexes and Assumptions' 1 Russell writes: 'This objective of the judgement is what (following Mr. G. E. Moore) I have called a proposition: it is to the objective that such words as true and false, evident, prob­ able, necessary, &c., apply.' Now Meinong's objectives and the Russell-Moore propositions of 1 904 have one thing in common, they are the objects to which judgements 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. Thus he remarks: 2 The validity of a truth . . . is in reality no kind of being at all. The truth of a proposition consists in a certain relation to truth, and presupposes the being of the proposition. And as regards being, false propositions are on exactly the same level, since to be false a proposi­ tion must already be. Thus validity is not a kind of being, but being belongs to valid and invalid propositions alike.

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 subsisting or exist­ ing. According to Meinong, only those objectives subsist which are facts: factuality is their specific modification of being; there are an infinite multitude of objectives which do not subsist. This last statement is rather loose, for, strictly speaking, there are no such objectives; what we should say is that these objectives are objectives, and that they are judged and assumed by such and such people. There is a more serious objection to the identification of an objective with a proposition: it is extremely un­ natural to say that any propositions are facts, and on Meinong's theory some objectives are facts. It is true t

Mind, vol. xiii, r 9 04, p. 3 50.

2

Prine. of Maths . , p . 45 0.

T H E T H E O RY O F O B J ECT I V E S

85

that Mr. Russell does i n one passage identify true pro­ positions with facts. He writes: The fundamental obj ection [to his own theory of truth and false­ hood] may be simply expressed by saying that true propositions expressfact, 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. 1

It must be conceded, however, that hardly any one 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. Thus Mr. W. E. Johnson says that 'any pro­ position characterizes some fact, so that the relation of proposition to fact is the same as that of adjective to sub­ stantive'.2 He also suggests that a judgement is true when the proposition in which it is expressed is in accordance with a certain fact, while any proposition in discordance with that fact is false. 3 Similarly we find in a paper by Mr. Ryle entitled Are There Propositions.? the following passage: 'A true proposition has to be defined as one which is in some sort of relation of correspondence to a reality or fact other than itself; and as we can think a pro­ position without knowing it to be true . . . in order to know it to be true we should also have to know the fact that it corresponded with.'4 Again Dr. McTaggart, in attacking the theory of propositions, writes: 'If the belief is true, it will correspond to the fact. It may also correspond to a true proposition, but then it will have two correspondences -to the true proposition and the fact.' 5 1

'Meinong's Theory', p. vol. r , p. 1 4. Ibid., p. I 6.

� Logic, 3

4

s

523·

Aristotelian Society Proceedings, 1 929-30, Nature of Existence, vol. 1, p. 1 4 .

pp. 1 06-7.

T H E T H E O R Y OF O B J E C T I V E S

86

Many other quotations might be givento show that those philosophers who have discussed a theory of propositions have regarded them as entities which might correspond to or accord with facts, but never as entities which could be identical with facts. As some of Meinong's objectives simply are facts, it is clear that it is wrong to identify objectives and propositions. And, if we consider ordinary linguistic usage, it is utterly impossible to substitute the word 'proposition', or even 'true proposition', for the word 'fact' wherever the latter occurs. We can say that a man's death was due to the fact that he forgot to turn off the gas, but we cannot say that it was due to the proposition that he forgot to turn off the gas. Yet according to Meinong objectives are involved in any causal connexion; 1 thus he remarks that Boyle's Law does not connect the qualities of air with the state of the thermo­ meter and the barometer, but the objectives which consist in the being of these qualities and states. It is clear that these objectives cannot be propositions; it is nonsense to say that the attainment of a temperature of 67° Centigrade by a certain gas is a proposition. It will clear up the difference between the doctrine of Meinong, and the views of those philosophers who believe in propositions, if we contrast their theories of truth. On the theory of propositions there is always a proposition before our minds whenever we judge or make some as­ sumption. Through this proposition we refer or profess to refer to some fact; that is, we claim that there is some fact to which our proposition corresponds. If the proposi­ tion really corresponds to a fact, it and the state of mind directed to it, are said to be true, if there is no such corre­ spondence, the proposition and the state of mind accom­ panying it are said to be false. The theory throws no light on the intrinsic difference between the proposition and the fact; how the-being-green-of-the-grass can suffer diremp­ tion into two entities, one of which is a proposition and the 1

0. Gegth. Gs. Abh. I I, p. 48 7 ; E rrw. Kaus., pp. 44, 45.

T HE T HEORY O F O B J ECT I VES

87

other a fact, is left utterly obscure. Surely the-being-green­ of-the-grass must be the same entity whether we regard it as an object of belief, or as the referent of a set of symbols, or as a circumstance in the mind-independent world ? On Meinong's theory there are no entities between our minds and the facts. 1 A proposition (Satz) is merely an objective regarded from a particular point of view ; we are considering it only in so far as it is formulated verbally, and leaving aside the question as to whether it is true or a fact. 2 Truth and falsehood are held to be properties of objectives, not of objecta or experiences ; in particular truth and falsehood are not properties of judgements. 3 But they are not properties of objectives considered as pure objects, but only in so far as they are apprehended by judgements or assumptions. Meinong writes: We come to the conclusion, then, that truth can only be predi­ cated of obj ectives, if, all other circumstances being favourable, they are being considered as the obj ects of suitable experiences. What somebody maintains or controverts, believes or disbelieves, surmises or merely supposes, may with complete naturalness be called true. I n this way truth is the property of obj ectives used in apprehension ( Erfassungsobjektive) ; inasmuch as the apprehending experience exists they may also be called pseudo-existent . . . . To sum up we may say obj ectives are true in the most natural sense of the word when (a) they are used in apprehension, and (b) they also possess factuality. 4

Meinong points out the subjective tinge of words like true and false ; they are 'relative terms' which can only be attributed to an objective in reference to some experience. Though it is most natural to regard them as properties of objectives, it is easy to shift them over to the experiences 1 This does not mean that every one who apprehends a factual objective, e.g. the exponential theorem, will apprehend it as a fact. This will only be the case if his apprehension is accompanied by the indescribable experience of evidence. Where evidence is absent we have surmises, assumptions, and j udge­ ments which may per accidens be correct, but which are not knowledge in the full sense. 4 Ibid., p. 40. J Mag., p . 3 8 . z 0. A., p. 1 00 .

88

THE T H E O R Y OF O B J E CT IV E S

involved, and such a shift in meaning does not matter very much. 1 Meinong's theory of truth is therefore a theory of identity or coincidence. The same objective which is factual is also pseudo-existent, that is, reveals itself in a certain judgement or assumption, and the conjunction of factu­ ality and pseudo-existence makes the objective true. There is no entity which is true by virtue of a correspondence with fact; the fact itself is true in so far as it is the object of a judgement. It is possible for Meinong to call a fact true because for him the word 'fact' is not a mere name, but involves a description. Facts are not an ultimate sort of entity, but are those objectives to which an ultimate characteristic, factuality, belongs. Hence an objective which has factuality, or which is, in common parlance, a fact, may very well also be true. Meinong admits, however, that it is possible to give a partial justification to the correspondence theory of truth if we distinguish between (a) an apprehended objective, and (b) the object of a given apprehending experience. 2 In (a) we are dealing with an objective as an objective, and its property of being apprehended is merely external and accidental; in (b) we are dealing with the objective as an object of an experience. Now, if it is possible to distin­ guish (a) from (b), the pure objective per se from the objective as object of thought, we may then say that (a) exactly resembles or corresponds to (b). Meinong seems to be proposing that we should make a distinction between, say, Newton as physicist, Newton as theologian, and Newton as Master of the Mint. These three entities may then be thought of as corresponding, and not as identical. This sense of 'correspondence' is utterly remote from any accepted sense of the word; nobody who speaks of correspondence merely means that some object plays two parts, or can be described from two different points 1

Mog ., p. 4 r .

2

Ibid., p . 42 .

T HE T HEORY O F O B JECT I VES

89

of view. And no philosopher who has believed in truth as correspondence would admit that by his correspondence of two entities he meant no more than that a single entity functioned in two ways. 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 unfactu­ ality. Since Meinong attributes to his objectives so many properties that no one would dream of attributing to pro­ positions, 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. X

We have dealt with the reasons given by Meinong for supposing that there are such things as objectives, and the properties which, according to him, distinguish them from the things of common sense. We are now forced by the principle of economy to consider whether we cannot explain these entities away, by regarding them as symbolic disguises for something more familiar. We shall therefore go on to deal with various ways in which one might attempt to show that objectives are not 'genuine entities'; if these prove unsuccessful, we shall be left provisionally with the conclusion that Meinong really has discovered an irre­ ducible type of object. In doing this it may be convenient to have a less un­ usual name to refer to the things that Meinong calls objectives. There are two such verbal forms at our dis­ posal; the first is the word 'circumstances', the second the

90

T H E T H E OR Y O F O B J ECT I V E S

phrase 'state of affairs'. That China is a Republic, that there is no perpetuum mobile, that dirigible airships exist, all these may indifferently be called circumstances or states of affairs. The latter word corresponds closely to the Ger­ man word Sachverhalt, which Meinong considers as a possible alternative to his word Objektiv, 1 and which had been used in this sense by Stumpf. But he objects to this word because it would be strange to say of an unfactual objective that it was a Sachverhalt. That the diagonals of a square are unequal or that whales are not mammals, would hardly be called Sachverhalte though they are ob­ jectives. I am not clear that a similar restriction applies to the English phrase 'state of affairs'. We can easily say that the visit of a pope to England would be a strange state of affairs, though we do not for an instant suppose that such a visit is likely to take place. 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. We can speak of the circumstance that Goethe and Schiller were friends, or of a fictitious circum­ stance occurring in a novel. We may note here that Meinong frequently uses the German word Umstand, the equivalent of 'circumstance', in cases where 'objective' would be equally admissible, though he does not explicitly comment on such a use. 2 With a little stretching of usage one can accustom oneself to phrases such as 'the circum­ stance that the sum of the first n odd numbers is a perfect square', or 'the circumstance that five and two are nine'. 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 a sheer gain, for it rids us of the tempta­ tion to psychologize objects. Circumstances therefore fall into two great classes, those circumstances which are facts and those which are unfactual. By the use of a familiar 1 2

a.

A., P · r o r . See, for instance, V. A., pp. 59, 60, 6 1 , 2 5 2 ; Geg. hrih. Ord. Gs. Abh. I I, p. 3 9 3.

THE TH EORY OF OBJECTIVES

91

word many arguments which would otherwise appear frigid and unconvincing may assume a more obvious evidence. We shall now inquire whether it is possible to reduce circumstances to existents, or characteristics, or relations, or to any complex of these. That a circumstance should be identified with any characteristic of an object, or with a relation between two or more objects, is only possible if we use the words 'characteristic' and 'relation' very loosely. It might be suggested, for instance, that what is really meant by 'the circumstance that the table is black' is simply the property blackness which is one of the qualities of the table. That cannot be so, however, since the blackness of the table, as a mere property, is not at all different from the blackness of the coal-scuttle or the blackness of the stove. A person who directs this thought to what he would call the-being-black-of-the-table is not thinking of black­ ness as a pure property, but of blackness in a very special position, namely as a determination of the table which stands before him. 1 This is even more clear if we consider the case of a subject who not only apprehends, but also takes pleasure in certain objects. To take pleasure in blackness as such, is not necessarily to take pleasure in the circumstances that this table is black. One might very well find blackness an agreeable quality and yet be pained by the fact that it occurred in this table, because it was unsuitable to the wood, or made the room too sombre, or because one was pained to find any instances of pleasant qualities. Again, one might be pleased that this table was black while one detested blackness, because one thought that blackness was suitable to the table or showed up by contrast other agreeable qualities in the environment. It is quite clear, then, that by the phrase 'that the table is black' we do not merely mean the characteristic blackness 1

Cf. V. A., P· 5 7•

92

THE THEORY OF OBJECTIVES

which is manifested in the table. I t might, however, be maintained that there were such things as particularized characteristics, that there was a sense in which this table possessed its own particular blackness, which was not the blackness of the coal-scuttle or the stove. Even if this were admitted, it seems to be impossible to identify what we have called the circumstance that the table is black with the particularized blackness which it and it alone possesses. For again one might be pleased with this particularized blackness as an instance of an excellent colour, and yet be pained by its presence in the given table. These same arguments apply mutatis mutandis to any attempt to identify a relational circumstance (or objective) with a relation. The circumstance that A is similar to B might be pleasant to a person who was fond of B, although similarity per se might give him no pleasure whatever. And, if there were such things as particularized relations, that is, if the similarity between one pair of similars were different from the similarity between another pair of similars, it would nevertheless be possible for a person to find the particularized similarity which subsisted between A and B quite an uninteresting object in itself, while he was very pleased by the fact that it held between A and B. We have now to ask whether such phrases as 'the-being­ black-of-the-table' can possibly mean the concrete existent table and nothing else, the sentence-form being a mere verbal disguise. Such a view would not account for cases where a person is delighted by the circumstance that the table is black, while he takes no pleasure in the table as a whole. There might be many disagreeable circumstances concerning the table-it might be of vulgar design, or of poor wood, or given to creaking, and his general attitude towards the table might be one of dislike. This need not affect the unmixed pleasure he takes in its being black. There is one case in which many thinkers have been tempted to identify what appears to be a circumstance

T HE T HEORY O F O B JECT I VE S

93

with the object it concerns. It might be suggested that 'the table exists' means exactly what is meant by 'the table', and that the word 'exists' merely gives expression to the fact that we are asserting or judging. This view was actually held by Brentano. 1 At first sight the distinction between a table and the existence of that table does seem a piece of hairsplitting. Yet, if we consider what we mean when we speak of 'the table', and when we speak of 'the existence of the table', it becomes clear that the latter is only one out of innumerable circumstances which concern the table. In the other circumstances what we call its 'nature' comes to light ; these are such circumstances as that it is brown, or heavy, or oblong, or made of cedar wood. These circumstances together make up what Meinong would call the so-being of the table. Now it is quite possible to know some of these circumstances and not to know others, to be pleased with some of them and not with others. The circumstance that the table exists is a perfectly distinct circumstance, to which perfectly distinct attitudes are possible. One might for instance be glad that the table existed, because one wanted the world of existence to comprise as many objects as possible, yet one might be quite indifferent to what sort of object it was. Again one might conceivably be pleased that the existent table had the properties it had, yet, owing to a nihilistic preference for the void, one might feel no joy in its exis­ tence. Such examples show that, however closely they are related, we can always distinguish between the circum­ stance that something of a certain sort is, and the circum­ stance that something which is, is of a certain sort. And once it is clear that the being of anything is only one circumstance among the innumerable circumstances that concern it, the temptation to identify it with the thing of which it is the being is weakened. 1

Psychologie, II. vii . 5.

T HE T HEORY O F O B JECT I VES

94

XI

We must now ask a much more difficult question, whether a circumstance is not possibly a complex into which various concrete objects, characteristics, and relations enter. Meinong frequently considers the objective from this point of view. Thus he says: 'The objective is in this way treated as a sort of complex, the objectum con­ cerned as a sort of constituent. From many points of view this agrees with our insight into the essence of the objec­ tive, which is at present so utterly defective. But that the analogy is only an aid in our initial confusion, and that we have no right to take it at all strictly, no one will deny. ' 1 Again he says: 'The objective does not stand apart from the objectum and beside it, but the objectum, in so far as the judgement apprehends it, always stands in an objective of which it forms a sort of constituent part.' 2 That an objective is hardly different from a complex was assumed by Mr. Russell in his article entitled 'Meinong's Theory of Complexes and Assumptions'. There he remarks: 'Complexes, as soon as we examine them, are seen to be always products of propositions: one might be tempted to describe them loosely as propositions in which the truth and falsity has been left out. ' 3 Again he remarks: 'On one view a complex is the same thing as a proposition and is always either true or false, but has being equally in either case; on the other view, the only com­ plexes are true propositions, and falsehood is a property of such judgements as have no objectives.'4 This theory of the fact as a sort of complex persisted in Mr. Russell's thought even after he had given us his belief in what he called propositions. Thus in the introduction to Principia Mathematica he says: 'We will give the name of "a com­ plex" to any such object as " a in the relation R to b" or 1

3

0. Gegth. Gs. Abh. I I, p. 493. Mind, 1 904, p. 346.

z 0.A., p. 47· Ibid., p. 5 I 2 .

4

T HE T HEORY O F O B JECT I VES

95

" a having the quality q" or " a and b and c standing in the relation S". Broadly speaking a complex is anything which occurs in the universe, and is not simple.' 1 The opinion which identifies facts (and propositions) with complexes is reflected in the ordinary logical practice of talking about the constituents of facts and propositions. In general we have an unfortunate tendency to treat any sort of many­ sidedness as an instance of the relation of a whole to its parts ; the analogy of a space which is made up of spaces colours our interpretation of practically any entity which exhibits distinct aspects. That Meinong cannot have meant his description of the objective as a sort of complex very seriously becomes clear when we consider his doctrine of complexes. Meinong considers a complex to be more than the mere collection (Kollektiv) of its constituents. 'If a and b are to compose a complex (Komplexion), that is, be parts of a whole, some connexion must subsist between them which makes them parts of a whole, i.e. they are constituents of a complex by virtue of a relation R, in which they stand to each other.' 2 Wherever certain objects are constituents of a complex they must also be terms of some relation or relations, which bind the constituents together and gener­ ate the complex. This fact is called by Meinong the Principle of the Coincidence of Parts (Partialkoinzidenz). Relations, however, fall into two great classes, according as they are capable of generating real complexes, i.e. complexes which exist, or are only capable of generating ideal complexes, i.e. complexes which merely subsist. Relations such as similarity, diversity, incompatibility, can bind a set of entities into some sort of unity, but, even when these entities exist, the complex generated by such a relation is incapable of existence. This seems sound sense; the similarities between Caesar and Cromwell, or the incompatibility between a republic and a monarchy, 1

Principia ed.

2,

p . 44 ·

2

Geg. hb"h. Ord. Gs. Abh. I I , p. 3 8 9.

T HE T HEORY O F O B J ECT I VES

do not generate complex existents Caesar-Cromwell or republic-monarchy, though they do bind these objects into some sort of unity. The complex produced by the similarity which holds between a and b is, according to Meinong, an ideal complex, 1 the word 'ideal' being used without any psychologistic implications. A relation which generates this sort of complex may be called an ideal rela­ tion. There are, however, some relations which, when they subsist between existents, generate complexes which exist quite as much as their constituents do. Such relations may be called real relations, and the complexes they form real complexes. Meinong gives as instances of such complexes the complex existent formed when a colour occupies a region. The coloured region or the localized colour is a genuine existent, and, on Meinong's view, its constituents are genuine existents too. Again a number of tones can form a chord, and this is also a genuine existent, and not a mere collection of tones. 2 The relation of a complex to an objective is treated in Ober Annahmen, where Meinong discusses the conditions under which we can be aware of complexes as complexes. Merely to be aware of the constituents of the complex, even if our awarenesses occur at the same time and stand in intimate psychological relations, is not to be aware of the complex at all. A togetherness of apprehensions is not the same thing as an apprehension of togetherness. Similarly, if we are aware of the complex, without being aware of its connexion with its constituents, we shall only see it as a vague whole, and not as a complex. 3 Thus an untrained ear has a complex of tones given to it, but is not aware of the complex as a complex, because it cannot discriminate the various notes which go to the making of that complex and see how they combine to form a whole. Meinong therefore comes to the conclusion that a complex 1 3

Geg. htih. Ord. Gs. Abh. II, p. 3 9 5. 0. A., p. 2 7 8 .

2

Ibid.

T HE T HEORY O F O B JECT I VES

97

can only be apprehended if, in addition to an unanalysed presentation of its separate constituents, we also are aware that these constituents are put together in a certain way. This is of course a fact, otherwise a factual objective. Only the person who clearly apprehends this objective by means of a judgement or assumption is aware of a complex as a complex. The relation of an objective to a complex is stated in the following passage : An obj ect of the form 'a and h in the complexion 1 C' only fails to reveal that an obj ective is necessarily involved in it because of the abbreviated verbal form in which it is expressed. B ut the presence of the obj ective makes itself apparent in verbal forms such as : 'a that, together with h, makes up a certain complex', or 'the complex that a and h compose', &c. Again 'relation R between a and h' also constitutes a complex state of affairs, as is clear from the Prin­ ciple of Coincidence ; it can in any case be treated here in exactly the same way as a complex : the phrase 'relation R between a and h' is really tantamount to the unabridged phrase 'R, that subsists between a and h', or 'a that stands to h in the relationship R'. I n general w e may say that, wherever a complex lies before us, an obj ective also lies before us as a constituent moment of it, and if any one wishes to apprehend the complex, he can only succeed if he also apprehends the obj ective. 2

In this passage it is perfectly clear that Meinong believes in two separate sorts of entities: the objective that a and b stand to each other in the relationship R, or the objective that a and b compose the complex C, and the complex C which is 'produced' by those objectives. Hence, though an objective is in some sense complex, it cannot be identified with a complex in Meinong's technical sense. It is also easy to see that the identification of an objec­ tive with a complex would have contradicted some of Meinong's previous statements. He holds, as we saw, 1 The word 'complexion' here means the condition of being woven to gether in a complex.

z -0. A., PP· 2 79, 280. 824187

H

98

T HE T HEORY O F O B J ECT I VES

that some complexes exist; the chord which is made up out of several tones exists as much as they do. Yet the objective or fact that those tones occur simultaneously, have certain intervals between them, and are somehow fused together, is an ideal object, which can only subsist but never exist. We may illustrate Meinong's point with a ludicrous example: a sandwich is a complex generated by the fact that there is a layer of butter between and in contact with two slices of bread. Yet there is an unbridge­ able gulf between the complex and the fact; we can eat the former but not the latter. Similarly a house is a complex generated by the fact that certain bricks stand in certain spatial relations to each other; yet we can inhabit the house while we cannot inhabit the fact. There is, of course, the view expressed by Mr. Russell that apparently complex existents such as Socrates, Piccadilly Circus, Twelfth Night, are not genuine entities at all, but merely classes of exis­ tents related in certain ways. 1 We shall discuss this doc­ trine in the fifth chapter in connexion with Meinong's doctrine of objects of higher order. For the time being we may say that, if Meinong and common sense are right, and such complex objects as sandwiches, houses, and per­ sons do exist, they are utterly distinct from the facts that the parts of those complex objects are combined in a cer­ tain manner. In any case we feel there is some a priori absurdity in holding that an entity like a fact is literally made up out of concrete things, characteristics, and relations. That all the lumps of snow in the universe are literally parts of the fact that snow is white, that the solid Athene of Pheidias is really a part of the fact that there once was such a statue, that the blazing bulk of the sun really enters into the fact that the sun is larger than the earth: suppositions of this sort seem crude and materialistic once we accept the 'ideal' nature of facts. 1

Monist, I 9 1 8, p.

5 I 2.

T HE T H EORY O F O B JEC T I VES

99

If we turn to particular types of fact there are further difficulties in the complex theory. We should have to hold, for instance, that the fact that it is not snowing had a special constituent corresponding to the word 'not' . What­ ever the true account of the fact that it is not snowing may be, it is hard to believe that it consists of the objective 'It is snowing' plus a negative. The point about the negative is that it excludes 'It is snowing' from being and sets in its place a certain absence ; the objective and the negative do not acquiesce quietly in each other's being as the complex­ theory suggests. Any one who walks out of his house ex­ pecting to feel the snow falling, and who then experiences the absence of snow, will realize that this circumstance has a peculiar simplicity, which forbids him to consider it as made up of parts. Of course it is a many-sided circum­ stance, and may be approached from many angles ; that it is a negative circumstance is one side of it, that it con­ cerns the falling of snow is another side of it. But this indefinable many-sidedness does not mean that it is a com­ plex of obj ects. We may therefore maintain tentatively that while the objective 'It is not snowing' is a negative objective, there is nothing in its material corresponding to the word 'not'. A similar treatment might be applied to the words 'existence' and 'being', and will solve many metaphysical puzzles. If such views are correct, there will be an analogy between the way in which a state of mind 'points' or refers to objects, and the way in which an objective 'concerns' its material. Just as an object is never a part of a mental state, though it may be the whole essence of the state to refer to that object, so, in the case of ob­ jectives, it will be their nature to concern themselves with many objects without prejudice to their ideal simpli­ city. We have therefore considered the possibility that what we call 'objectives' are verbal disguises for characteristics or relations or concrete realities, or for complexes of these.

1 00

T H E T H E O R Y O F O B J E C T IV E S

In all of these cases the reduction met with difficulties. It therefore seems likely that objectives are genuine entities with a peculiar status in the universe. XII

We have not, so far, discussed Mr. Russell's view that 'a judgement does not have a single object, . . . but has several interrelated objects'. 1 If I understand this view rightly, Mr. Russell means that when I say that I believe that the Atlantic Ocean is blue, there is no such object before my mind as the blueness of the Atlantic Ocean. All that is before my mind are the two objects, the Atlantic Ocean and blueness (with the possible addition of the characterizing relation). These objects do not form a unity by themselves, but only in conjunction with my mind. Thus blueness is believed-of or predicated-of the Atlantic Ocean by me. I cannot see that this view has plausibility. When I consider the blueness of the Atlantic Ocean, I am not considering two separate objects, but a unity of some sort. And it will not do to say that the objects form a unity by virtue of a mental act of believing or predication or 'syn­ thesis', because the sort of unity which I am thinking of is not a mental unity at all. I am considering the Atlantic Ocean and blueness united in a purely objective way, in such a manner, namely, that blueness is a property of the Atlantic Ocean. It may, of course, be true that in order to have this fact or supposed fact before my mind I must perform a synthetic act of believing or predicating blue­ ness of the Atlantic Ocean. If, however, this connexion between the Atlantic Ocean and blueness by my act of believing is not merely a pleasant pastime, but an act which knows or claims to know, it must have, what all knowledge has, ' die allgemeine Eigenschaft iiber sich hinaus zu "transzendieren" '. It is not anything which our minds 1

Principia, ed.

2,

p. 43.

T H E T H EO R Y OF O B J E C T I V ES

IOI

have done or are doing which is given to us when we judge that the Atlantic Ocean is blue, but simply a physical circumstance. And a physical circumstance will likewise be before us if we believe that the Atlantic Ocean is a sea of mercury, or that it is pink, or that it has never been navigated. It seems to me therefore that either (a) belief is never knowledge, but an arbitrary mental synthesis of objects, or (b) belief is a two-termed relation in which the object is a circumstance or objective. That we have a relation to an objective as a whole will of course not preclude us from having distinct intentions to the various objects which the objective concerns. Meinong brings these points forward when he explains, in the second edition of Ober Annahmen, why he abandoned the terms 'thetic' and 'synthetic func­ tion of judgement', which were used in the first edition. In that edition an objective of being, 'A is', was held to be apprehended by a 'thetic' judgement; an objective of so­ being, 'A is B ', by a 'synthetic' judgement. Such terms are misleading because 'it does not lie in the power of a person judging to posit objects. Perhaps it is the case that when I pass judgement on two objects A and B, and my judgement is of the form "A is B", there is really some sort of connexion between these objects, which is the result of my judging, and which I can destroy or reinstate by my own actions; but in general the word "synthesis" has not been used in this innocent fashion'. 1 Meinong is referring to the doctrine, characteristic in particular of Kantian idealism, that the relation of snow to whiteness, which it has in the objective 'Snow is white', is some sort of mental relation of being thought together by a mind. Mr. Russell's theory seems, on the epistemological side at least, to have all the difficulties of the Kantian 'synthesis'. 2 I 2

0. A. , P · 3 42 .

This section owes a great deal to Professor Stout's article entitled'Russell's Theory of Judgement', reprinted in Studies in Philosophy and Psychology, p. 239 et seqq.

IV T HE M O DA L M OME N T

I

I

this chapter we shall deal with a small but irritating problem, which provides Meinong's critics with formidable objections to his theory. We shall find in discussing it, that the conception of AujJersein, which we sketched in our second chapter, requires considerable modification . We pointed out in the previous chapter that it was one of the fundamental differences between objec­ tives and objecta that the latter have being only by virtue of objectives which concern them, whereas the former do not possess their being by virtue of superordinate objec­ tives. Mount Everest, as a pure object, is constituted by a set of determinations of so-being, and would be what it was even if it were not there at all; but it stands in an objective of being, which happens to be a fact, and is therefore brought into the realm of the actual . The posi­ tion with regard to the objective 'Mount Everest exists' is quite different; it is not a fact because there is an objec­ tive 'It is a fact that Mount Everest exists', but it is a fact in its own right. On any other assumption we should be faced with a vicious infinite regress; each objective would depend for its factuality on an objective of higher order, and there would be nothing to give validity to the whole crazy structure. 1 No doubt we can distinguish, even in the case of the objective, between its characteristic content, between what, to use a bad metaphor, it 'says', and the factuality of this content. If this were not the case, no fact could ever be N

1

0. A., pp. 70, 7 r ; Mag., p. 29 1 ; above, Chapter III, § vi.

THE MODAL MOMENT

103

given to our thought without our knowing that it was a fact. But even if a factual objective is distinct from its factuality, it is much more intimately bound up with this factuality than an objectum is with existence ; factuality is a genuine property of the factual objective, whereas existence is not, strictly speaking, a property of an objec­ tum which exists. Because it is a fact that Mount Everest exists, it is a fact that this is a fact, and so on ad infinitum ; the basic factuality of the first objective gives factuality to all superordinate objectives. II

In the last chapter we said that factuality was the kind of being which was peculiar to objectives. This statement, according to Meinong, is not absolutely accurate. Even an unfactual objective can possess being of a sort, though this being is itself unfactual. I can for instance imagine that it is a fact that 2 + 2 = 5 ; I have then before me, as the result of my assumption, a certain object, namely, the fact that 2 + 2 = 5 . Even if I judge that it is not the case that 2 + 2 = 5 is a fact, I have had to conceive the factu­ ality of 2 + 2 = 5 . There must therefore be a being or rather a being-the-case which even unfactual objectives can possess, and we must distinguish between the genuine being-the-case which we call factuality, and this strength­ less substitute. We shall translate Meinong's distinction between the mere Sein (being) of an objective and its Tatsachlichkeit (factuality), by saying that every objective may have a 'watered-down' (depotenzierte) factuality/ by virtue of objectives in which factuality pertains to it, but that only some objectives have a 'full-strength' factu­ ality. Meinong holds that there must be a factor, which he calls the modal moment, in which the difference between i

Mog., P· 2 9 r .

I 04

THE MODAL MOMENT

full-strength and watered-down factuality consists. Full­ strength factuality minus the modal moment yields watered-down factuality. Watered-down factuality plus the modal moment yields full-strength factuality. 1 It is indifferent to watered-down factuality, as a pure object, whether the modal moment is added to it or not; as Meinong says: 'Neither being nor non-being loses any­ thing of its peculiar character by being determined as unfactual: unfactual being remains what it is, namely, being, and the case of unfactual non-being is precisely similar.'2 It will throw light on this difficult doctrine if we turn from objectives to the analogous but simpler case of objecta. Meinong believes, as we saw, that the determina­ tions of a non-existent object belong to it in the same way as properties belong to genuine existents. The round square may be incapable of existence, but it is a fact as 'hard' as any other that the round square is both round and square. In his review of the Grazer Untersuchungen zur Gegen­ standstheorie, Mr. Russell made two important objections to this doctrine.3 The first is that, if we admit that the round square is really both round and square, we have taken from the principles of logic their unlimited validity. This is a serious objection, but it is dismissed by Meinong on the ground that no one would ever think of applying these principles to anything but the actual or the possible.4 Exceptions to logical principles which are confined to the realm of impossible objects are not an important limita­ tion of those principles.5 Mr. Russell's second objection causes more difficulty : if the round square is really round, then it is also true that the existent round square really exists. 6 Hence by purely analytical necessity we are able to demonstrate the existence 1

4

Mog., p. 266.

Ste/lung, i, p. 62.

Ibid., p. I I I . s Mog., p . 2 7 8 . 2

3 6

Mind, r 905, p. 5 3 2 . Mind, r 90 5, p . 5 3 3 .

T HE M O DAL M O MEN T

1 05

of an impossible object, which is a novel application of the ontological proof. Meinong meets this difficulty by denying that the analytic judgement, whose realm is that of objectives of so-being, can also be significantly applied to that of being. 1 As we saw above, the pure objectum is indifferent both to being and non-being; these distinctions come to it from outside, and are only to be found in the objectives which concern it. 2 It follows that, strictly speaking, we cannot make existence, or any other sort of being, part of the nature of an objectum. Meinong is here taking up a position similar to that of Kant; qua objecta a hundred non-existent thalers are exactly like a hundred existent thalers. I can, however, cum grano salis, as Meinong remarks, treat existence or being as if it were part of the so-being of an object, 3 but the grain of salt reveals its presence by the fact that this purely suppositious existence is merely a watered-down substitute for existence in the full sense. Meinong proposes to say that an object is existent (existierend) when it merely has the watered-down variety of existence, and that it exists (existiert) when there is no such watering-down. The God of Anselm is an existent and so is the wealth which lies before us in dreams, but neither exists in the full sense, because genuine exis­ tence does not belong to the sphere of so-being. The existence of the God of Anselm or the existence of the wealth in dreams are both existences in which the modal moment is lacking. On Meinong's theory there is a watered-down variety of subsistence as well; thus the equality which subsists between two and seven is a genuine object in whose nature subsistence has been included, but this is at best a sub­ sistence which has lost its sinews. Only the presence of the modal moment in the objective could transform it into a full-strength subsistence. 1

3

Mag., p. 2 78 . Mog., p . 2 8 2 .

2

0. Gegth. Gs. Abh. II, p . 493 .

T H E MODAL MOMENT

1 06

We may note here that the modal moment is present in all genuine possibilities as well as in all facts. 1 That it is possible for two lines to enclose a space, or possible for an organism like man to live in a vacuum, are in a certain sense possibilities, but their possibility is something watered down, in which the modal moment is lacking. The modal moment reveals its presence as soon as we pass from such 'possibilities' to the possibility that a triangle should have equal sides or that a human being should subsist on a purely vegetable diet. Meinong believes that possibility is a property which admits of differences of degree; in short, he identifies it with probability, pro­ vided that this word is shorn of its subjective associations. Every degree of possibility can be given to us with or without the modal moment. If I assume that it is very likely that a coin that I am tossing will come to rest on its edge I am apprehending a very high degree of possibility, but the modal moment is lacking. In the same way I should be apprehending a low degree of possibility minus the modal moment, if I were to assume that it is very unlikely that I shall ever die. Such watered-down possi­ bilities may be contrasted with the absolutely evident, measurable possibilities of the mathematical theory of probability. The modal moment is in all likelihood incapable of intrinsic variation; 2 but its addition to the various levels of possibility merely puts something genuine in the place of something spurious. III

Mr. Russell's problem is therefore solved: the existent round square is existent, but it does not exist, because its existence lacks the modal moment. Unfortunately a 'second wave' rises up to overwhelm Meinong's theory. Suppose I assume that the objective 2 + 2 = 5 has factuality 1

Mag., p.

266.

2

Ibid.,

pp. 2 66, 292 .

THE MODAL MOMENT

107

plus the modal moment, then it is clear that I am assum­ ing something more than that 2 + 2 = 5 has watered-down factuality. Shall we hold that the modal moment is itself capable of being watered down, that it too has a ghostly counterpart which requires a second modal moment to lend it full reality ? It is clear that this path leads to the infinite regress ; we should have an infinite series of strengthless modal moments, each appealing to another moment which was equally feeble. From this situation Meinong saves himself by holding that we cannot, by means of a judgement or an assumption, attribute the modal moment to an objective which does not possess it. 1 In Ober Annahmen he had laid it down that we have unlimited freedom in our assumptions; 2 we can assume that squares are round, or that seven is less than two, or that Graz is north of Vienna. In each case, by means of our assumption, we lift an entity out of the infinite abundance of Au/Jersein, and make it an object of thought. We can even assume that it is afact that squares are round, or that it is possible far two straight lines to enclose a space, and can thereby apprehend objectives endowed with a watered-down factuality or possibility. But the freedom of our assumptions is limited in one important respect: we cannot by any mental feat lift out of AuJJersein a fact that squares are round, or a possibility that two straight lines should enclose a space, in which the modal moment is present. The most fantastic and insane assumptions can present genuine objects, but the attempt to assume the presence of the modal moment where it is not present is necessarily abortive, and apprehends no object whatever. The posi­ tion of a person who attempted to make such an assump­ tion would resemble that of some one who attempted to think of the very thought he was thinking ; in both cases, according to Meinong, not even an impossible object would be presented but only a complete void. 3 2

0. A., p . 348.

J

See Em. Preis., pp.

22,

23.

THE MODAL MOMENT

I 08

There can be little doubt that, as far as mere assump­ tions are concerned, Meinong's theory is right. An as­ sumption is merely a contemplative experience, in which we set objects before us and look at them, but have no serious convictions about them. 1 Even if we assume that something exists or is a fact, it is clear that the existence or the factuality that lies before us is in some way deficient. We are moving in a sphere in which the modal moment can never be given to us, because the whole question as to the genuine existence or factuality of the objects before us has been left aside. Only when our attitude involves conviction, or the desire to attain conviction, can there be any hope of apprehending the modal moment. A mind that is capable of belief knows what is meant by genuine being, whereas a mind that is confined to assumptions can never know it. Meinong is, however, unwilling to think that mere con­ viction is intrinsically adequate to the apprehension of the modal moment. Even if I believe that the round square exists with the most insane tenacity, the existence which is before my mind is to some extent deficient. We cannot believe that the round square exists in the same way that we can believe that we are seeing black marks before us, or that red is different from green. Our belief is not, in the last resort, complete and absolute belief, because it lacks the moment of evidence, the peculiar, luminous accompaniment of certain judgements.2 A person who is convinced about something, and whose conviction is completed by evidence, seizes the modal moment in his object ; no other experience can seize anything but a simulacrum or a surrogate. The impossibility of apprehending the modal moment by any other means becomes clear when we consider that the modal moment is unique among objects because it is not indifferent to being ; it cannot be dissociated from the fact of its being there, like a mountain or a tree. The soul in the Phaedo could not be separated from life because it 1

See

Mog., p. 255.

2

Ibid., p. 2 56.

THE MODAL MOMENT

1 09

was the life-giving principle itself; in the same way the modal moment cannot be separated from real being be­ cause it is the very element whose presence turns a pure object into an object which is really there. We cannot therefore apprehend it in any way which would leave it an open question as to whether it was there or not. IV

A 'third wave' remains to be faced. We said above that, if an objective does not possess the modal moment, it is impossible to think of it as having the modal moment. This seems to be a self-contradictory statement, for, in the very thought of its impossibility, we have conceived the impossible thing itself. According to Meinong, we can think or believe that a round square is existent, in the sense defined above, but we can neither think nor believe that it exists. How can Meinong make such a statement without thinking of a round square which exists as well as of one which is merely existent ? He answers this question by pointing to the fact that we can often refer indirectly to something which is not, in the full sense of the word, present to the mind. If I think of the square root of 5 I am referring to a specific real num­ ber, but this number, being an infinite class, can never be apprehended in the way in which 5 itself is apprehended. It is the ultimate object of my reference, but it is only given to me as the X which stands in a certain functional relation to the number 5 ; in other respects its nature is not given to me. Meinong believes that it is in some such indirect way that we apprehend the difference between a round square which is merely existent and one that, as we say, exists. We cannot add the modal moment to the watered-down existence of the round square, but we can perceive the modal moment in the full-strength existence of other objects, the moon, for instance, and we can then

THE MODAL MOMENT

assume that the round square is in this respect like the moon. 1 When we speak of a round square which exists, as opposed to one that is merely existent, we find a feeble surrogate for the modal moment by assuming that the round square is, as regards the modal moment, in the same position as genuine existents. This doctrine is cumbrous and difficult, but not obvi­ ously fallacious. It is not wholly unplausible to suppose that full-strength factuality or existence is given to us in only a few facts or objects belonging to our inner life or to purely formal sciences, and that we refer to the inacces­ sible factuality of other objectives, e.g. those about elec­ trons, or other minds, or the past, by imagining them as bound up with this kernel of elementary certainties. The three waves have therefore been weathered, but the ship looks considerably less tidy in consequence. V

We may briefly consider a theory which has been put forward by Professor Mally, and which removes many difficulties in Meinong's theory, without abandoning the general standpoint of the theory of objects. Meinong took over from Mally the principle of the independence of so-being from being, and incorporated it into his theory of Auj]ersein. According to this principle it is a fact that the round square is round, that the man who squared the circle squared the circle, and so on. The constitutive determinations of an object must pertain to it at all costs, whether the object exists or not. As Russell observed, this principle has the disadvantage of limiting the validity of logical principles, and also gives rise to the puzzles for which Meinong has proposed such incredibly ingenious solutions. If such a theory can be dispensed with, we must certainly do so.

T H E M O DA L M O M E N T

III

Mally's theory rejects entirely the principle of the independence of so-being from being, and bases itself on a more careful analysis of the relation of the object to the objectives which concern it. In every objective an object is in some way determined, and the way in which it is determined is called a determination (Bestimmung). A Bestimmung is more or less identical with what is meant by the 'propositional function' of Russell; it is an incomplete or indeterminate objective. Thus 'being white' or 'being greater than two' are determinations; if we allowed them to determine definite objects they would yield objectives. On the view of Mally every determination determines an object, but not every determination is satisfied (erfiillt) by an object. The determination 'being two-legged and featherless' determines the abstract determinate 'featherless biped', which is usually called a 'concept', but it is satisfied by nearly every human being. On the other hand, the determination 'being round and square' determines the abstract determinate 'round square', but it is not satisfied by any object. The object which satisfies a certain determination is really characterized by that determination; thus it is a fact that Plato is two-legged and feather­ less. But the determinate of a certain determination need not really possess that determination. The round square is not really round, nor is it a square at all; the only properties it really possesses are those of being determined by a certain determination and by all its implications. The possession of these properties is, of course, not contrary to the laws of logic. Nor is it necessary to deny to the round square as a pure determinate every vestige of being; it is to be found among formally possible combinations, but its determinations can never be satisfied. On the theory we are examining, there will still be an ultimate distinction between factual and unfactual objec­ tives, but there will be no sense in which an unfactual objective can have a watered-down factuality. If I say:

112

T H E M O D A L M O ME N T

'It is a fact that 2 + 2 = 5 ', I apprehend the determinate of two determinations, that of being the equality of the sum of two twos to five, and that of being factual, but there is no object which satisfies these determinations. Similarly the existent golden mountain is the determinate of the determination 'being an existent golden mountain', but this determination is also unsatisfied. With these modifi­ cations the whole notion of AuJJersein changes ; it is no longer a realm of full-blown objects unfortunately deprived of being, but something more nearly analogous to a space of points, which may or may not be occupied by actual objects. The round square is not really a round square but a locus in this space, concerning which we can say a priori that it will never be filled. We cannot develop this important theory here; its interest for us lies in the fact that it allows us to dispense with the modal moment and its attendant perplexities. If there is no sense in which a golden mountain exists, or 2 + 2 = 5 is a fact, we do not require a modal moment to distinguish between genuine and spurious existence or factuality. 1 1 For the theory of determinates see Gegenstandstheoretische Grundlagen der Logik und Logistik, by E. Mally (Leipzig, 1 9 1 2) , pp. 64, 76.

V O B JE C T S O F H I G H E R O R D E R I

E saw

in our study of objectives that Meinong drew a fundamental distinction between two varieties of being, existence and subsistence. Such things as classes, configurations, simi­ larities, and facts subsist, whereas tables, chairs, and mental states exist. In what this difference lies it is im­ possible to say, but it impresses itself immediately on the minds oflaymen and philosophers alike, and is as irreducible as the difference between yellow and blue. 1 We may take as a typical instance of a subsistent the group which is constituted by the four nuts which lie before me on the table. This is undoubtedly a genuine object and not a state of mind : its apprehension may demand a collect­ ing activity of thought, 2 but such a mental activity is not part of the object apprehended. 3 I am not considering my state of mind, but a certain group of physical objects. This group has properties quite different from those of any individual nut : it is not brown and round as they are, nor will it yield to nut-crackers: at the same time it is four in number, which the individual nuts are not. In its rela­ tion to the actual world it seems to be in a more favourable position than the golden mountain or the philosopher's stone; there is some sense in which it is on a table which exists, whereas we could search the world in vain for a golden mountain. But, in spite of this, we cannot add it to the individual nuts A, B, C, and D as something which exists as each of them does.4 1 3

0. A., p . 7 4 • Gs. Abh. II, p. 474.

824187

2 4

Geg. hrJh. Ord. Gs. Abh. I I, p. 3 8 8 . Geg. hoh. Ord. Gs. Abh. I I, p. 395.

1 14

O B J EC T S OF H I G H E R O R D E R

The group is, as it were, the very condition of there being a plurality of nuts at all; if, per impossibile, it were not there, each nut would remain shut up in itself, and would have nothing to do with anything else. Such an assump­ tion is really meaningless: each object is essentially 'one among objects'. We cannot, however, regard this funda­ mental presupposition as itself an existent object, beside the plurality of objects which presuppose it, for then it would presuppose itself. But we can, when once the four nuts have been given to us as a plurality, reflect on what is involved in such a 'givenness': in this case, we shall become aware of the group of four nuts as something distinct from the nuts, though far less capable of forcing itself upon our attention. The similarities and diversities between existent objects, and the various orders and patterns into which they assort themselves, are in the same way implied in the existence of those objects and cannot be set beside them: the reflection which disen­ tangles them is also quite different from the perception which directs itself to a totally new object. The objects we have mentioned are not merely a pecu­ liar class of objects: if this were the case we should have no reason to distinguish their mode of being from that of other things. Parrots are very different from ideas, but this does not entitle us to distinguish between the sort of being which is peculiar to parrots and that which is peculiar to ideas. But, in the case of the group of nuts and each of its members, we see plainly that they are not only very different in nature, but that they enter into the world in different ways, that they 'are there' in different senses; for this reason, it is impossible to say of the group that it exists as its members do. 1 I have not reproduced Meinong's argument which occurs in 0. A., p. 74. There he says that an airship which exists now subsisted at some previous date. That it can have these different determinations at different times, proves that i t i s not the sort of object, but the sort of being that is i n question. Meinong believes that all existents subsist, whereas some subsistents do not exist. If there is any 1

O B JECT S O F H I G HER ORDER

1 15

There are other points of view from which the differ­ ence between existents and subsistents forces itself upon us: the first concerns their relation to time. Existents are necessarily bound down to given dates, whereas we cannot significantly associate such dates with subsistents. The similarity which subsists between two people who live together may be said to grow greater with the passage of time, but by this we only mean that at different sections of their history different relations subsist between them. That such similarities should subsist between objects having the nature of the people in question, is timelessly necessary. 1 With such necessity goes an accessibility to under­ standing or a priori evidence. There are of course certain entities whose subsistence is not intelligible, e.g. purely contingent facts, but in a large number of cases subsistence reveals itself to rational insight, without recourse to par­ ticular experiences. The equality between two groups, the distance between two points, the degree of similarity between two colours, are felt to belong inevitably to the entities concerned.2 But that such and such an object exists, or that, in the actual world, the existence of A is always followed by the existence of B, may serve as examples of things which rational insight alone can never discover. Based on the distinction between existence and sub­ sistence is the distinction between 'real' and ideal objects. An ideal object is in no sense of the word a mental object­ like any other object it has a nature which is independent of our knowing-but it is the sort of object which can subsist but is incapable of existence. 'Real' objects, on the other hand, are objects whose nature does not preclude them from existing; in this sense the golden mountain sense in comparing the difference between existence and subsistence with that between yellow and blue, then we must also believe in the incompatibility of these two sorts of being. To me it seems as nonsensical to say that an airship sub1 0. A., pp. 75, 76 . sists as that similarity exists. 2 See Geg. Mh. Ord. Gs. Abh. II, pp. 394, 395·

1 16

OBJECTS O F H IG H E R O RD E R

which never existed, and the Colossus of Rhodes which exists no longer, are as much 'real' objects as Mount Everest or the Arc de Triomphe. 1 The notion expressed by the word Realitiit is an ex­ tremely important one, but it is impossible to give the same meaning to the English word 'reality'; this could certainly not be applied to something non-existent. The distinction between the abstract and the concrete comes nearer to Meinong's distinction of the ideal and the real, for there is some sense in speaking of a golden mountain as concrete even if it doe� r ot exist. II

We have dealt with one important class of ideal objects, the objectives; in this chapter we shall be principally con­ cerned with other classes, those of relations, complexes, and characteristics. Meinong's doctrine on these subjects is very involved and does not readily fall into a coherent system. He holds, for instance, that some relations and complexes exist, and are real objects, whereas others merely subsist, and must therefore be classed among ideal objects. The position of characteristics is even odder; as pure characteristics divorced from particular embodiments, they subsist timelessly, but they exist in their instances. In his early psychological work on abstraction Meinong was chiefly concerned to prove the genuineness of analysis. When I distinguish two qualities A and B in a certain object, he insists that I am apprehending real differences and articulations in its nature; I am not introducing fictitious divisions or merely uttering empty words. This defence of analysis is a necessary foundation for Meinong's treatment of ideal objects, but it does not, by itself, throw much light on their nature. As the whole tendency to depreciate analysis and abstraction, and to regard it as in 1

See Geg. hoh. Ord. Gs. Abh. II, p. 7 7 .

O B J ECT S O F H I G H E R O RD ER

I I7

some way a falsification of the object, is much less pre­ valent to-day than it was when Meinong wrote, many of his earlier treatises have lost in interest. Every act of abstraction 'presupposes a plurality of elements in the content of the idea which is given to it; every act of determination must arrive at such a plurality'. 1 Abstraction, if it is not erroneous, does not break up an entity into elements which do not exist in it; it only throws a clearer light on elements which a more confused idea failed to reveal. If an object before us were absolutely simple there would be no place in it where analysis could find a footing; if then we seemed to discover elements in it, this could only be because we had substituted for the original simple object a complex of objects which stood in no intelligible relation to it. The theory that analysis is an importation of subjective distinctions into an original objective simplicity attributes the most astonishing creative powers to the mind. A certain set of physical stimuli awaken in us the experience which we call the hearing of a buzz: analysis supervenes, and we are able to disentangle a number of voices. On the theory in question, the mind operates in a wholly magical way: in spite of the fact that the physical stimuli are precisely the same, it is able to change a single object into a number of totally different objects. It is not even correct to speak of a change: there is only a substitution. The buzz disappears, and the voices appear, and this complete alteration in the object of perception is solely due to an act of mind. 2 Such a theory is not a priori impossible, but it is incredible because of its physical difficulties, and also because it fails to give any account of our lively conviction that the object which we are analysing is really the same object that we saw vaguely before we commenced our analysis. If analysis is creative in this inexplicable way, all investigation of the immediately given is perfectly hopeless. 1

Relth. Gs. Abh. I I, p. 75 ·

2

Psy. An. Gs. Abh. I, p. 3 r 7.

I I8

O B J ECT S O F H I G H ER O R D E R

Meinong assumes therefore that the sphere of the · judgement is narrower than that of the mere idea. Many objects may be given to us even where it is impossible for the existence of such objects to be acknowledged in explicit judgements. 1 Every idea has a certain 'weight', and by this Meinong means a tendency to evoke in us a judgement in which the object of that idea is recognized. 2 The idea of a loud noise, especially when it is intuitive, has a greater 'weight' than the idea of a soft noise, the idea of an uninteresting object less 'weight' than the idea of an interesting object. The whole subject is treated in modern psychology under the heading of 'attention', and need concern us no further. The important point is that analysis is not a creation or a substitution, but merely the bringing into the sphere of j udgement of objects which were formerly presented by ideas, but were not definitely acknowledged by the mind. Where such an acknowledgement is easily effected, the object appears to undergo no alteration; its moments distinguish themselves naturally before us, in fact seem to demand recognition. But, wherever objects are not easily accessible to the judgement, we may have the illusion of a substitution. If we analyse the timbre of a note into its constituent overtones, the timbre as a quality may disappear. 3 This can only mean that our ability to perceive the complex of tones as a complex has temporarily been paralysed; we cannot see the wood for the trees. But in no case does analysis replace the given by something absolutely new. III

In his article 'Abstrahieren und Vergleichen'4 Meinong defends the view that a plurality of characteristics are genuinely present in objects, as against the view of 1

J

Psy. An. Gs. Abh. I, p. 3 3 6. Ibid., p. 326.

4

z Ibid., p . 343 .

Gs. Abh. I, viii, p. 445 et seq .

O B JECT S O F H I G H ER O R D ER

119

Cornelius, who had attempted to rehabilitate the distinctio rationis of Hume. On this view we need not believe that a concrete object really has distinguishable elements, e.g. shape and colour, as the ordinary talk about characteristics suggests. An object may be perfectly simple and struc­ tureless, but, when we compare it with other objects, it will exhibit various resemblances to them; these resem­ blances, which are relational properties, are attributed to the object as its characteristics, and may become so closely associated with it as to create the illusion of an internal complexity. What we call characteristics are really dis­ tinctions of reason, which are carried over into the object because it is the meeting-point of a large number of resemblances. As against this revival of Hume's doctrine by Cornelius Meinong maintains that it is simply contrary to experience to hold that all abstraction involves the recognition of similarities. If this were the case, I should never be able to abstract from its setting some property that I had not experienced previously; only when I had seen a number of objects which were similar in a certain respect, would I be able to hypostatize and give a name to the respect in which they were similar. But experience gives no support to this demand. I am perfectly able to distinguish totally new characteristics, e.g. new shapes or melodic forms, from the familiar colours and tone-qualities with which they are bound up. 1 The case which Meinong gives involves an important principle, and is quite sufficient to annihilate the similarity­ theory. What are called the characteristics of objects are no doubt often arrived at by means of comparison, but we find that in some cases we arrive at a characteristic by noticing a similarity between two objects, in other cases by noticing a diversity. I may arrive at the characteristic 'red' by comparing a geranium with a pillar-box, but I may 1

Abst. u. Vergl. Gs. Abh. I, p. 4 5 1 .

1 20

O B J E CT S O F H IG H E R O R D E R

equally well discover it by comparing the geranium with its leaves. Unless we wish to adopt the far-fetched theory that by the characteristic 'red' we mean a class of relational properties, comprising one exact similarity and an infinite number of peculiar diversities, we must concede that we are dealing with a single term which stands in many relations of similarity and diversity. The fact that more than one relational route leads us to the statement 'This is red', forbids us to identify the object arrived at with any of the routes. Meinong also points to the complexities which the similarity-theory involves. We assume that two objects A and B can be similar in more than one respect; they may be similar in shape as well as in colour. To account for this state of affairs we may have recourse to the hypothesis that there are as many distinct sorts of similarity as, on the ordinary theory, there are distinct characteristics. Simi­ larity-in-respect-of-shape will be a distinct relation from similarity-in-respect-of-colour, and both will presumably be ultimate and unanalysable kinds of similarity. Meinong had long ago shown, in his early studies of Hume, that such a view had no theoretical advantages; to get rid of shape and colour in order to put shape-similarity and colour-similarity in their places does not simplify any problem. 1 But, quite apart from this, we have very great difficulty in observing any such differences in the relations in question; exact similarity ( Gleichheit), in particular, seems to be the same relation whatever the terms are between which it holds. That a difference in relations which is so hard to observe should be the foundation of such clear distinctions as those denoted by the words 'red' and 'blue', 'high' and 'low', seems incredible.2 We may also attempt to make the similarity-theory work by means of a construction not unlike that which Mr. Russell makes use of in his theory of number; this 1

Krit. Nom . Gs. Abh.

I, p.

60.

2

Abst. u. Vergl. Gs. Abh. I,

p. 45 8.

O B JECT S O F H I G HE R O R DER

121

will enable us to dispense with the assumption of many distinct similarities. On the Russell theory of number a class has a certain number if it belongs to a class of classes all of which are similar to a given class. The theory works because the relation of similarity between classes, using this word in the technical sense of one-one correspondence, does sort classes out into mutually exclusive classes of classes, which classes of classes will perform all the func­ tions usually assigned to numbers. We might attempt a similar reduction of characteristics. We might hold that the statement 'A has the characteristic P' really means that A belongs to a certain class of objects all of which are similar to a given object X, whereas 'A has the characteris­ tic Q ' means that it belongs to a certain class of objects all of which are similar to another given object r. But this construction will not enable us to distinguish between P and Q unless X and are wholly dissimilar objects; if they are similar in any way, then, since similarity is a transitive relation, the two classes will be confounded, and with them the two characteristics. Unfortunately we can­ not point to any wholly dissimilar objects, similarity to which will divide objects into a number of distinct classes. Classes can be wholly dissimilar, in Mr. Russell's techni­ cal sense, but all objects are more or less similar in the ordinary sense of the word. Consequently the whole con­ struction breaks down. 1 Meinong admits, however, that there is a small group of cases to which a theory of the Hume-Cornelius type is really applicable. If we speak of a set of objects as 'coloured' or as 'reddish', we do not mean that these objects have in common an ordinary characteristic called 'colour' or 'reddishness'. If there were such characteris­ tics, the nature of each object would be infinitely complex: an object which had a certain definite shade of red would also have the characteristic of being 'reddish-yellowish',

r

1

See Abst. u. Vergl. Gs. Abh. I, p. 4 55 .

O B J E C T S O F H I G H ER O R D ER

1 22

'reddish-bluish', 'reddish', and so on ad infinitum, with every wide and narrow range of shades. Such a theory is extraordinarily unplausible, and is also without experi­ mental confirmation. We can pay attention to the colour of an object and ignore its shape, but we cannot pay atten­ tion to the red-yellow component in an object which has a particular shade of red. 1 Meinong believes therefore that such words as 'reddish' or 'coloured', when applied to a number of objects, do not imply that those objects have a first-order characteristic in common, but that they, or rather their characteristics, have certain similarities. 2 ' X is coloured' means ' X has one or other of the determinate shades P, Q, R, S, &c.', which are ordered by their similarities. Colour is therefore one of two things: a characteristic of higher order which consists in the possession of one or other of a set of first­ order characteristics, or an incomplete or indeterminate characteristic, which is determined as regards its member­ ship of a certain class, but undetermined as regards its position in the class. In the next chapter we shall deal with the nature and functions of incomplete objects; we may observe here that they are incapable of being, but that they may be given to our thought, and may enable us to deal with the corresponding complete objects. There is no reason, therefore, to think that a certain specific shade of red is a highly complex entity, into whose constitution such elements as colour, reddish-yellowish­ ness, redness, &c., enter. The apparently simple qualities of our sensuous experience may really be as simple as they seem. According to Meinong there is one limitation to this principle: if a set of qualities are ordered in more than one dimension, they cannot be simple, but must have at least as many components as they have dimensions.3 Since 1 2

3

See Reith. Gs. Abh. I I, p. 76; Empfindung Gs. Abh. I, pp. 1 1 4-1 6. See Abst. u. Vergl. Gs. Abh. I, p. 480 et seq. Farbenktirper Gs. Abh. I, p. 5 1 5 .

O B JECT S O F H I G H ER ORDER

1 23

tones can vary in pitch and intensity, they must contain a pitch-component and an intensity-component to account for these independent orders. The distinction between the content and the act-element in mental states rests similarly on the fact that mental states exhibit an independent variability in two directions; 1 we can direct our minds in various ways to the same object, and to different objects in the same way. It is not easy to see how Meinong would apply this doctrine to the case of spatial positions, or, as he calls them, local determinations ( Ortsbestimmungen). 2 Since spatial positions are ordered in three dimensions each position must be a complex of three determinations; in a universe without absolute co-ordinates it is impossible to see what these could be. In any case it shows excessive con­ fidence in an a priori argument to hold that a characteristic must be complex, because it is a member of a class of characteristics which are ordered in more than one dimen­ sion, even when the characteristic reveals no inner com­ plexity to the most careful examination. It would be more reasonable to admit that a characteristic was many-sided, like a point which is the meeting-place of several lines, without holding that it was made up out of several distin­ guishable moments. IV

Since the distinctions at which analysis arrives are all genuinely 'there' in the object examined, Meinong began by regarding concrete objects as complexes of characteris­ tics, much as Locke and Hume thought of them as bundles of simple ideas.3 A concrete object may be literally made up of a certain colour, a certain shape, a certain spatial and temporal position, and other determinations, bound together in a peculiarly intimate way. 4 This doctrine leads 1 J 4

2 Reith. Gs. Abh. I I, p. 47. Em. Preis., p. 57. See e.g. Krit. Nom. Gs. Abh. I, p. 1 9 . See Phantasie Gs. Abh. I, pp. 2 07, 2 44; Geg. hoh. Ord. Gs. Abh. I I, p. 395.

OBJECTS O F H IG H E R O R D E R

1 24

to curious consequences. The first is that it is not really possible, in the strict sense, for two objects to have the same characteristic; it is only possible for them to have exactly similar (gleiche) characteristics. 1 The acanthus­ form that I meet with in a Greek tern pie is not the same as the acanthus-form in a Renaissance building, even if they are exactly alike; the same holds with regard to the tone quality of a fiddle on two separate occasions. The reason for this view is obvious. If the characteristic X is really a constituent of the complex XYZ, it cannot also be a constituent of a complex widely removed in place and time, and of a totally different sort. The relation which binds characteristics together into a complex concretum is very intimate, and is presumably transitive. If, there­ fore, X occurred in two complexes xrz and XVW, where rz and VW are space-time determinations, these space-time determinations would be bound together in the same intimate way as are X, r, and Z. We should there­ fore have one complex concretum XYZVW before us, which would involve incompatible space-time properties. We are therefore forced to hold that, if characteristics are constituents of the objects they characterize, they cannot be common to many objects. Meinong does not deal with the difficult consequence that, if the blue of every blue object is a distinct but exactly similar blue, every part of a blue square will have its own blue, and so we shall have before us in the square a very large, or even a transfinite number of blues. The second consequence of Meinong's complex-theory is the doctrine, to which he is remarkably faithful, that the characteristics of existents exist as much as they do. It would indeed be very curious to believe that a whole could exist, while the entities which composed it were non­ existent.2 Attributes come into existence and pass away 1 2

Reith. Gs. Abh. II, pp. 74, 1 3 2 ; Krit. Nom. Gs. Abh. I, p. See Geg. hi:ih. Ord. Gs. Abh. I I , p. 3 8 2 .

22.

O B J ECT S O F H I G H ER ORD E R

125

with the things they characterize. No one, says Meinong, would maintain that an attribute of a concrete object could continue to have being, when the object to which it was attached had passed away. 1 His original theory is there­ fore quite different from that of Mr. Russell, who holds that characteristics subsist, but are incapable of existence. At a subsequent stage of his thought Meinong was led to distinguish between two senses in which we may speak of a characteristic : when I utter the word ' green' I may either be referring to a characteristic as such, regardless of its occurrence in this or that object, or I may be thinking of it as the attribute of a particular object. In the former sense ' green' subsists timelessly, like similarity or diversity, and is not affected by the coming into being or passing away of objects that are green; in the latter sense, there are as many ' greens' as there are green objects, and each of these ' greens' comes into existence and passes away with the object that possesses it. Meinong seems to believe that a timeless green can acquire existence by being present in a concrete existent, of which it is a distinguish­ able but inseparable part. 2 The theory is very like that of Professor Moore in his early paper entitled Identity, who distinguishes between a characteristic and an instance of that characteristic. 3 At a still later period of his thought, Meinong held that a pure characteristic, such as green in abstracto, is not a complete object, and is as such incapable of either existing or subsisting. 4 Presumably only the attributes of existents have genuine being, and the charac­ teristics of which these attributes are instances have only a sort of derivative being in their instances. 5 1 2

Krit. Nam. Gs. Abh. I, p. 22. See Erfgl., p . 26 ; Mag., p. 1 6 9 .

1 90 1 , p. 1 0 3 et seqq. r 8 5. He there remarks that abstract characteristics are incomplete objects, and has said on p. r 79 that these are undetermined in respect of either existence or subsistence. s To this type of being Meinong gives the name of 'implexive being'. See next chapter. 3

4

Aristotelian Society Proceedings,

Mog., p .

1 26

O B J E C T S OF H I G H E R O RD E R

In his later work Meinong explicitly abandons the complex-theory of the concretum. We may quote a rather important passage from Ober die Erfahrungsgrundlagen unseres Wissens. Meinong says: Every one knows that we cannot have a colour without exten­ sion, and that we cannot have them together without all sorts of tactile and other qualities. This lack of independence in single qualities readily suggests the view, as I have discovered in my own case, that the natural independence and peculiar nature of 'sub­ stance' or 'the thing' lies in the fact that it is a complex of mutually dependent qualities. Only after a long time, when I considered the matter from the point of view of the theory of obj ects, did it become clear to me that such a conception completely overlooks the true, characteristic notion of 'the thing'. This notion makes itself evident when, without regard to other relations of dependence, we contrast the meanings of the words 'green' and 'something green' ( Grunes). It is important to note that, in this antithesis, the moment of simplicity or complexity is quite irrelevant, as is also the question whether, in the case of something complex, we must think of a single bearer of several properties, or of a complex of entities, each of which is the bearer of a single property. Any one who, for any reason, adopts the latter conception, and thinks he is putting pro­ perties together, when he is really putting things together and making a new complex thing, may easily imagine that he has built the thing out of its properties. 1

Meinong has here criticized, in a very fundamental way, the Locke-Hume theory of the 'thing' as a complex of simple 'ideas'. It is only possible to build a lump of gold out of such things as yellow colour, brilliancy, fusibility, &c., if we have already particularized these characteristics ; it is an instance of yellow which we bind up with an instance of brilliancy, and so on. If we do not substitute instances for abstract characteristics, we can only construct a wholly abstract object which might have being in any world. Even spatial and temporal positions, if they are qualities, as Meinong seems to believe, 2 might occur in many 1

E ifgl. , p. 2 7 .

2

Reith. Gs. Abh. I I, pp. 47, 50.

O B J E C T S O F H I G H ER ORD ER

1 27

worlds, and are powerless to individuate the bundle of abstracta we have before us. 1 'Something green' is not different from 'green' merely because it has other pro­ perties as well as green; it is different because we have passed over from an abstract logical 'nature' to an indi­ vidual instance of that 'nature'. In what this transition lies it is impossible to say, but it is quite plain that it has nothing to do with an increase of complexity. 2 It seems clear that if Meinong had developed this important idea all the perplexities of his doctrine of charac­ teristics would have disappeared. An instance of a charac­ teristic C may theoretically be quite as simple as C: it is not made up of C and a featureless substrate, or of C and other characteristics. Of the two hypotheses that Meinong mentions (a) that a many-sided thing is made up of in­ stances of its several characteristics, (b) that a many-sided thing is an instance of its several characteristics, the latter is infinitely more acceptable. A green, square, heavy thing, will not then be made up out of three pieces, 'green', 'heavy', and 'square', but will be an indivisible simple instance of three characteristics. Its parts will be the various individual segments into which it can be divided, not the properties of which it is an instance. We shall no longer require particularized greennesses and squarenesses which occur in different objects; one greenness and one squareness will fulfil their functions. The notion that objects separated in time and space share in one charac­ teristic as two houses share in a wall, will be eliminated, with all the difficulties that it occasions. We shall then be able to give up the unnatural doctrine that characteristics 1 We may note incidentally that the reason why positions in space and dates in time do individuate, is that they are in themselves individual, and not charac­ teristics at all. This place is not different from that place because there is anything i n its nature which distinguishes it from that place, but merely because it is this place and not that. The whole search for a principle of individuation is absurd; it is a priori impossible that anything which is not itself a 'this' should make •thises' out of entities which are not 'thises'. 2 E rfgl., p. 2 7 .

1 28

OBJECTS OF H IGHER O RDER

exist like houses or trees. At the same time, we shall not have denied that concrete objects are genuinely many­ sided, though such many-sidedness is not to be identified with a resolubility into related parts. Meinong's defence of analysis will be unaffected; only his interpretation of the results of analysis will be changed. In Meinong's whole treatment of the problem of characteristics we see at once how astonishingly acute and how strangely narrow his philosophical insight is. He is capable of simply seeing facts which are for most people hidden under masses of words, but at the same time he clings with extraordinary conservatism to any result at which he has previously arrived. His method is one of infinitely gradual advance in isolated issues; only very occasionally does he attempt to gather together the results of his scattered researches. The fascination of his thought is the fascination of the minute. V

We now turn to the theory of relations and complexes. In the earlier phases of his thought, Meinong emphasizes very strongly the inner incompleteness of relations, the fact that they are dependent on terms between which they can hold; similarity, for instance, can only have being if there are two things which it can bind together, and its whole being seems to lie in this function of binding. According to Meinong, this peculiarity places relations on quite a different footing from 'things' or even charac­ teristics. 'It is important', he says, 'to observe the antithesis between the various contents 1 of ideas in regard to what we may call their inner self-sufficiency ( Selbstiindigkeit) . Let us compare the content of a particular idea of colour or sound, with the content of the idea 1 ' Contents' here means 'objects'. The distinction had not yet been made by Meinong in 1 8 94, when these words were written .

O B J ECT S O F H I G H E R ORD ER

129

o f similarity o r any other relation . Colour i s n o doubt not altogether self-sufficient ; it is, in fact or by necessity, bound up with other contents such as place, extension, &c. But one could call such a lack of self-sufficiency an external lack, in comparison with that inner incompleteness ( Unfertigkeit) which pertains to the notion of a relation without its terms ; "red" and "sweet" may be bound up with accompanying facts, but they are, as it were, something in­ wardly rounded off and complete (ein in sich gleichsam .Ahgesch/0s­ senes) . ' 1

It cannot be doubted that Meinong is pointing to some important difference which every one readily experiences. In spite of the theoretical and symbolic advantages of treating characteristics and relations on a similar footing, and explaining their differences by pointing to the number of their terms and the way in which they inhere in those terms, we are inclined to regard qualities as considerably more substantial than relations: we can look at qualities and enjoy them, whereas the apprehension of a relation between two objects can only be consummated in a mental transition. The former can be held before the mind, the latter cannot. We may note here that by a ' relation' Mei­ nong does not mean some abstract link like the R in the symbolic expression xRy : he only recognizes such an entity much later and gives it the name Relat. By a 'rela­ tion' we must rather understand the actual relatedness of the terms, their being bound together ; it is because he approaches relations from this factual point of view, that their lack of self-sufficiency makes itself so obvious. In his early study of the theory of relations, 2 Meinong attempts to clear up the notion of a ' foundation' of a rela­ tion which occurs in Locke, and which Locke describes as the cause or occasion for comparing two objects. 3 Of such a foundation Locke gives as examples the marriage ceremony which is the foundation of the conjugal relation, 1 2

Psy. An. Gs. Abh. I, p. Gs. Abh. I I, 1 .

824187

322.

See also Geg. hoh. Ord. Gs. Abh. I I, p. 3 8 6. J Essay, II. xxv. 1 . K

1 30

O B J E C T S OF H I G H E R O R D E R

or the colour 'white' which is the foundation of the fact that Caius is whiter than sandstone. Meinong sees in this notion of Locke's nothing that can be made definite, 1 but thinks that the only necessary presuppositions of a relation are its terms, or, as he calls them psychologistically, the ideas of its terms. These are therefore properly to be called the foundations or fundamenta of the relation. 2 The notion of Locke arose because we often say we are com­ paring two objects, when we are really not comparing those objects as wholes, but their colours, or shapes, or other properties. It is these, and not the objects, which are the real fundamenta of the relation. On these fundamenta a relation is entirely dependent : a relation without fundamenta would be 'like a comparison in which nothing was compared'. Relations are 'founded' objects, or objects of higher order; just as the higher storeys of a building rest on lower storeys, so do relations depend on the terms they bind together. They are superi­ or a, and every superius demands one or more inferiora. 3 The inferiora, according to Meinong, show no such depen­ dence on the superiora; they can in some cases exist when their superiora do not, and even where superiora are neces­ sarily founded on certain inferiora, we can apprehend the inferiora without apprehending the superiora. 4 The funda­ menta of a relation may of course be other relations, and these may have other relations as fundamenta, but some­ where or other such a process must come to an end : we shall be brought down to the bed-rock of terms which are not themselves relations. 5 It is an a priori principle, that every superius demands not only inferiora, but also infima. This thesis lacks plausibility if by the word 'relation' we refer to entities like the relations of Russell, which 2 Ibid., p. 43. Reith. Gs. Abh. I I, p. 3 3. The possibility that a superius should have only one inferius is recognized i n Em. Preis., p. 1 0 5 , where obj ectives are explicitly recognized as obj ectso f higher order. s Reith. Gs. Abh. II, p. 44. 4 Geg. ho'h. Ord. Gs. Abh. I I, pp. 3 8 6, 3 87. 1

3

O B JECT S O F H I G HER ORDER

131

Meinong calls Relate. That the relation ' between' demands one more term in any instance-group in which it is found than the relation ' before', is a relation between these two relations which would subsist even if nothing whatever were between other things, or came before anything else. But Meinong's 'relations' are not these bloodless ab­ stracta: they are the actual relatednesses of definite enti­ ties: the ' betweenness' which connects Linz with Salzburg and Vienna, or the ' beforeness' which binds the years I 899 and I 900. That relations in this sense demand fundamenta can hardly be doubted. Meinong rejects all those forms of thought in which everything is held to be merely 'relative' to other things ; whether he is quite right in this, will be discussed later. In the case of space, for instance, he says that the side­ by-sideness (Nebeneinander) of its parts is undoubtedly a relative circumstance, but we can only have such a side-by­ sideness if the parts are there to be side-by-side. Every region in space must therefore have something which distinguishes it from every other region, and this Meinong calls its local determination ( Ortsbestimmung). 1 These local determinations are difficult to describe, because they are ultimate data like colours or sounds. But a person who has seen the Cathedral at Cologne and the Minster at Strasbourg has come into immediate cognitive relation with two distinct local determinations. Their distances from each other and from other determinations are a necessary consequence of their being the determinations they are, just as the similarities between shades of colour are a consequence of their being the colours they are ; 2 in neither case, however, does the term lose itself in its relationships. Temporal relations, such as 'before' and 'after', are similarly based on absolute temporal determinations. If I saw the fire in the Ring Theatre in r 88 r I was also aware 1

Ibid., P · 47.

2

Ibid., p. 49.

132

O B JECT S O F H I G HER ORDER

of its temporal determination; then I called it 'present' because it was exactly similar to the temporal determina­ tion of my experience in 188 1 , now I call it 'past' because my experience now and the event in question have widely dissimilar ternporal determinations. But in either case I am approaching an imperishable moment of a real object; the route by which I approach it is irrelevant. 1 We may note in passing that Meinong's whole contention is little affected by anything which the scientific theory of rela­ tivity has adduced : we have only to substitute space-time determinations for Meinong's space determinations and time determinations, and replace his necessary distances by 'intervals'. The necessity for absolute fundamenta can accordingly be pointed out even in the difficult case of the relations of space and time. VI

Meinong began by treating relations as wholly psycho­ logical entities : 2 it was natural that, at a period when objects were treated as 'contents' of ideas, the relations between those objects should themselves be treated as mental. Relations are held to be the ways in which we relate the contents of our mental states; they presuppose an original 'togetherness' ( Zusammenbestehen) of two con­ tents in our thought, 3 and then a further activity in which the contents are 'brought together' in a particular way. 4 He holds that the similarities and diversities which we observe between various contents are brought into being by our acts of comparison, and they are therefore called relations of comparison. 5 The relations of space and time are nothing but peculiar cases of similarity and diversity : we compare two local or temporal determinations and are able to say not only that they are diverse, but also the 1 2

4

Reith. Gs. Abh. II, p. 50 ; Geg. hr.ih. Ord. Gs. Aph. II, p. 457. Reith. Gs. Abh. II, p. 1 55. 3 Ibid., p. 3 9. Ibid., p. I 3 8 . 5 Ibid., P · 7 3 .

O B J ECT S O F H I G H ER ORD ER

1 33

degree of their diversity. 1 This degree of diversity be­ tween two determinations is what is ordinarily called their distance: it is an indivisible magnitude, which we pretend to measure when we are really only measuring the spatial stretch between the determinations.2 It follows that the whole spatial and temporal order of the contents of our experience is imposed on them by our acts of comparison : it is we who make the stars seem near or far from each other, and who build the raw material of sensation into organized patterns. Turning to incompatibility, Meinong holds that the incompatibility of two contents A and B simply is their ability to awaken in us a self-evident judgement that A and B cannot be associated with the same local and temporal determinations.3 Causality, according to Meinong, is a complex relation in which two components are discernible: one is temporal succession, which is nothing but a species of diversity or distance, the other is necessity, which Meinong at this period identifies with the impossibility of the contradictory state of affairs.4 The former relation is generated by our acts of comparison: the latter by the evident judgement that we pass. Meinong's original position was therefore one of extreme psychologism adopted without criticism from the reigning schools of thought: whatever unity and structure we discover in the objects of our experience was thought to be unquestionably the work of the mind. It is interest­ ing to see how he attempted to overcome the unplausibili­ ties of this theory, and how he was led to recognize that some relations at least cannot possibly be generated by the mind, since they are in the mind itself. Meinong was led to ask how, on the psychologistic theory of relations, we can justify our 'carrying over' into the non-mental world of relations such as similarity, 1 3

Ibid., p. 79· Reith. Gs. Abh. I I, pp. 8 8, 8 9.

2

4

Web. Ges. Gs. Abh. I I, p. 2 8 7 . Ibid .. P· I I 8 .

1 34

O B JECT S OF H I G HER O R D ER

diversity, causality, spatial and temporal distance, which have been recognized to be purely mental. No one who believes, as Meinong at this period believed, in a world of transcendent objects which are the underlying, unseen causes of the objects immanent in our minds, would wish to deny that these transcendent objects are distinct from each other, that they resemble each other with different degrees of closeness, that they are separated from each other by intervals analogous to our subjective time and space intervals, that they exercise causal influences upon each other and upon our minds. Yet if these relations are products of mental activity, such a belief seems to be merely nonsense. Meinong held, however, that the mental character of relations is so far from making their application to the non-mental world nonsensical, that it rather favours it. vVhen A and B are compared with each other we generate a relation X between them: when they are not compared no such relation can subsist. But in spite of this fact, it is still the case that A and B are such that, if they were put into relation with each other, X would be the necessary result. In this sense we can say of two objects that they exhibit a certain degree of diversity before they are com­ pared and even if, in fact, they are never compared: they are at all times the sort of objects which could serve as fundamenta for an idea of diversity. 1 We can make such statements about contents of our experiences and there is no reason why we should not also make them about the inaccessible causes of our ideas; if they could be given us and if we chose to relate them in various ways, they would exhibit all the similarities, distances, diversities, and causal dependencies that we ordinarily attribute to them. It is because these relations are products of mental activity that we are certain that they would subsist in such cases as well; if they were non-mental realities we could have 1

Reith. Gs. Abh. II, pp. 1 43, 1 44.

O B J EC T S O F H I G H ER ORDER

I

J5

no certainty with regard to cases where our experience failed us. This is very like a Kantian argument, but it should have led Meinong immediately to a purely realistic conclusion. Though similarity, diversity, distance, and causality are figments of the mind, the power of objects, in conjunction with the mind, to generate such figments, is in every case a relation which does not depend for its subsistence on mental activity. If we believed anything else we should be landed in a vicious infinite regress, since the generation of a relation presupposes that the fundamenta and the mind are properly adjusted to each other. Meinong no­ where maintains that the mental activity which generates relations is a purely arbitrary one ; all relations can be recognized in judgements, 1 and it is not given to us to judge as we will. When three local determinations A, B, and C lie before us in a visual field we are forced, if we bring them together at all, to arrange them in one way, to place B between A and C, for instance. It is clear there­ fore that these determinations stand in fixed, objective relations to each other and to our minds before mental activity supervenes ; they permit certain definite mental relations to be generated between them, and not others. This leads us to ask why, if in any case there are genuine mind-independent relations, similarity and 'betweenness' may not be among them. Meinong, however, was led to recognize the existence of real relations by the consideration of another group of cases. He takes the relation between an idea and its 'content' ; we must remember that by the 'content' he understands at this period the 'immanent' or 'mental' object. If I have an idea of a mountain, can we maintain that the act-element in the idea and the mountain as content are not really together until they are brought together by a special mental activity ? Does an idea of a 1

Ibid ., 1 56.

1 36

O B J ECT S OF H I G H ER ORD ER

mountain only actually become an idea of a mountain when I compare the two and bind them together in thought ? The absurdity of such a view is obvious; as Meinong says, the relation in question is not produced by any new activity, but is perceived by us, when we examine our ideas, in the same passive way as the absolute funda­ menta of a relation are perceived. 1 The same holds in the case of a complex mental process in which more than one intention is discernible. If I think of X and desire it, there are present in my mind an idea and a desire; shall we hold that the intimate unity into which they appear to be fused is nothing but the product of some mental synthesis ? Do I really need to consider an idea of X and a desire for X and judge that they are causally connected or what not, in order to have a unitary mental process in my mind ? 2 And if the awareness of a complex object XYZ implies that its components are bound together in thought, shall we hold that the ideas of X, r, and Z are only connected because they have been similarly synthetized ? 3 It is obvious that such a conten­ tion would lead to a vicious infinite regress. The more strongly we maintain that the unity between objects (or 'contents') is a product of mental activity, the less possible is it to regard the mind's own unity as the product of such activity. Meinong therefore came to the conclusion that there are two fundamentally different sorts of relation; there are real relations or 'relations of receptivity', which are in no sense produced by us, but merely recognized in real things, and ideal relations or 'relations of spontaneity', which are generated between objects by an activity of the mind. 4 Of the former we only know mental cases, but there is a vague possibility that there are unknowable relations of this sort between non-mental things.s The 1 4

Reith. Gs. Abh. II, p. 1 3 8.

Ibid ., p. 1 42.

Ibid. 3 Ibid., p. I 3 9 · s Ibid., pp. 1 42, 1 48 .

2

O B J ECT S O F H I G H E R O RD E R

137

latter we can generate between the contents of our ideas, and can carry them over hypothetically into the non­ mental world. VI I

It is not necessary to criticize this cumbrous hypothesis; when the content and object of an idea were distinguished in 1899, the whole temptation to confuse a relation with the mental act of relating dropped away. This act of relat­ ing, however, retains its importance as the means by which relations are given to our thought: 1 Meinong even holds that view that certain relations can only be given in this way. We can experience what blue or green is : but we cannot see similarity or diversity with our eyes or hear them with our ears. When we compare things seen, we generate, produce ideas by means of this activity; through such ideas we apprehend similarity or diversity in the same way that we apprehend blue or red through sensations. 2

The apprehension of certain relations does then really demand a 'bringing together in thought' ; but it must be observed that we do not produce similarity, or diversity, or causality, or any other relation by such a synthesis: we only produce the ideas of them. The production of such ideas merely involves the selection by thought of one superius, which, together with a vast number of other superiora, subsists between a number of objects, whether we think of it or not. 3 The whole treatment of relations therefore falls out of the sphere of psychology into that of the theory of objects: similarities and distances are obvi­ ously things which can as little be 'lived through' as houses or trees. Whether Meinong's production-theory is ten­ able, whether for example the distances in a visual field can only be given to us by a special mental activity, and Em . Preis. , p. 79. 0. A., p. 1 0 . See also Ameseder, Untersuchungen, Psychologie, p. 2 2 2 et seqq. 1

2

p. 48 et seq q . ; Witasek, 3 E ifgl. , P · 99 ·

I J8

O B JECT S O F H I G HER ORDER

are not already involved in the most immediate data, is a difficult question which will be discussed later. We must now note that, even when relations were taken out of the psychological sphere, the distinction of real and ideal relations, which was formerly based on mental activity or passivity, was retained, although it was given a new basis. Before we deal with this point, we must turn to Meinong's theory of complexes. In Ober Annahmen Meinong distinguishes between a complexion and a com­ plex, just as he distinguishes between a Relat and a rela­ tion, though neither of these distinctions are made in his earlier writings. A complex is any object which is made up out of several objects: in short it is a 'unity'. A com­ plexion, on the other hand, is the fact that certain entities form a unity, the being-together-of-these-things. When tones form a melody, the melody is a complex, the coming together of the tones is the complexion, as a result of which the complex arises. A complex of objecta is an objectum, whereas a complexion of objecta is an objective. 1 In the same way a relation is a relatedness of several terms, and is always an objective; a Re/at is the determining factor in this objective. The diversity between blue and green is a relation, but diversity as a pure determination, conceived in abstraction from all actual cases, is a Re/at. 2 As we remarked before, the relations of Russell are more like the Relate than the relations of Meinong. The connexion between relations and complexes con­ stitutes an important principle, which Meinong calls the principle of the coincidence of parts (Partialkoinzidenz). If a and h are to compose a complex,3 that is, be parts of a whole, some connexion must subsist between them, which makes them parts of a whole, i .e. they are constituents of a complex by virtue See 0. A., p. 279 ; Gs. Abh. I I, p. 474 , n. r 7. 0. A., p. 283; Gs. Abh. II, p. 474, n. r 7. 3 Meinong writes Komplexion, but the distinction between Komplex and plexion had at this period not been clearly made. 1

2

Kom­

O B J ECT S OF H I G HER O R DER

1 39

of a relation R, in which they stand to each other . . . . If, on the other hand, we begin by supposing that a and h stand to each other in the relation R, it is immediately evident that this cannot mean that a relative something is set beside two things which are possibly absolute, and that the three things together make up an objective collection. 1 We must rather say that a and h belong to a whole by virtue of the relation R in which they stand. If therefore there is a relation between a and h , there is ipso facto also a complex which has the terms of the relation as its constituents. 2 The principle was more simply formulated by E. Mally in the following words : 'Every complexion coincides with a relation (Beziehung) between its inferiora, every relation coincides necessarily with a complexion of its terms.' 3 The coincidence of parts mentioned by the principle consists in the fact that the same terms which are con­ nected by the relation are also constituents of the complex. Meinong believes that the terms are the fundamental entities in question, the relation presupposes them and them only, whereas the complex which supervenes pre­ supposes both the terms and the relation. The first part of the principle would hardly be ques­ tioned : we cannot conceive of a complex object unless there is some relation which binds its various constituents together. Thus if we consider the human body, or a rail­ way train, its various parts are bound together by spatial contact and by causality. Even in the case of such objects as patterns and melodies, which seem to exhibit unanaly­ sable qualitative differences, we recognize that such differ­ ences are based on the distances or intervals or other relations which hold between the various parts of the object. These relations appear to be so intimately fused with their terms, that it is less easy for us to see them than the unity which they produce. 4 I An 'obj ective collection' is a group : the word 'objective' serves to distinguish it from the collecting activity. 2 Geg. hoh. Ord. Gs. Abh. II, p. 3 89. 3 Untersuchungen, p. 1 5 3. • See Psy. An. Gs. Abh. I, p. 326.

OBJECTS OF H IGHER O RDER

The second part of the principle of coincidence is much less evident: it is not at first clear why relations should not hold between a set of entities, without its being the case that these entities are constituents of any complex. Thus there are innumerable relations between the throne of Queen Victoria and the throne of Aurangzeb: they exhibit many diversities and similarities, they are separated by certain space-time intervals, the one is perhaps more beautiful than the other, and so on. Yet only a philosopher would conceive the notion that there is somewhere in the world a complex of which the throne of Queen Victoria and that of Aurangzeb, and nothing else, are parts. Of course there are complexes such as the universe, or certain big spatio-temporal slices of the universe, of which the two thrones are constituents, but the relations between the thrones do not seem able to make any genuine complex out of them alone. Common sense would probably hold that some relations, e.g. distance or incompatibility, were so far from generating a complex out of their terms, as rather to separate those terms, and make their coming together in a complex impossible. We say that two notes cannot form a melody because the temporal interval be­ tween them is too prolonged, that two emotions cannot fuse into a complex emotion because they are incompatible. Meinong admits that this is in many cases the common­ sense conception; thus he says that the similarity which holds between red and orange has so little to do with the inner nature of its terms, that we fail to recog­ nize and have no name for the complex which it gene­ rates. 1 We may note here that the reason which Meinong gives for believing that every relation generates a complex is not a satisfactory one; he thinks that, if there is no such complex, the terms and the relation will merely lie side by side; we shall have red, blue, and diversity, but not red 1

Psy .

Kom . Gs.

AM. I, p . 295.

O B J ECT S O F H I G H E R O R D E R

141

and blue in diversity. This argument is certainly worth­ less: the objective or fact that red and blue are diverse, which is not a complex, nor made up of the objects which it concerns, 1 is quite capable of fulfilling all the unify­ ing functions which are here attributed to the complex. There seems no obvious reason whatever why objects which are related to each other should form any sort of 'whole'. Meinong is able to maintain his position because he believes that there are two sorts of 'wholes', real wholes which are capable of existence, and ideal wholes which can only subsist. As examples of real wholes Meinong points to the complex formed when a certain colour is bound up with a certain local determination; the coloured region which results is as genuine an existent as its constituents. This example naturally dates from the period when Meinong believed in the complex-theory of the 'thing'. He also points to a musical chord, which is made up of tones, but which is also a genuine existent and not merely the simultaneous existence of a number of tones. In the realm of the mind, also, various ideas, feelings, desires are bound together in complex states of mind, which are every bit as real as their constituent elements.2 With such real wholes we may contrast ideal wholes, which are incapable of existence. The most elementary of these is the whole which Meinong originally called an objective collection, and which he afterwards called a group (Menge). Graz and Linz and Salzburg form a group, whose binding relation, the least intimate of all relations, is denoted by the word and : Meinong speaks of this relation as the 'sowohl-als-auch' or 'as-well-as' relation. 3 Like other relations, it was originally confused with the mental process by which it is apprehended, which is in this case the act of collecting, but it was afterwards clearly 1

3

See above, Chapter I II, xi . Psy . Kom . Gs. Abh. I, p. 2 9 2 .

2

Geg. hoh. Ord. Gs. Abh. I I, p. 395.

r42

OBJECTS OF H I G H ER O RD E R

distinguished by Meinong and its non-mental character recognized. 1 The group consisting of Graz, Linz, and Salzburg is a genuine object and a part of the furniture of the earth, but it would be absurd to range it as an existent alongside of the cities which are comprised in it. Relations of space and time also generate complexes which are incapable of existence. Cassiopeia's Chair or the Scorpion are actually there in the heavens: it is not we who create them when we pick them out, but they have not the same crass reality as the stars which compose them. The masses of Palestrina are genuine wholes but they lack the existence of their component tones. The same applies to all movements, configurations, processes, and dispositions: they are ideal objects which are founded on related existents. 2 To discern them is easy in some cases and difficult in others; a pattern occurring in an empty field is readily recognized, surrounded by other patterns it may be entirely lost. We must, in some cases, actively bring together our awareness of several existents in order to see what pattern they form, but this makes no difference to the fact that the pattern was there before we seized it. When we have admitted that there are ideal wholes such as groups and patterns, we need go only one step farther to hold with Meinong that all relations which do not generate real wholes generate ideal complexes. The system of colours, ordered by the extent of their similarity or diversity, is an ideal whole, and so is the postal system of Italy, or the river system of China, or the liberal party, or the communion of saints. They are in no sense imagin­ ary entities: they are not on a level with dragons or round squares, but it would be nonsense to say of them that they existed. VIII

We can now complete our account of the difference between ideal and real relations. Both generate complexes 1

Geg. hon. Ord. Gs. Abh. I I, pp. 3 8 8 , 474.

2

Ibid., p.

400 .

O B JECT S O F H I G HER ORDER

1 43

by virtue of the fact that they hold between a group of entities, but the complexes determined by ideal relations can only subsist, whereas the complexes determined by real relations are capable of existence. Meinong also speaks of ideal relations as Beziehungen and real relations as Verhaltnisse. 1 We might translate the former by the phrase 'relations of standing', since they are the various ways in which one entity can stand to another, and the latter by 'relations of connexion'. It seems clear that Meinong has pointed out a very important distinction. Relations of standing leave their terms untouched; they do not bind them into any intimate unity. Aristotle long ago pointed out that the mere fact of resemblance between a Platonic Form and a sensible object does not bring the two into any real relation: a man could be exactly like Socrates although Socrates was quite unconnected with his origin or activities. The stars go their ways unaffected by the similarities and distances they have to each other, and the same is the case with regard to a state of mind and the objects to which it is adequate.2 No doubt there may be other relations which bind these objects into real unities; quite possibly, as Meinong him­ self believed, 3 causal action involves real relations as well as the ideal relations of succession, and perhaps, also, all things are mystically one in God. But, however this may be, it remains the case that similarity, distance, diversity, adequacy do not, by themselves, generate real unities. Relations of connexion, on the other hand, involve some sort of fusion of a number of entities, not fusion of their intrinsic natures, which is impossible and meaningless, but participation in a whole which is quite as real as its parts, and which may even have a superior self-sufficiency, inas­ much as its parts can only exist in it, or some similar whole.4 It seems quite clear that states of mind, and 1

J

Ibid., P · 3 96. See Gs. Abh. I I, p.

2

1 8 0.

4

0. A., p. 266 . See Phantasic. Gs. Abh. I, p.

208.

144

O B J ECT S O F H I G H ER ORD E R

possibly certain fundamental physical objects, are real wholes in a sense in which constellations or groups, if we treat them as constellations or groups, and ignore other more intimate relations which inhere in them, can never be real wholes. Meinong also believes that the Relat which is involved in a relation of connexion is a real object, as the Relat involved in a relation of standing never can be. Mental togetherness exists as much as the states which are together, but similarity as a pure Relat is incapable of existence. The other important difference between ideal and real relations lies in the fact that the subsistence of the former between a set of terms is an a priori necessity, whereas the subsistence of the latter is a brute empirical fact. 1 The similarity between two colours or the distance between two local determinations is something which could not be otherwise, our terms being what they are. The name 'founded object' which was originally applied to all rela­ tions and complexes, was afterwards explicitly restricted to the ideal relations, and the complexes they determine. 2 Only these are, in the strict sense, founded on the terms they relate. Real relations, like ideal relations, are superi­ ora, or objects of higher order, but we cannot see why they should hold between their inferiora. The conjunction of this instance of red with these local and tern poral deter­ minations, or of this idea with that feeling, is in fact very intimate, but regarded from the point of view of logical necessity, it is utterly external. It is not at all clear that these statements harmonize with Meinong's general position, to be discussed later, that all determinations of a real object are parts of its nature, 3 but we may say, at any rate, that real relations between objects are not deducible from other determina­ tions of those objects in the way that ideal relations are. 1

3

Geg. hoh. Ord. Gs. Abh. I I, p. 398 . See below, Chapter V I , ii.

2

I bid., p. 399.

O B J E C T S O F H I G H ER ORD ER

1 45

In this sense the former are brute facts, whereas the latter are 'founded' on the objects they relate. Ideal relations, therefore, presuppose Relate which are ideal, generate complexes which are incapable of existence, and hold with necessity between their terms: real relations, on the other hand, presuppose real Relate, generate complexes which exist, but the fact that they hold between their terms remains a brute fact. IX

According to Meinong, a relation is never a constituent of the complex it generates. 'The relations which a com­ plex involves in itself, essential as they are, should never be counted among its constituents.' 1 Thus if an idea and a feeling unite together to form a peculiar mental state, we may regard the idea and the feeling as constituents of that complex state, but we cannot treat the relation which unites them, whatever it be, as a constituent. It is a mode of union, not something united. In the same way, when similarity relates A and B we should not regard similarity as a constituent of this ideal complex; it is merely the way in which the constituents stand to each other. We can, however, discover other ideal complexes of which the relations in question are constituents. Thus, involved in the complex AB which similarity generates, we can discover another complex whose constituents are A, B, and similarity, which are bound together by another relation or relations. Meinong and Mr. Bradley think that we require two relations, one to bind similarity with A and another to bind it with B. We can, however, make use of a single triadic relation of 'holding between', and say that similarity 'holds between' A and B. We can go on to discover another complex whose con­ stituents are A, B, similarity and the new relation which connects them. This will presuppose another relation 1

824187

Geg. hb"h. Ord. Gs. Abh. I I, p. 3 9 1 . L

OBJECTS OF HIGHER ORDER

which is not a constituent of the complex. In this way we can discover an infinite series of complexes involved in any complex; each complex in this series will arise out of its predecessor when the relation which makes it a com­ plex is turned into a constituent of another complex. As Meinong remarks, 1 there is no more difficulty in all this than in the infinite divisibility of a line. The regress would only be vicious if the original relation of similarity did not really relate the two entities A and B, and we were trying to patch up the breach by a series of relations all of which were infected with the same weakness. 2 X

We may now compare Meinong's theory of complexes with an opinion held by Mr. Russell when he wrote The Principles of Mathematics, and also when he wrote his articles entitled Logical Atomism. 3 This is briefly the view that there are no such· things as complexes, and that all statements apparently about complexes can be resolved into statements about what would ordinarily be called their ultimate constituents. In The Principles of Mathe­ matics Mr. Russell discusses the view of those people who think that space is a 'unity', by which they mean an exis­ tent complex, and not a mere class of points, and finishes up by saying that for him 'it is sufficient to observe that all unities are propositions or propositional concepts, and that consequently nothing that exists is a unity'.4 In Logical Atomism he tells us that he does not believe that many objects which we treat as complex really exist : Socrates, Piccadilly, and Twelfth Night are given as in­ stances of such spurious objects. 5 Mr. Russell seems to mean that such things as 'space' or 'Piccadilly' are not really single objects at all, but a great number of separate Geg. hb'h. Ord. Gs. Abh. II, p. 3 90. Monist, 1 9 1 8- 1 9 . s Monist, Oct. 1 9 1 8 , p. 5 1 2 .

1

J

4

2 See 0. A., p. 2 6 1 . Prine. of Maths., p . 467.

O B J E C T S O F H I G H E R ORD E R

1 47

objects, which no doubt stand in various relationships to each other, that is, function as the material of many rela­ tional facts, but do not constitute complex objects, as they are usually supposed to do. The fact, objective or proposi­ tion becomes the one unifying factor among the ultimate objects which, we assume, underlie the spurious objects of common sense. We may illustrate Mr. Russell's view by considering Cassiopeia's Chair: to a person acquainted with the geo­ graphy of the heavens, it seems to be a genuine object about which many assertions can be made. It might, for instance, be maintained that there are minute alterations in the shape of this constellation, and that in a few million years it will have become perceptibly distorted. If, for the moment, we treat the stars in the constellation as ultimate objects, then the view in question forces us to substitute for our statement about Cassiopeia's Chair a statement about those stars, to the effect that at time t1 , they are at certain definite distances from each other, and that at time t2 they are at other different distances from each other. Cassiopeia's Chair vanishes; it is not required in a com­ plete description of the world. In the same way we might hold that facts apparently about Piccadilly would resolve themselves into facts about the relative positions at certain dates of certain atoms, electrons or other more primitive entities, and facts apparently about Socrates would resolve themselves into facts about the sequences and coexistences of certain ultimate mental and physical events. The real and ideal compl exes of Meinong will both be eliminated: only his objectives will be unaffected. If simplicity and economy of hypothesis were in all cases the test of truth, there can be no doubt that the theory of Mr.Russell, which is also that of Dr. Wittgenstein, would be very satisfactory. But it seems to come into conflict with the plain fact that a person who looks at Cassiopeia's Chair does see something, and that this

OBJECTS OF HIGHER O RDER

something is not to be identified with a number of stars, or with a set of facts about the relations of those stars. It will not do to say that the unity of the constellation is merely a mental product; whether the mind unifies its ideas or not, it is quite clear that it is not thinking about its ideas or their unity when it considers Cassiopeia's Chair. The unity which lies before the mind is an objective unity, whether such a unity has being or not. And once we admit that such a unity is a genuine object, there is less temptation to doubt its being. The belief that what we see when we look at Cassiopeia's Chair really is there, reposes on two grounds : that it is free from contradiction, and that it is foolish to doubt our judgements of percep­ tion when they are free from contradiction. The same arguments apply to Mr. Russell's general doubts about the being of classes; that classes are objects is simply indubitable, and there is no reason to exclude them from being, even if certain classes involve contradictions. If we turn from ideal to real complexes the theory of Mr. Russell has even less plausibility; that a complex mental state, whose unity we live through, is really a number of atomic experiences and a set of facts about their mutual relations, or that a similar account applies even to so poor a unity as Piccadilly, is quite incredible. These too are objects, they appear to exist, and there is no good reason to be sceptical about this appearance. If, however, we leave aside the whole question of the existence or subsistence of complexes, they will still retain their place in the theory of objects as pure objects, indifferent to being. XI

We may conclude our account of Meinong's theory of relations and complexes, by asking whether his whole view of them as 'objects of higher order' is really tenable. What such a view suggests is a sort of atomism; there is the

O B JECT S O F H I G HER ORDER

1 49

basic stuff of the world, its raw material of simple self­ sufficient terms, and on this foundation a superstructure of dependent entities is built. The natural criticism which such a view suggests is that complexes which involve many terms in relation are often more self-sufficient than their constituents. The logical priority of the latter does not affect the existential priority of the former. We find that this very objection was made against Meinong's treatment of the continuum by Schumann. According to Meinong, a continuum, e.g. a red square, is an object of higher order, just like a constellation; but its inferiora are continuously bound together and are not discrete like those of the constellation. The puzzling question then arises as to what these inferiora really are, to which we can only answer that they are local determina­ tions associated with redness. Meinong holds that local determinations are not points, because they are parts of a continuum. 1 But if a local determination is not a point, each local determination will be an object of higher order, founded on other less extensive local determinations, and in this way there will be no ultimate infima, on which the whole superstructure can repose. Against this whole conception Schumann brings for­ ward the Aristotelian theory of the continuum, that the red square in question is an undivided unity, and that the parts into which we divide it in thought are merely ficti­ tious.2 The divisible is not the actually divided. Against this Meinong makes the very natural objection that the divisible must have within itself material for distinction : it must exhibit inner differences, or else division will have no foothold.3 But Meinong thinks that these differences need not be parts in the strict sense; they may not even separate themselves into natural unities. They are 'indeter­ minate constituents', and by calling them indeterminate 1 2

Relth. Gs. Abh. II, p. 48 ; Geg. hiih. Ord. Gs. Abh. II, p. 426. 3 Ibid., p. 42 3 . I bid., p. 42 0 .

O B J ECT S O F H I G H E R O R D E R

Meinong does not merely mean that we have not picked them out, or determined their boundaries, but that they are, by their very nature, incapable of determination. 1 A continuum is therefore an object of higher order founded upon indeterminate constituents. It seems quite impossible to attribute any meaning whatever to the phrase 'indeterminate constituent' ; Meinong's statements about the continuum appear to rest on a popular but self-contradictory conception of con­ tinuity. It seems clear that a red square contains, as actual, determinate parts, all the conceivable sets of parts which our divisions merely emphasize; it is also clear that it has precisely the same sort of reality as its parts, and is as much a unity as they are. Hence the whole notion of a superius basing itself insecurely on a multitude of hurrying atoms is inapplicable. It is quite arguable, as Kant argued, that the space-time system as a whole is existentially prior to its various parts; it is not founded on them, but they distinguish themselves in it. The whole conception of objects of higher order is based on the notion that relations, however necessary they may be, somehow lie outside of the terms they connect. The terms must be there with full-blown natures, before the relations can supervene. We find Meinong moving away from this conception when he holds that a relation is really nothing but a very curious characteristic which can­ not inhere in one object alone, but only in a number of objects, and in each only in so far as it inheres in the others as well. 2 If such a view is carried through, relations will not be objects of higher order, nor will they be less self­ sufficient than characteristics. The characteristic 'red' has being in this instance of 'red', the relation 'greater than' has being in the couple Vienna-Graz. We must in the case of a relation distin­ guish not only the terms in which it inheres, but the way 1

Abst. u. Vergl. Gs. Abh. I, p. 466.

2

V. A., p. 2 8 3 .

O B J E C T S O F H I G H ER O R D ER

151

i n which i t inheres in them. An asymmetrical relation like 'greater than' inheres in the members of its instance-group in two different ways: it inheres in Vienna in one way, and makes it greater than Graz, it inheres in Graz in another way, and makes it less than Vienna. Symmetrical relations on the other hand inhere in all their terms in the same way. A triadic relation like 'betweenness' which is partly sym­ metrical, might be written in the following form: R 1 22 • R 1 2 2 (a, b, c) would mean that a is between b and c. Since R inheres in b and c in the same way, denoted by the index 2, we see at once that R 1 22 (a, b, c) is equivalent to R 1 2 z (a, c, b), but not equivalent to R 1z2 (b, a, c) . 1 Once this view of relations is adopted, we have no reason to exclude them from the nature of the terms which they relate. We can ignore the relationships of an object in the same way as we can ignore some of its properties: but just as trivial properties are not really accidental, so relations are never really external. All alike go to the building up of the so-being or nature of concrete objects. We may note, also, that the possibility of objects whose whole nature consists in relation to other objects, is theoretically allowable. As an example of such a purely relativist system, we may take an instance-group of a relation R which is n-adic and asymmetrical throughout. We should then have n entities each of which was distin­ guished by the fact that R inhered in it in a different way, so that confusion between them would be impossible. But it is hard to say whether such purely relative entities have existence anywhere. 1 This paragraph owes a great deal to the chapter on Relations in Mr. W. E. Johnson's Logic, vol. i.

VI T H E T H E O R Y OF I N C O M P L E T E O BJ E C T S

I

I

F we consider any object, what we call its 'nature' will be given to us in a series of objectives of so-being. There is an important division of such objectives into objectives of how-being ( Wiesein), and objectives of what-being (Wassein) . As instances of the former we may point to such facts as 'X is blue', 'X is extended', while ' X is a cube', 'X is a lump of clay', will serve as instances of the latter. The complete knowledge of a concrete object would involve a knowledge of every fact of how-being and every fact of what-being that concerned it. 1 Even if we were able to enumerate every such fact, our knowledge would still be imperfect, however, for the nature of an object is as much determined by the circum­ stances that are absent from it as by those that are present. To know an object completely, therefore, we should have to know exactly what it was not and how it was not. If we confine our attention to those objects which have some sort of being, it is clear that they will all have natures that are infinitely many-sided: we shall never be able to know more than a vanishingly small portion of their so-being. By the law of excluded middle, which holds without reserve for objects that have being, any object will be 1 Mog., p. 1 69. From such statements one may infer that Meinong includes every property of an object in its 'nature'. When Meinong wrote Ober Moglich­ keit und Wahrscheinlichkeit he had not yet considered Russell's doctrine of 'illegitimate totalities'. He deals with this to a certain extent in Ober emotionale Priisentation, § 2. To satisfy the requirements of the theory of types, we should have to say that every first order fact of so-being about an object was involved in the nature of that object.

T H E THEO RY O F I N COM PLETE OBJECT S

1 53

involved in the nature of any other object. The object B will be involved directly in the nature of the object A as part of its what-being or how-being, or it will determine that nature indirectly by its absence. Thus the objects denoted by the words 'brown' and 'article of furniture' are comprised respectively in the how-being and the what­ being of some table; whereas the objects denoted by 'poly­ phonic' and 'sea-serpent' enter into the nature of a table, in so far as they are absent from its how-being or its what­ being. Every object represents two possible determina­ tions for every other object; it must either fall inside the other object's what-being (or how-being), or fall outside of it. In the case of such objects as have being, one of these alternatives is always a fact. II

It will be interesting to consider here whether Meinong means to include in the nature of objects which stand in certain relationships, the fact that they stand in those relationships. Is it part of the nature of my desk to be possessed and cherished by me, to stand in a certain part of my room, to have undergone certain accidents which involved a spilling of ink, to be worth a certain sum of money, and so on ? On this point Meinong's attitude is rather vacillating, but it is only possible to make sense out of a great deal of his theory by assuming that relational properties do enter into the nature of the objects that they characterize. In the Relationstheorie relations are apparently treated as if they fell outside of the terms which they connect; Meinong says that the subsistence of a relation presup­ poses the being of its terms, 1 and this seems to imply that the natures of these terms would be completely given even if we ignored the relations between them. Such a view is suggested also by the many passages where Meinong 1

Reith. Gs. Abh. II, p . 44; Em . Preis., p . 70.

1 54

T HE T HEORY O F I NCO M PLETE O B JECT S

insists that it makes no difference to the nature of an object whether it is apprehended or not, or in what way it is apprehended. 1 But, as against this view, we may point to the passage in Ober Annahmen where Meinong explicitly declares that relational objectives are ordinary objectives of so-being.2 A relation like diversity determines two objects that are diverse, in precisely the same way as the characteristic 'green' determines a green object; the only difference lies in the fact that diversity inheres in a plurality of objects, and can only inhere in each of them in so far as it inheres in the others at the same time, whereas a characteristic need not inhere in more than one object. If the difference between relations and characteristics is as unimportant as this passage suggests, if relations are only 'polyadic universals', whereas characteristics are 'monadic universals', then there seems no good reason to exclude relations from . the natures of objects, while we include characteristics in them. The diversity which holds between A and B is not, of course, comprised in the nature of A taken separately, or in the nature of B taken separ­ ately, but in the natures of both A and B taken together it is undoubtedly comprised. And if we turn to the relational properties which this relation generates, there is no reason whatever to exclude them from the natures of the objects in question; the diversity from B which A possesses, and the diversity from A which B possesses, do not belong to them in any other way than their ordinary characteristics. Meinong's whole treatment of concrete existents im­ plies that there is simply no sense in drawing a line between their internal and their external properties. He could not agree with Professor Moore's restriction of the 'nature' of such objects to their qualities, as opposed to their relational properties, 3 nor admit that 'being a man' is an 1 2

See e.g. Geg. hiih. Ord. Gs. Abh. I I, p. 457; Mog., p. 1 77 . 3 See Philosophical Studies, p . 308 . 0 . A., p. 2 8 3 .

T HE T HEO R Y O F I N CO M PLETE O B JECT S

1 55

internal property of King Edward VI I, whereas 'being the father of George V' is one of his external properties, which he might conceivably not have possessed. Meinong says of a writing-table, which is standing before a certain window at the present moment, that it is, strictly speaking, nonsensical to say that it could be standing anywhere else at the moment in question. The table is a concrete real object, which is determined in every possible way: the possibility that it should be standing in some other part of the room is excluded by the fact that it is standing where it stands. 1 This argument will only be cogent if Meinong means to include its position at a certain date in the nature of the writing-table; otherwise it will merely be a confused piece of reasoning of the sort finally exposed by Professor Moore. It is indeed a very arbitrary procedure to pick and choose among the properties of a concrete 'this', and to group one set of properties in its essence, another set among its accidents. And to include the most unimportant qualities of an object in its nature, while we exclude its most important relational properties, has no justification whatever. Plato's birth in Attica, his association with Socrates, his foundation of the Academy, surely do as much to constitute him the object he is, as his height or his build or the colour of his hair. No doubt the selection of certain properties of an object, and their treatment as essential properties, is both permissible and inevitable, but, in so far as such selection occurs, the concrete object is left behind. We are substituting for a complete object one that is incomplete and abstract, and there is no sense in asking whether two such objects are identical. The writing­ table that could indifferently have stood, or not have stood, before a certain window at a certain date, is an abstract entity which we cannot identify with the historical table of our actual experience. 1

Mog., p.

222.

1 56

T HE T H EORY O F I NCO M PLETE O B JECT S

Meinong says that subsistents are completely deter... mined in every possible way, just as existents are. 1 This implies that all their relational properties will enter into their natures, as the relational properties of existents enter into their natures. It is not altogether clear that this view is correct. If we consider a relation of similarity between two objects, or an ideal complex like a melody or a certain fact, there seems to be a clear distinction in their case between what they are intrinsically and what properties they possess extrinsically. That similarity is in this case accompanied by incompatibility, that this melody is built upon violin tones, that this fact has certain implications and is known by certain people: all these circumstances seem to lie entirely outside of the objects that they con­ cern. In the case of ideal objects we seem to be able to separate kernel and husk, whereas concrete existents do not naturally admit of such divisions. III

Objects which exist or subsist are, according to Meinong, determinate in every possible respect; we can never know their natures fully because it is impossible for us to run through an infinite number of facts of so­ being. But there is another class of objects which have a finite number of determinations, and which are perfectly accessible to our thought. These objects are called by Meinong incomplete objects (unvollstiindige Gegenstiinde) : their chief importance is epistemological, since, as we shall see, it is by their means that a reference to completely determined real objects is rendered possible. But in this chapter we shall be concerned with them mainly as a peculiar variety of object; all genuine objects are worthy of study whether they exist or not.2 An incomplete object lies before us whenever we think, in a perfectly general way, of 'something that is so-and-so'. 2 Ibid ., p. 1 8 1 . 1 Mag., p. 1 69.

T H E T H E O R Y OF I N C O M PL E T E O B J E C T S

1 57

It is clear that a person who thinks of 'something that is black' is thinking of an object, but it is also clear that such a person need not consider whether the object that he is thinking of is square, or heavy, or what other pro­ perties it has. If he likes, he can treat the object of his thought as 'closed', 1 in which case it not only has no further properties, but is incapable of having any. If the man were asked whether the object that was before his mind was heavy he would have no alternative but to answer ' No'. To think of something that is black and heavy is quite different from thinking of something which is merely black, and this difference ultimately goes back to a difference in the objects concerned. In general, however, we treat objects of the form 'some­ thing that is so-and-so' as if they were 'open', not 'closed'; that is, we leave blanks in their constitution which we fill up as we proceed. Thus, if a certain object were given to us in a judgement of perception as something black, and was afterwards discovered to be heavy and square, we should ordinarily consider that such 'filling up' did not affect its identity. This view is a loose one which Meinong afterwards rejects. But, accepting it provisionally, we can see that such a process of 'filling-up' necessarily involves that there is something to which the successive additions are made, and this something must be a certain sort of object. Meinong illustrates the 'filling-up' process by the case where we predicate extension of a coloured object. The relation between colour and extension is, in the case of existents, one of necessary implication; an existent cannot possibly be coloured if it is not extended. But it is the very intimacy of this relation which, by contrast, shows up the total independence of the object 'something that is coloured, from the property 'extension'. The very fact that something is added to the first object when it is 1

See 0. A., p.

271.

158

THE T H E O R Y O F I N C O M PLETE O B J E CTS

'filled-up' by the second, shows that i t differs from all such additions. Meinong remarks: Over against that which we must think or ought to think together with the blue obj ect whenever we think of it, there stands some­ thing which we, in our thinking, really think of as its nucleus (Kern) or essence, if the thought 'something blue' is really present. This obj ective nucleus is clearly given in the meaning of the expression 'something blue' ; and this meaning is nothing but an obj ect attached to the word in a peculiar way. And if we now ask whether the moment of extension is in any way comprised among the properties of this obj ect, and if we do not in any respect whatever go beyond it, and thereby pass on to another obj ect, the only possible answer to such a question is an unqualified 'No'. 1

We could only avoid Meinong's view, that a man who thinks of 'something blue' is thinking of an incomplete object, by holding that such a man is really thinking of all the blue concreta in the universe. Such a theory would only apply to cases where a judgement of existence was involved; if I believe that a blue object exists, I also necessarily believe that one or other of the actual objects in the universe is blue. Such a belief is therefore in some way directed to concrete objects. But if a person merely thinks of something blue, without having any beliefs about such a thing, it is impossible to identify the object of his thought with any actual blue concretum. This is even clearer if we imagine some one to think of something immortal or timeless, where it is doubtful whether any concrete existent possesses such a property. Against the theory in question there is also the plain evidence of inner perception; if I think of 'a square', or 'a house in a tree', or 'a sphere of ivory 1 0 metres in diameter' (to use an example of Meinong's), it seems clear that I am not setting before my mind a plurality of objects; such a thing can only be maintained if the objects of my thought are held to be quite different from what they 1 Mag ., pp. r 0, r 7 7r.

THE THEORY O F IN COMPLETE OBJECTS

1 59

appear to be. It is incredible that a person who puts before his mind the apparently single object indicated by the words 'a house in a tree', is really thinking of the vast assemblage of all houses that ever were or will be in trees. Nor is it an objection against Meinong's theory to say that such an object as 'a house in a tree' cannot exist unless we specify completely what sort of a house it is. The fact that it is possible to add specifications to 'a house m a tree', proves that it is something, i.e. an object. IV

Objects of the form 'something that is so-and-so' do not, strictly speaking, possess any characteristics beyond those specified in their 'so-and-so'. 'Something that is black' is not and cannot be extended or heavy. We might therefore be led to assume that 'something that is black' lacked heaviness or extension; indeed such a conclusion might seem to be forced on us by the law of excluded middle. This would lead to the conclusion that 'something that is black' was not merely a non-existent, but also an impos­ sible object, for a black object cannot lack extension. That it was an impossible object would not, of course, affect its status as an object, but it would place it in the same un­ important class as the round square. And if 'something black' definitely lacked all other properties, it would be impossible to treat it as an 'open' object which we gradu­ ally filled up, because such filling up would involve a contradiction. Meinong considers the case of 'the triangle' in abstracto ; it is clear that this is not equiangular, and it might there­ fore be thought that it lacked equiangularity. But we generally assume that a particular triangle has all the properties that are possessed by 'the triangle', though it has others as well; only on such an assumption is there any point in using the same name for both of these objects.

1 60

THE T H E O R Y OF I N C OM PLETE O B J E CT S

If, however, 'the triangle' comprises among its properties the lack of equiangularity, as a state of mind comprises among its properties the lack of extension, this lack will be present in all particular triangles, and an equiangular triangle will be as impossible as an extended state of mind. As this is not the case, we cannot hold that 'the triangle' lacks equiangularity, any more than that it possesses it. 1 That such a state of affairs involves a contradiction was maintained by Berkeley, and led to his celebrated on­ slaught on abstract ideas. And the abstract ideas which Locke had described were self-contradictory inasmuch as they were held to have all as well as none of the peculiari­ ties of their several particular embodiments. 2 Locke's theory of abstraction violates the law of contradiction as well as the law of excluded middle. Meinong, in holding that his abstracta have none of the properties of their particular embodiments, only violates the law of excluded middle, which has always been the most shaky of the laws of thought, and the most difficult of precise application. It is not that it is not a true law, but that it is hardly ever possible to be sure that we have got hold of a simple negative. Mr. Russell has made us familiar with the doctrine that the two statements: 'The present King of France is bald', and 'The present King of France is not bald', are both false, as both imply that there is a King of France. The proper negative of 'The present King of France is bald' is 'It is not the case that the present King of France is bald,' which leaves open the possibility that there is no King of France. A similar solution for his difficulties is proposed by Meinong. The objectives 'A possesses B' and 'A lacks B' are both objectives of so-being; they appear to be 1

Mag., p.

1 72 .

Locke: 'It must be neither oblique nor rectangle, neither equilateral, equi­ crural, nor scalenon, but all and none of these at once.' Essay, IV. vii. 9 . 2

T H E T H E OR Y O F I N CO M PL E T E O B J ECT S

161

simple contradictories, but are in reality contraries. When A is a concretum, i.e. a fully determined object, there is no harm in treating the objective 'A lacks B ' as if it were the contradictory of 'A possesses B'. When, however, A is not a concretum, but an undetermined or abstract object, such a proceeding is invalid, and leads to Berkeley's fallacious refutation. 'A lacks B ' is called by Meinong a not-so-being (Nichtsosein) ; it is an objective of so-being like any other, only in this case the absence of B functions as a property. On the other hand, ' It is not the case that A possesses B ' is the not-being of a so-being (das Nichtsein eines Soseins) . Since A need not be a concretum, it may be in a position where certain sorts of so-being cannot touch it at all. We can describe this independence of either the posses­ sion or the lack of certain properties in two ways : the less satisfactory is to say that the law of excluded middle breaks down in the case of certain objects. Thus Meinong says: I f we say that an obj ect .A is determined in respect of an obj ect

B when we are entitled to say either that it is B or that it is not B,

then 'something blue' is undetermined in respect of extension, and the principle embodied in the law of excluded middle (which, as we saw, holds in the case of all that is actual or subsistent), a principle by virtue of which every obj ect must be determined in respect of every obj ect, has no longer a j ustifiable application to the obj ect 'something blue in the abstract' . 1

Such an account is unsatisfactory because there is un­ doubtedly a sense in which 'something blue' is not ex­ tended, i.e. if we mean by saying that it is not extended, that it is not the case that it is extended. This use of the negative Meinong calls a wider form of negation (erweit­ erte Negation). 2 With such a wider negation, the principle of excluded middle applies without restriction. In all i 824187

Mag.,

P· r 7 1 .

2

M

Ibid., P · I 74 •

1 62

T HE T HEORY O F I NCO M P LETE O B J ECT S

circumstances it is either the case that A is B, or not the case that A is B, but A may fail to be B in two ways. The more familiar way is when A is an object, concrete or abstract, which is determined in respect of B. In this case its failure to be B implies that there is a gap or defect in its nature at the point where B might have been present. In the other case, where A is not determined in respect of B, its nature has no such definite gap, but a blank of a different sort which presupposes nothing, like the blank­ ness of one of those compartments in a logical diagram which we may proceed to 'empty' or to 'occupy' at will. It is very difficult to give any clearer account of the difference between these two types of absence. We can only say that 'A lacks B' excludes B from A, prevents A's possession of B from being a fact, whereas 'It is not the case that A is B' will in some cases allow B to be added to A.Such an addition of B to A results of course in a new object, more determinate than A, of which A is the nu­ cleus; one type of absence destroys an object's capacity for acting as such a nucleus, the other type of absence does not. V

Meinong distinguishes therefore between objects which are subject to the law of excluded middle in its narrower form, i.e. which are determined in respect of every object, and those which are not.The former are called completely determined or complete objects, the latter incompletely deter­ mined or incomplete objects. 1 The reason why these objects have tended to hide themselves from us is that we are so exclusively occupied with concrete existents, and these, as we have seen, are always completely determined. Meinong points out the very interesting fact, to which we shall recur later, that nearly all knowledge of con­ crete existents is by means of incomplete objects. In this 1

Mag., p.

178.

T HE T HEORY O F I NCO M PLETE O B JECT S

163

lies the imperfection and vagueness of our knowledge of concreta ; practically all of them are known not by 'acquain­ tance' but by 'description'. To use an example of Aris­ totle's, if I see Callias walking past, I may only be aware of him as 'a man' ; the incomplete object 'man' here func­ tions as an instrument by means of which I vaguely grasp the vastly complex concrete object ' Callias'. It is because we use incomplete objects in this wholly instrumental way, that they appear to be swallowed up in the concreta which they help to elucidate. Our preoccupation with the concrete makes it an ex­ tremely unnatural proceeding to consider anything inde­ terminate for its own sake, and indeterminate objects are usually treated as mere first sketches for their more definite successors. Our impatience with the indefinite is so great that if we are presented with a set of concepts, those of early thinkers, for instance, which are indeterminate in many ways, we proceed to fill them up with interpretations of what we suppose those thinkers 'really meant'. There is also a universal amnesia by which we are quite unable to recall a first imperfect impression of something, once the blanks in that impression have been filled up with further details, an amnesia which explains the inability of the learned or the sophisticated to imagine the experiences of the ignorant or the naive. Further, we tend to confuse the indeterminateness of certain objects with vagueness, and because vagueness is a quality of our ideas which we desire to eliminate as far as possible, we imagine that it is also desirable in every case to replace indeterminate by determinate objects. But the indeterminateness of objects such as 'man' or 'the dog' or 'something blue', has nothing that resembles vagueness or confusion. If I am aware of Callias by means of the incomplete object 'animal', I have a vaguer knowledge of him than if I know him by means of the object 'man' ; the range of entities with which he may be confused is also

1 64

THE THEORY O F I N COMPLETE OBJE CTS

wider. But though I may apprehend Callias more or less vaguely, the indeterminate objects which I use in such apprehension are not in any way vague. The nature or so-being of an object like 'man' or 'the dog' unfolds itself up to a certain point and there it ceases; we cannot pene­ trate farther because there is nothing more to know. Beyond its nucleus of constitutive properties, the object allows of completion in an infinite number of ways; but to complete it would be to pass on to the apprehension of other objects in which it is embedded. There is no reason why we should not study indeterminate objects as in­ determinate objects, instead of passing beyond them in this way. We may say that Meinong has restored to philosophy a class of objects whose nature and status had been tem­ porarily obscured by nominalist criticism. In principle they do not differ from the genera and species of Aristotle. It is usual to regard genera and species as classes to which certain individuals belong, but it is clear that this is not the meaning of Aristotle on the majority of occasions on which he uses the words. The species 'man', he tells us, can be predicated of the individual man, and the genus 'animal' is part of the nature of the individual man. When he makes such statements he cannot be thinking of classes, which could never enter into the nature of an individual or be predicated of it. His words only mean anything if the species 'man' and the genus 'animal' are taken 'in intension'; on this interpretation both are indeterminate objects, which act as a nucleus round which further determinations, called by Aristotle 'accidents', can gather. We may here point to an ambiguity in the word 'uni­ versal', which has often confused philosophical discus­ sions. Sometimes we mean by 'universal' objects such as 'man' or 'triangle', i.e. incompletely determined objects, and sometimes such entities as qualities and relations, e.g. humanity or triangularity. Mr. Russell has practically

T HE T H EORY O F I N C O M PLETE O B JEC T S

165

restricted the application o f the word t o objects o f the latter type, but it is still often used in the other sense. Now it is clear that there is an unbridgeable gulf between an object such as 'man' in abstracto and the complex of properties called 'humanity'. It cannot be right to use the same word 'universal' for objects so different. The point is that indeterminateness is a peculiar property which may be met with in objects belonging to any category. There are indeterminate characteristics, relations and objectives, and there are indeterminate objects which, if they were fully determined, would be concrete things. Thus quali­ ties such as colour, pleasantness, extension, shape, which Mr. Johnson calls 'determinables', are indeterminate characteristics. The relation of exceeding is an indeter­ minate relation. The objective ' X is hurt', usually called a propositional function, and treated as a purely symbolic convenience, is an indeterminate objective. 1 The object 'man' is an indeterminate object belonging to the same category as do concrete things. If the phrase were not strange and apparently contradictory, we might call it an indeterminate concretum. Finally the object 'something', or the variable X, is the most indeterminate of all objects, since even its category is indefinite. We see therefore that there is a dimension of determinacy in which an object of any type may vary; indeterminateness is not in any way peculiar to qualities and relations. We may note further that there are wholly determinate qualities and relations as well as wholly determinate con­ crete things. If we take a certain perfectly definite shade of blue, it is clear that we cannot 'complete', or 'fill in', such an object in any way. Its nature is an utterly simple, qualitative moment, which, however much it may be buffeted about the universe, and whatever circumstances I Professor Mally calls such indeterminate objectives Bestimmungen (deter­ minations) . Thus 'being hurt' or 'being greater than 1 0' are Bestimmungen. See Studien zur Theorie der Moglichkeit und Atmlichkeit, Report of Akademie der Wissenschaften, Vienna, 1 922, p. 8 .

1 66

T HE T H EORY O F I N C O M PLETE O B JE C T S

it may enter, is unaffected and unalterable. No one who knows such an object at all can know it more deeply or better. If this quality be bound up with other qualities, such an addition is essentially an external connexion, a putting together of entities whose natures are not com­ pleted by such a process. Thus the union of the shade S with the determinate shape R does not substitute for an indefinite shade or shape a shade or shape that is more definite. We must conclude therefore that 'universals', in the sense of indeterminate objects, are quite different from 'universals' in the sense of qualities or relations. This difference was perfectly clear to Aristotle when he dis­ tinguished between those things which can be predicated of a substance but which are never in a substance (i.e. genera and species), and those things which could be in a substance but could never be predicated of it (i.e. qualities such as whiteness). VI

Meinong considers whether incomplete objects may be said to have existence or subsistence. This cannot be the case, because everything that has being is fully determined in every possible way. It might be natural to suppose that they are therefore non-existent or non-subsistent. This view Meinong also regards as a mistake, unless non­ existence or non-subsistence be interpreted in a wider sense than usual. Thus the non-existence of 'man' in general is quite different from the non-existence of a character in a novel, assuming that the novelist had an infinitely powerful imagination, and had described this character in every possible respect. The non-existence of the fictitious character is a definite exclusion from existence of a certain object, whereas the non-existence of 'man' simply means that it is not in any way determined whether 'man' exists or not. And, presumably, we can only add a

T HE T HEORY O F I N C O M PLETE O B JE C T S

167

genuine being to an object after its nature has been fully determined. The lack of being of a partially determined object is due to the fact that there is simply no sense in asking whether a thing has being, when it is not fully determined. Being can only supervene on a fully deter­ mined so-being; in the less advanced stages of determina­ tion it has no application at all. To quote Meinong's own words: Here we see, in addition to the indeterminateness of so-being, with which we are already familiar, an indeterminateness in respect of being. Incompletely determined obj ects, or, more precisely, obj ects incompletely determined in so-being, are, we may say, indeterminate in respect of being, except in so far as their particular nature excludes being. The disj unction demanded by the law of excluded middle 'Either .A is or .A is not', like the disj unction which concerns so-being, presupposes that .A is determinate in respect of being. 1

Meinong also considers the objection that incomplete objects must exist in some cases, since they are parts of complete objects which exist. The balls on my billiard table are spheres and exist; are we not forced to say that 'the sphere as such' exists, since it is a part of the being of each of these concrete billiard balls ? In a similar way, did 'the dirigible airship' not become an existent when Zep­ pelins were first constructed ? 2 Meinong's answer to this difficulty is to affirm the principle on which his doctrine of content and object is based, the principle, namely, that an object A may involve or refer to an object B, without its being in any way true that B is a p art of A. 'The sphere as such', and the billiard ball on my table, are intimately related; some properties are common to both of them, and some are possessed by the billiard ball alone. But we can­ not conceive of the billiard ball as literally made up of 'the sphere as such' and a number of properties, in the same 1

Mag., pp. 1 79, r 80.

2

Ibid., p.

2 IO.

168

T H E T H E O RY O F I NCO M PL E T E O B J ECT S

way that a knife is made up of a handle and a blade. All the parts of a concrete object are quite as concrete as it is itself; we cannot conjure it out of a complexion of abstracta. It is further clear that 'the sphere as such' could only become a part of my billiard ball by surrendering its original indeterminacy, that is, by ceasing to be the object that it is. Just as an empty schema cannot really be a part of a schema that is filled in, because this would involve a loss of its essential emptiness, so an indeterminate object cannot, strictly speaking, be present in a determinate object with­ out losing its indeterminacy. If an object is indeterminate, it is contrary to its nature to be bound up with external determinations; we may replace it by a more determinate object, but we cannot really determine it further. Hence all the talk about 'filling in' involves a rather dangerous metaphor. 1 We may remark in passing that Meinong's argument refutes all those theories which hold that in some way incomplete things will be 'gathered up' into something more complete. If it is true that we have broken arcs on earth and a perfect round in heaven, then it cannot also be true that the things in heaven are really the same as the things on earth, for in losing their brokenness the arcs would sacrifice their identity. If it is wrong to regard any incomplete object as a part of a complete object, it is nevertheless evident that there is an intimate relation between an incomplete object and all the complete objects which are determined where it is undetermined. Meinong says that 'the sphere' is im­ plektiert (involved) in the billiard ball of my friend, and in all other spherical objects; we may translate this by saying that an incomplete object is embedded in certain complete objects. It is always possible for us, by ignoring a deter­ mination of a complete object, to replace it by one of the incomplete objects which are embedded in it. And the 'embedding' is such an intimate relation that we constantly 1

See Mag., p.

2 24.

T HE T H E OR Y O F I N CO M P LETE O B JECT S

1 69

use the incomplete object and the complete object as sub­ stitutes for one another. It might then seem to be reasonable to attribute some sort of being to an incomplete object which was embedded in an existent complete object; there seems to be a differ­ ence between 'the sphere' and 'the two-sided plane figure', in that the former is embedded in some existents, the latter in none. Vve saw that it is impossible to say that any in­ complete object 'exists' or 'has being' ; nevertheless, as Aristotle says, the word 'being' has many senses, and we may say of something that has not got being in the primary sense, that it is in so far as it is somehow involved in that which primarily is. Meinong says: We have said that the incomplete obj ect does not really exist or subsist i n the obj ect in which it is embedded ; nevertheless the existence or subsistence of the latter affects to some extent, if not the so-be i ng, at any rate the being of the obj ect embedded, so that we can conceive the notion of implexive being, or, more precisely, of implexive existence or subsistence. 1

An incomplete object like 'the sphere', which is embedded in existents or subsistents, is somehow part and parcel of the actual world, it is realized, exemplified, manifested, whereas an incomplete object like 'the plane figure with two sides' lacks even this parasitic sort of being. Whether Meinong is right in regarding what he calls implexive being as a genuine variety of being, fit to be set beside existence and subsistence, is not easy to say. But it is a valuable service to have pointed out that the sense in which we say that 'the dog' exists, or that 'the philosopher's stone' exists, is quite different from the sense in which we say that this dog or that lump of gold exists, and that the first of these senses is derivative and definable in terms of the second. Meinong adds to the conception of implexive being the corresponding conception of implexive so-being. Just as I

Ibid., PP·

2 1 1, 2 1 2.

1 70

T HE T HEORY O F I N CO M P LETE O B JECT S

we say that an incomplete object has implexive being because it is embedded in objects that have being, so we may say that an incomplete object has a given property in its implexive so-being, because it is embedded in existent or subsistent objects that have that property. The 'triangle as such' is not isosceles; but some of the triangles in which it is embedded have this property, and so it may be said to have it derivatively or implexively. We point to this derivative possession when we say that it is possible for 'the triangle' to be isosceles. If we take such a property as having the sum of its angles equal to two right angles, this will be found in all actual triangles, and will therefore enter necessarily into the implexive so-being of the triangle. How Meinong makes use of these distinctions to throw light on the theory of probability will be discussed in a later chapter. VII

We have examined incomplete objects and discussed their apparent immunity from the law of excluded middle, as well as the way in which they participate in the building up of the actual world. But their principal practical impor­ tance is epistemological: it is through their means that we apprehend all complete objects. Meinong says: 'Com­ plete objects are then only accessible to us via incomplete objects, which, as we saw, manifest themselves as a sort of aid towards the apprehension of complete objects.' 1 Two things seem perfectly clear; that, when an ordinary man, who is not a student of Gegenstandstheorie, says 'I met somebody in the street', or 'I see something that is brown', he does not mean that he has either seen or encountered an incomplete object. He is talking about complete objects, which exist if anything does. At the same time it is hard to believe that complete objects are really given to his thought, since he has the vaguest conception of the nature 1

Mog., p.

223 .

T H E T H E O RY O F I N CO M P L E T E O B J ECT S

171

of the things he is referring to. And if the man is think­ ing neither of a complete nor an incomplete object, then what sort of object is before his mind ? Something must be given, and this something, whatever it is, must be investi­ gated by the theory of objects. Meinong approaches this difficult question along a psychological route. According to him, all conscious reference to objects involves either a judgement or an assumption. A mere Vorstellung is a passive experience, in which the reference to an object is only potential, not actual. 1 Now we saw that a judgement or assumption can only apprehend an object as the material of some objec­ tive; 2 it is only by apprehending some circumstance about an object that the object can be referred to (gemeint). For our purposes the most important way of referring to an object is what Meinong calls the reference by way of so-being (Soseinsmeinen) ; we think of an object by judging or assuming that there is something that has a certain property.3 This mode of reference may be contrasted with another mode which Meinong calls the reference by way of being (Seinsmeinen) ; in this mode we think of an object A by making the simple judgement or assumption that A exists. 4 It is clear that such a judgement or assumption, unless it is helped out by judgements or assumptions of so-being, will not reveal to the mind the nature of A. The object will be given as a vague 'something or other', but we shall not be able to make out what it is. Meinong considers that certain rudimentary forms of perception are cases of the pure reference by way of being. He says: The inner aspect of such (perceptual) experiences will not permit us to doubt that we have in them no combination of a reference by way of being with a reference by way of so-being, but a relatively original form of reference through being, a form unaffected by I 3

0. A., P· 2 2 5 . Ibid ., p. 2 6 9 .

2 4

Ibid ., p. 47. Ibid., p. 2 3 9.

172

T HE T HEORY O F I N CO M PLETE O B JECT S

abstraction . . . . Concreta are therefore the proper field for the pure reference by way of being. 1

A man can perceive a brown door or a red carpet with­ out perceiving that it is a brown door or a red carpet ; 'things' are before us in perception without our needing to describe them to ourselves. This is hard to understand if we consider that the essence of an object is constituted by its so-being and not by its being, but Meinong holds that the nature of an object is given implicitly or potentially, and will immediately reveal itself as soon as our perception becomes more analytic. 2 We shall then pass from a rudi­ mentary perception of things, to the perception of things endowed with certain properties and distinguished as doors and carpets, or as brown and red.3 In such developed per­ ception the reference by way of so-being is fused with the reference by way of being ; we are no longer confronted with a mere 'this', but with a 'this' which is 'so-and-so'. Meinong attempts to show that the reference by way of so-being is a much more powerful instrument than the reference by means of being. The latter type of reference is confined to two limited fields: the 'things' which are before us in perception, and certain highly abstract and fragmentary entities. It will present us either with the unanalysed 'thises' of the baby's world, or with such float­ ing properties as blueness. The reference by way of so­ being can, on the other hand, present to our thought any type of entity; every object can be given to us indirectly as the possessor of certain properties. Meinong holds, as we saw, that the presentation of any object presupposes a peculiar modification of our inner experience which he calls a content. Thus, if I am to see the brown colour of a door, I must live through an experi­ ence of a peculiar quality, an experience which is not itself brown, but which is such that, if I were to judge any object to be present, this object would be presented as brown and 1

0. A., p. 2 70.

2

Ibid . , p. 2 7 5 .

3

Mag., p. r 93 .

T HE T HEORY O F I N CO M PLETE O B JECT S

1 73

nothing else. If now a judgement of being operates on this brown-content-(we use this name since language has no proper names for contents)-it will throw before my mind an appropriate object. As all judgements of so-being are either absent or implicit, I shall be merely confronted with brownness as a peculiar objective something ; its own nature, or its relation to anything it qualifies, will be quite obscure. 1 If now a judgement of so-being supervenes, it will use the brown-content to present brownness, not as an independent entity, but as an element in the nature of a many-sided object. I shall no longer be aware of a mere brownness, but of something that is brown. This second type of reference is much more important than the first because we can pass by its means beyond what is immediately given to entities which can only be given in part. Meinong contrasts the direct acquaintance we have with brownness, when we use the first type of reference, with the indirect way in which the brown thing is given to us. Brownness is really present to us, whereas the brown thing has only a sort of quasi-presence ; 2 it is only given to us in so far as it is brown, while the rest of its nature is quite obscure. Now complete objects can only have this quasi-pre­ sence, since their natures are infinitely complex and can only be grasped in part. No doubt there is a sense in which concrete objects are given to us in perception as mere 'somethings', but such presentation is utterly vague. All clear apprehension of objects occurs via their so-being. Meinong says: . . . this quasi-presence of an obj ect referred to by means of its 1 Meinong holds that brownness is an incomplete object, but that it differs from other incomplete objects because 'it is not per accidens undetermined, but necessarily undetermined in that it cannot admit such determination' (Mag., p. 1 8 5). It seems to me that there is no good reason to regard brownness, if by this we mean a perfectly specifi c shade of brown, as an incomplete object . It may be incomplete in the sense that it must be conjoined with other properties in order to exist, but this is not the inner incompleteness of 'the square' or 'something 2 Mag., p. 1 8 8. heavy'.

1 74

T H E T H EO R Y O F I N C O M P LETE O B J E C T S

so-being allows us to understand why this type of reference is not confined to incomplete obj ects like the reference by means of being. In the latter type of reference content and obj ect are, as it were, so close to each other, that a complete content would certainly be necessary for the apprehension of a complete obj ect, or rather a complex of contents, each of which was adequate to one determina­ tion of the obj ect ; this reference is excluded by the infinite number of the members of such a complex. As opposed to this, in the case of a reference by means of so-being, it suffices if the content only 'points' to the obj ect to be apprehended. This happens in the case of every obj ect that has a certain determination or complex of determinations, which the content underlying the reference is able to present (i.e. present in the full sense, in a simple reference by way of being). What other properties the obj ect thus determined may possess is quite irrelevant : but, if it is in fact a complete obj ect, then the reference is to a complete obj ect. 1

To sum up, we may refer to an object by assuming that X exists or by assuming that something has certain pro­ perties. In the former case X must be present in the full sense, and must, accordingly, if it is to be clearly pre­ sented, be some very elementary sensuous fragment ; in the latter case only the properties of the object referred to need be fully present; hence the object may be of any degree of complexity, and may even reach the infinite complexity of a complete object. Not all objects can be literally present to us, but we can 'point' to objects that are not present and so give them a quasi-presence. 2 VI II

We now return to the question which prompted this digression. Suppose I make the judgement or assumption that there is something that is blue, I can set before my mind in this way a certain object. If I am investigating Mag., p.

1 8 8. In this account I have somewhat simplified Meinong's doctrine, and have passed over certain inconsistencies and difficulties. The full treatment of the problem of reference is postponed to Chapter VIII. 1

2

T HE T HEORY O F I N C O M P LETE O B J EC T S

1 75

the theory of objects this may be an incomplete object, a blue thing whose nature is wholly indeterminate in every other respect. But, if I am thinking of actual objects in the real world, or even fictitious objects in a novel, it is quite clear that the object which acquires a certain quasi­ presence by means of the reference is thought of as com­ p lete. On what does this difference depend ? How are we enabled to make a reference to something of which only the most shadowy simulacrum can be clearly conceived ? Meinong puts this difficulty in the following form: Both complete and incomplete obj ects are accessible to the reference by means of so-being; the question therefore arises, on what it really depends, whether this reference actually functions as an apprehension of complete or of incomplete obj ects. 1

And he solves the problem in the following way: My answer to this question is so fundamentally simple that it is hard to get rid of the fear that it is too simple. I f my reference by means of so-being is to apprehend a complete obj ect, nothing seems more natural than to include the moment of completeness explicitly in the reference. If I wish to refer to something wholly determined by the thought of 'a square brown thing', without necessarily knowing the determinations which distinguish it from something else, I need only think of 'a determinate square brown thing'. 2

We make a reference, which would naturally present an incomplete object, present a complete object, by stipulat­ ing explicitly that the reference is to something complete. The obvious objection to this solution is that the mere addition of another determination to an incomplete object is powerless to make it complete. A determinate square brown thing seems to be still an incomplete object, because we have left quite undetermined in what ways the square brown thing is determinate. It may be heavy or light, large or small; beyond the fact that it is square, brown and determinate, nothing is decided. Meinong's answer to 2

Ibid.

1 76

T H E T H EO R Y O F I N C O M P LETE O B J E C T S

this is to point out that 'determinateness' is not really a property of the same simple type as squareness, brown­ ness, or heaviness. The determinateness or indeterminate­ ness of an object is not a part of its nature in the same immediate way that other properties are. Meinong takes over from Ernst Mally the view that there are two fundamentally different types of properties of objects, those which are nuclear (konstitutorisch) and those which are extra-nuclear (auj]erkonstitutorisch). 1 The property of simplicity is an interesting example of an extra-nuclear property. There are some objects, e.g. a certain specific shade of red, which it would be usual to regard as simple. If, however, we treat this simplicity as part of the nature of the shade we are involved in a contra­ diction; the nature of the shade involves two 'moments', being-red and being-simple, and is therefore complex. Again, if simplicity be an element in the shade of red, all objects that are characterized by the shade will be also characterized by simplicity, which is absurd. Meinong disposes of these Megarian subtleties by holding that the simplicity of the shade of red cannot be treated as a con­ stitutive part of its nature, or even as something consecu­ tive upon this constitutive part; it is a property of higher order which is 'founded on' the nature of the object. The determinateness or indeterminateness of an object is similarly an extra-nuclear property; accordingly, when an indeterminate object is embedded in a determinate one, the latter does not participate in its indeterminateness. We see, therefore, that the passage from 'square thing' to 'determinate square thing', is different in kind from the passage from 'square thing' to 'square heavy thing'. In our ordinary thinking we are so exclusively absorbed in the world of existence that we do not trouble to em1 See Mag., p. I 76. I have translated konstitutorisch as 'nuclear' because Meinong distinguishes it from konstituti-v, which it is best to translate as 'constitutive'.

T H E T H E O R Y OF I N C O M PL E T E O BJ E C T S

1 77

phasize the complete determination which, we presume, belongs to the things we are referring to. 1 And it would be a fiction to imagine that we were.first presented with an incomplete object, e.g. 'a square thing', and that we then proceeded to make it do duty for a complete object by adding to it the property of 'completeness'. No such process is introspectively revealed; the man who thinks of a square thing, unless he is philosophizing, normally thinks of it as complete. There is a content in his experi­ ence which is used to present a complete object through a certain segment of that object's so-being, but this con­ tent can also be used to present the corresponding in­ complete object. The same experiential material, used in one way, sets before us the determinate square thing; used in another way, it sets before us the indeterminate square thing. As contents are in general obscure and inaccessible, it is convenient to substitute for them the objects which they have the potentiality of presenting. Hence we may call the indeterminate square thing which the content could present an auxiliary object (Hilfsgegenstand), and the deter­ minate square thing which the content does present the ultimate object ( Zielgegenstand). .Adopting this frankly fictitious mode of speaking, we may say that, where the ultimate object of our reference is a concrete existent or one of its aspects, we refer to it by means of an indeter­ minate auxiliary object. 2 It is through one or other of the incomplete objects embedded in it that a complete object is given to our thought. It is by means of the incomplete object 'man' that the son of Callias, who is much more than merely a man, is given to us. We think of a certain con­ crete object as 'a table', or 'a piece of furniture', or 'a brown thing', or 'a hard thing on which we have bruised ourselves in the dark', and so on. The auxiliary object is able to do duty for the ultimate 1

824187

Mog., p. 1 9 1 .

2

N

Ibid., p . I 9 6.

1 78

T HE T HEO R Y O F I NCO M P LETE O B JECT S

object because the extra-nuclear property of determinate­ ness has been added to it. As so modified the auxiliary object becomes a peculiar type of object, intermediate between the complete and the incomplete object ; to this Meinong gives the name of the completed object ( vervoll­ stiindigte Gegenstand). He writes as follows: Through such surrogative or fictitious presentation, whose nature we have j ust described in a manner which is, we hope, not too cursory, we arrive at obj ects which occupy a peculiar inter­ mediate position between the incomplete obj ects, &c., which are fully present to us, and complete obj ects which are for the most part not fully present. These (intermediate) obj ects are, of course, in­ complete obj ects, but, in comparison with the present and quasi­ present obj ects correlative to a given content, which are merely incomplete, 1 they may be called completed incomplete obj ects. I n so fa r a s they function a s auxiliary obj ects, they may also b e called completed auxiliary obj ects. 2

We have therefore three objects: the complete object which is the ultimate object of our reference, the merely incomplete object by which we make our reference, and which is therefore called an auxiliary object, and the modi­ fied form of the latter, in which determinateness is explicitly added to it, and in which it 'does duty' for the complete object. This last variety of object is the completed object. Since a completed object is merely a substitute, and not an object in its own right, it manifests an elasticity not to be found in the merely incomplete object. Thus 'plane figure with three sides' and 'plane figure with three angles' are, as merely incomplete objects, rigidly distinct ; the corresponding completed objects are not, however, dif­ ferent, any more than the corresponding complete objects are. The incomplete objects 'something that is A', 'some­ thing that is AB', 'something that is ABC', &c., are a series of objects, each of which may replace another, but 1

2

There are, of course, quasi-present objects which are complete. Mag., p. 203 .

T HE T HEORY O F I NCO M PLETE O B JECT S

1 79

none of which can in any sense be identical; on the other hand, the completed objects 'something definite that is A', 'something definite that is AB', 'something definite that is ABC', &c., may be treated as identical, as they all do duty for the same complete object. 1 We have therefore discovered what is before a man's mind when he sees 'some one' in the street, or knocks into 'a hard thing'. It is not an incomplete object that he has seen or knocked into, nor, on the other hand, are complete objects fully present to his thought. But what are present are incomplete objects modified by the deliberate addition to them of the property 'complete determination'; thus modified they can act as surrogates or deputies for the corresponding complete objects, and are known as com­ pleted objects. This account raises formidable difficulties. In the first place it rather looks as if the completed incomplete object is quite as impossible as the round square. It does not matter if the incompleteness of 'square thing' is extra­ nuclear; if we substitute for this object 'determinate square thing', we are thinking of an object as both deter­ minate and indeterminate. It seems as absurd and useless to think of something in this way as to think of a simple object as complex. Surely the extra-nuclear property of determinateness does either in fact belong or not belong to an object; it is not in our power, without error, to add it where it is absent or conjure it away where it is present. Only where the nucleus is in fact completed, can the extra-nuclear property of completeness supervene; to at­ tach it to an incomplete object is to attempt to transform that object into something utterly different. There seems to be no way out of this impasse except to admit frankly that such a procedure is contradictory. To treat the incomplete object 'square' as a complete square is not in principle different from assuming that a square 1

See Mog., pp.

20 1 , 202.

1 80

T H E T H E O RY O F I N C O M PL E T E O B J E C T S

is round. This does not, however, mean that such self­ contradictory and impossible objects may not fulfil a useful function in knowledge. We are forced to employ incomplete objects as representatives of complete objects, because the latter would otherwise, owing to the infinite complexity of their natures, be quite inaccessible to us. Such use of incomplete objects involves a fiction, and a fiction which is, in the last analysis, a contradiction. But, as our knowledge deepens, and we fill up the completed object with more properties, the discrepancy between the indeterminateness of its nuclear properties and its extra­ nuclear determinacy will become less and less, until, when the object is completely known, the completed object will be fused in the complete object, which at first it only represents. Completed objects are like sketches which do duty for complete objects; such 'doing duty' is, strictly speaking, an absurdity, as no entity can play the part of another. But as the completed object is 'filled in', the sketch becomes more and more like its original, till at last it loses itself in the original. In some such way we shall have to get over the difficulties involved in the mental reference to the concrete, difficulties which, of course, are not peculiar to the doctrine of Meinong. IX

The theory of incomplete objects throws a flood of light on the distinction between analytic and synthetic judge­ ments. Meinong refuses to believe that this distinction is merely a subjective one; there is some objective differ­ ence between the judgements 'Triangles are figures' and 'Metals are conductors'. 1 But the difference is not a purely objective one since it cannot be right to predicate some­ thing of an object which does not really pertain to it. If we consider merely the objective material of our j udgements, 1

Mog., pp. 205,

206.

T H E T H E O RY O F I NCO M P L E T E O B J ECT S

181

all judgements are analytic, for all assert of objects what is comprised in their nature. A via media between pure subjectivism and pure objectivism is found, when we remember that one and the same complete object may be approached through various auxiliary objects. If I refer to the complete object X by means of the auxiliary object then the judgement that X is / will be analytic if / belongs to the nucleus of synthetic if it does not. The j udgement that triangles are figures is analytic, because the auxiliary object 'triangle' has 'being a figure' as a constitutive property, whereas the judgement 'Metals are conductors' is synthetic, because conductivity is not in­ cluded among the constitutive properties of the incomplete object 'metal'. Whether a judgement is analytic or synthetic will there­ fore depend on the auxiliary object which we have selected to do duty for a complete object ; it is a subjective distinc­ tion in so far as our choice of an auxiliary object is arbitrary, objective in so far as such an object is definitely an object. It seems clear that Meinong could quite consistently have held that there are analytic and synthetic objectives as well as judgements, though synthetic objectives could only be about incomplete objects. If a property / is among the constitutive properties of an incomplete object X, then the objective ' X is /' will be analytic. If, however, f only belongs to X implexively, i.e. as a property of the com­ plete objects in which X is embedded, then ' X is / implexively' will be a synthetic objective. An analytic objective is one of so-being in the proper sense, a synthetic objective one of implexive so-being. Auxiliary objects can function in various important ways. We can use the incomplete object 'man' to think of ' ' ' ' I • ' ' ' ' a certam man , or any man , or a }} men , or some men . Meinong gives no detailed discussion of what these verbal differences express or mean. He thinks, however, that

r,

r,

1

Mog., p.

207.

1 82

T HE T HEORY O F I NCO M PLETE O B JECT S

the notion of a class (Kollektiv) is involved in the meaning of 'any man', 'all men', or 'some men'. There is a class of complete objects we are approaching through an in­ complete object which is embedded in all of its members. This view would explain why the judgements 'All men are mortal' and 'All rational animals are mortal' are not identi­ cal judgements, though they determine the same set of objects in the same manner. The reason is that this set of objects is apprehended through a different auxiliary object in each case. It seems clear that it would be possible to make a similar distinction between the two general facts 'All men are mortal', and 'All rational animals are mortal', quite regardless of our mode of apprehension of these facts. For the fact that all men are mortal is not identical with the fact that mortality pertains to A, B, C, D, E . . . (where A, B, C, D, E . . . are all the men that there are). If this were so, it would certainly be identical with the fact that all rational animals are mortal, since this concerns the same set of objects. But the statement 'All men are mortal' means the fact that the incomplete object 'man' does not have implexive being without also having 'mortality' in its implexive so-being, and this fact is dis­ tinct from the equivalent fact about the incomplete object 'rational animal'. In this way the distinction between two such general facts could be shown to be a purely objective one, which does not depend on our judgements or on our subjective approach. X

We may conclude this chapter by suggesting a few fundamental modifications of Meinong's theory which are involved in the doctrine of 'determinates, held by Profes­ sor Mally, and discussed in Chapter IV. On Meinong's theory, if I think of something blue, there is an object of

T HE T HEORY O F I N CO M PLETE O B JECT S

183

my thought, which is really blue, but which is undeter­ mined in every other respect. This doctrine is difficult in two ways. It is in apparent conflict with the law of excluded middle, and we can only get rid of this conflict by various subtle and unplausible distinctions. Then, in the theory of apprehension, it gives rise to serious diffi­ culties, because it is, strictly speaking, absurd to treat an incomplete object as if it were complete. On the theory of Mally, the object 'something that is blue' is merely the determinate of the determination 'being blue'; it does not satisfy this determination. The only objects which satisfy the determination of being blue are concrete blue exis­ tents. 'Something that is blue' is not really blue; the only property it really possesses is that of being determined by the determination 'being blue'. This property or deter­ mination, which is one of higher order, it satisfies. It will be clear that 'something that is blue' cannot be ranged alongside of blue things, as a number of the same species; tradition has recognized this by speaking of it as a 'concept'. On the theory we are discussing, 'some­ thing that is blue' is not in any way an incomplete object in the sense that its nature is partly undetermined; it is a perfectly definite determinate of certain determinations, and has a being of its own. It only looks indeterminate if we confuse it with the object which satisfies its deter­ minations. In apprehending concrete existents we do so by means of the determinates of certain determinations. We grasp through the determinate at the object which satisfies a set of determinations. The determinate is not referred to specifically, it is apprehended in being grasped through (wir erfassen es durchgreifend), while the object which has the determinations is actually referred to, and is appre­ hended in 'being hit' (wir erfassen ihn trejfend). If I think of 'something that is blue' I 'hit' all the blue objects that there are, but I do so by means of a determinate which

1 84

T H E T H E O R Y O F I NCO M PL E T E O B J ECT S

subsequent reflection will present explicitly. All reference to objects is like a shot in the dark; the determinate gives us our direction, but the cognitive situation, considered in itself, need not tell us whether we have hit anything at all. We may discover that there was nothing to hit in that direction, in which case only the determinate remains in our hands. It seems to me that a theory like the one we have sketched solves most of the difficulties of the doctrine of Meinong, without ignoring the points of which he was so acutely conscious. But it might have its own difficul­ ties, which we cannot go into here. 1 1 See Gegenstandstheoretische Grundlagen der Logik und Logistik, by E. Mally, p . 76.

NOTE. It is perhaps worth observing that Meinong's distinction between the auxiliary and ultimate obj ect does much the same work as Frege's distinction between Sinn (Sense) and Bedeutung (Reference) .

VII TH E M ODAL PROPE RTI ES OF OBJE C TI V ES I

B

the modal properties of objectives Meinong understands such characteristics as necessity, pos­ sibility, factuality, truth, probability, which tradi­ tional logic regards as properties of judgements. That they are not properties of the experience we call judgement is clear; if I judge that something must be the case, or is the case, or might be the case, or is likely to be the case, it is not my experience of judging that is a fact, or a necessity, or a possibility, or a likelihood; my judge­ ment simply exists as a mental happening, and no such descriptions are applicable to it. It is the object of my judgement to which such properties attach, and this is, as we have seen, always an objective. That modal properties attach to objectives, and that judgements have nothing to do with them, rests, according to Meinong, on direct empirical evidence. 1 We see them in the objectives, not in our experiences. He criticizes 2 Marty's psychologistic definition of being or existence as 'the property of being capable of a valid acknowledgement' (was mit Recht anerkannt werden kann); as little as the nature of colour consists in the physical and psycho­ physical events which condition its manifestation, so little does the nature of being lie in being acknowledged, or being capable of being acknowledged. An examination of factuality, possibility, and necessity reveals a similar independence of all psychological factors. This does not, Y

1

0. A., P· 84.

2

Ibid., p . 62.

1 86

T HE M O DAL PR O PE R T IES O F O B JECT I VES

however, mean that it may not often be useful to approach such ultimate things from the psychological side; it may enable us to describe entities and qualities which would otherwise have to be simply apprehended, and which might seem obscure because so little could be said about them. II

Factuality occupies the central place in Meinong's list of modal properties, ousting the property generally called truth. Truth is not, properly speaking, a separate modal property of an objective; nor has it the fundamental im­ portance which philosophers have generally attributed to it. 1 As we have seen, an objective is true if it is factual, and if it functions as the pseudo-existent object of some apprehending experience; it is only an important property because an objective which is a fact can be given to us without its factuality being given, i.e. we often light on facts without knowing that they are facts. There is nothing in an assumption, or in a judgement which has no self­ evidence, to show that the circumstance it presents is really a fact. Consequently the factuality of an objective appears to lie beyond and outside it, and this gives rise to the rather misleading notion of two things, a pseudo­ existent objective and a fact to which it corresponds, that is, to the correspondence theory of truth. In so far as it is legitimate to distinguish between an objective as apprehended, and an objective as an entity with its own independent nature and being, to that extent this corre­ spondence account is valid. But the property which thinkers have been seeking to express when, like Aristotle, they spoke of being as truth, or of the timelessness of truth, is clearly not truth but factuality. 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 some one. I

0. A., p. 95·

T HE M O DAL P R O PER T IES O F O B JECT I VES

1 87

Factuality is to objectives what subsistence and exis­ tence are to objecta; the being peculiar to objectives is being-the-case. Unfactuality is simply the absence of factuality; Meinong refuses to admit that it is a positive quality like factuality. Since, strictly speaking, there are no negative characteristics/ an objective is unfactual when it is the material of a negative fact of the form ' O is not factual'. Meinong repudiates the famous comparison of Mr. Russell, in which truth and falsehood are said to differ as a white rose differs from a red rose. 2 In Mr. Russell's account truth and falsehood are practically equivalent to factuality and unfactuality, as neither has anything to do with our apprehension. Meinong refuses to admit that falsehood or unfactuality can be regarded as something positive on an equal footing with truth and factuality. 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. The factuality of an objective is given to us by a peculiar quality of our experiences which we know as evidence (Evidenz); 3 the judgements of inner perception are experiences which have this quality. III

We now turn to Meinong's treatment of necessity. That this is a property of objectives and not of objecta is fairly obvious. We cannot call a house, a feeling, or a colour necessary; 4 it can only be necessary that such things should exist, or have certain properties. The absence of necessity from the realm of objecta explains why some philosophers have doubted whether the word 'necessity' has any objective reference at all. Like being or negativity, I 2

3

See above, Chapter III, viii. 'Meinong's Theory', Mind, 1904, p. 52 3 ; 0. A., P· 95· Mog., p. 148.

4

Ibid., p. 233.

188

T H E M O DAL PRO PERT IES O F O B JECT I VES

necessity is something that eludes presentation by a mere Vorstellung. According to Meinong the necessity of an objective is given to us when our judgement of it is not only accom­ panied with evidence, but also with understanding (Ver­ standnis). 1 By 'understanding' Meinong does not mean the understanding of words, but a peculiar experience in which a fact no longer confronts us as something which merely is so, but as something whose being so is natural and 'intelligible'. We can understand that red is different from green, that the sum of the first n odd numbers is a perfect square, or that every objective implies itself. All our a priori formal knowledge is accompanied by under­ standing. On the other hand, it may be perfectly evident or very nearly evident to us that the sun is shining, that we are making an assumption, that we had breakfast yesterday, that gold is fusible, but we do not understand these facts. 2 Even if we could explain them by showing that they were Laws of Nature, or consequences of Laws of Nature, such explanation would only be a sort of quasi-understanding. There are then two classes of objectives, those that can be understood and those that cannot be understood. Meinong refuses to take the view that the difference between these two classes is merely a difference in our mode of knowing them; it is not merely a fact that the former set of objectives are formally certified, and the latter empirically certified. The fact that they have to be certified in different ways, that in the first case evidence is combined with understanding, whereas in the second case it occurs alone, proves that there must be some ulti­ mate, non-psychological difference in the objectives con­ cerned. Meinong says: Little as we are able to state in what understanding, so familiar to us as an experience, really consists, one point is quite clear, that 1 2

Mog., p. 2 3 5 ; 0. A., p. 9 r . Mag., pp. 1 4 2 , 2 3 5 .

T H E M O DAL PRO PERT I E S O F O B J ECT I V E S

1 89

our understanding must find its foothold in that which we under­ stand, which means that, in all cases of understanding, the nature ( Beschaffenheit ) of the obj ective which is to be understood plays an essential part. 1

It follows that the experience of understanding, interest­ ing as its analysis is, belongs to psychology and not to the theory of objects. The latter is only concerned with those properties in objectives which make some of them acces­ sible to understanding, whereas others are left inaccessible. The understanding-experience plays the part of a content by means of which an objective characteristic, necessity, is given to us. Here, as in other cases, the direct approach to the object is easier and more natural, whereas the ap­ proach by way of the presenting experience is definitely a circuit. Meinong says: 'An objective is necessary in so far as it is evident a priori, but it is this evidence which reveals necessity to us, not any consideration of evidence (as an experience).'2 We do not need to practise intro­ spection, and be sure that a certain judgement is accom­ panied by understanding, in order to see the necessity of the objective which is judged. Meinong also points out that an analysis of understand­ ing is much more difficult than an apprehension of neces­ sity. It is comparatively easy to see and understand that the sum I + ½ + ¼ + ¼ + . . . tends towards 2 , but the nature of such seeing is elusive and obscure. We might say perhaps that a man who understands this objective feels a great difficulty in believing or even assuming any­ thing else; if this is a correct account of understanding, then our reliance on mathematical necessities is much greater than our reliance on the experience of understand­ ing. For we can see that the fact in question must be so, whereas we are far from sure that our powers of belief or assumption may not increase in the future, or that other I 2

Ibid., pp. 1 4 1 ' 1 42 . 0 . A . , p . 92 .

1 90

T H E MODAL P R O P E RT IE S O F O B J E CTIVES

people may not have totally different powers. 1 We have therefore to consider necessity on its own merits as a property of objectives, leaving aside altogether the way in which it comes to be given to us. There are two ways in which we may be said to under­ stand an objective, and, corresponding to these two sorts of understanding, there are two sorts of necessity. There is a prevalent conception which identifies all understand­ ing with knowledge which is based on some ground ; 2 this goes back to Aristotle's view of scientific knowledge as the knowledge of causes, which is not content to know that something is so, but must also discover why it is so. Meinong points out that not all rational evidence or understanding is mediate, as this demand for 'grounds' supposes ; there is also an immediate rational evidence. That diversity should hold between red and green is not only evident but also intelligible ; yet understanding of this fact does not depend on a knowledge of other facts which serve as grounds for such a diversity. 3 Meinong also points out, as Aristotle does, that to require knowledge of a ground for the understanding of every fact, would lead to an infinite regress, and, since we cannot understand a fact on this hypothesis without understanding all its grounds, we should never have perfect understanding of any fact. From such a conclusion Meinong and Aristotle save themselves by an appeal to an immediate under­ standing. Meinong gives a long discussion of what we may call relative understanding in the sixth chapter of Ober Annah­ men. He says : We can say that a person draws a conclusion, if he arrives at some new conviction by adducing one that he already possesses, i .e. by looking to this conviction (im Hinhlick auf diese) . The experience denoted by such a word seems to me to be sufficiently characteristic to require no further description. 4 1 Mag., p. 2 34. 2 Ibid., p. 2 3 5. 4 0. A . , p . I 7 6 . 3 Ibid., p . 2 3 6 .

T H E M O D A L P R O P E RT I E S O F O B J E CT IV E S

191

This Hinblick or looking-to some other cognitive experience, is to be found in all mediate understanding. An important point lies in the fact that the experience which is 'looked­ to' need not be at all evident, but may be a mere assump­ tion. The postulates from which certain theorems are deduced may be mere postulates, or, if they are believed, such belief may lack evidence or understanding. Mediate understanding is only perfect, however, if the ultimate judgements which it looks to are both evident and accom­ panied by understanding. There will then be two species of necessity correspond­ ing to the two varieties of understanding: sim pie necessity, such as the necessity that red should be diverse from green, and conditional or relative necessity, such as the necessity that a given stone should fall when it is released, if we take this to be necessary relative to the law that all unsupported bodies fall. 1 Meinong does not really require this doctrine of two types of necessity, a point which he admitted in the latest phases of his thought. The historical reason for the doctrine lies in the fact that, when Meinong wrote Ober Annahmen, he did not believe that what are called hypo­ thetical judgements are really single judgements at all; they are 'operations which include a judgement'. 2 We pass a judgement while we look-to some other judgement of assumption, and this is all that the verbal form ' If P, then Q ' expresses. This view made it difficult for him to hold that there are objectives or facts of the form ' If P, then Q ' . At a later stage of his thought Meinong was led to admit objectives of the form ' If P, then Q ', or ' Q on the condition P', as a peculiar and irreducible type of objec­ tive; they are called by him objectives of with-being (Mitsein), and with-being or being-on-a-condition takes its place beside being and so-being.3 The recognition of this type of objective led Meinong finally to abandon his 2 0. A., p. 20 6 . 1 Mog., p. 23 7 . J Mog., p. 1 55.

1 92

T H E M O DAL PR O PE R T I E S O F O B J ECT I V E S

view that there were relative necessities. Q is not neces­ sary relative to P, but the objective of with-being 'If P, then Q' or ' Q on the condition P' has absolute necessity. 1 He therefore substituted for the conditional necessity of a simple objective, the absolute necessity of a conditional objective. IV

Having characterized necessity by means of our sub­ jective approach to it, i.e. the experience of understanding, Meinong attempts to probe more deeply into its nature. He points to the interesting circumstance that when we say that something is necessary, we usually mean that it is a fact but also something more than a fact. 'If one follows accepted usage in talking of necessity, we are in every case (at least we think so) confronted by factuality, but this factuality has a determining addition' (Beisatz) . 2 This Beisatz can obviously not be an increase in factuality, for factuality does not admit of differences of degree; nothing can be more factual than fact. 3 Undoubtedly, it is this equality of all facts as regards their being facts which has led many philosophers to banish necessity from the universe. That the sum of the first n odd numbers is a perfect square is neither more nor less the case than that lions prefer a flesh diet. We have therefore to look elsewhere for the Beisatz which, when added to mere fact, will make of it necessity. Meinong finds on consideration that what is really wanted is not a Beisatz at all; the difference between necessity and simple factuality lies in the way in which factuality is in each case bound up with the factual objective. In every case of necessity the factuality of an objective inheres in it in an extraordinarily intimate way, whereas in cases of purely contingent 'brute' fact this intimate relation ap­ pears to be absent (though of course it may really be 1

Erw. Kaus., p. 8 0 .

z

Mag., p.

123.

3

Ibid., p.

I 22.

T H E M O D A L PRO PERT I E S O F O B J EC T I V ES

1 93

present unknown to us). 1 The relation of factuality to an objective in cases where necessity is absent, if there are such cases, may be described by saying that factuality merely adheres to the objective. 2 We therefore distinguish a necessary objective in that its factuality is inhesive (inhasiv) ; in cases of pure contingency factuality is only adhesive (adhasiv). 3 Objectives which are accessible to understanding have a material and a form which stand in an ' internal' relation to their factuality; they are such that they must be facts. If there are any purely contingent objectives, then their factuality, if they are facts, is alto­ gether external to their material and their form; there is nothing in their Beschaffenheit which decides whether they are the case or not. While all necessary objectives have inhesive factuality, Meinong holds that the sphere of inhesivity is much wider than that of necessity. Thus the factuality of Laws of Nature, which we establish by induction, inheres in those laws, in the same way that factuality inheres in mathemati­ cal theorems. This sort of inhesive factuality cannot be called necessary, because we cannot understand it a priori ; yet, since it is a genuine sort of inherence, we may say that such laws have empirical necessity or quasi-necessity. 4 Such quasi-necessity is of course quite distinct from the con­ ditional necessity of a certain fact relative to the law which governs it; it is not only particular cases of a law, but also the law itself which has quasi-necessity. This doctrine of Meinong's accords with common sense and the procedure of science, and is in conflict with nothing but the sceptical sensationalism of Hume. If we mean anything by referring to certain general facts as 'laws', we certainly seem to mean that their factuality sticks to them in some more intimate manner than the I

J 4

2 Ibid., p. 2 2 1 . Ibid., P · 1 4 2 . Meinong uses these strange words in preference to 'inh erent' and 'adherent'. Mag., p. 238 .

824187

0

1 94

T H E M O D A L PRO PERT I E S OF O B J ECT I V E S

factuality of any chance circumstance. The law that every case of